Texas Transportation Institute Oil and Gas Energy Developments and Changes in Crash Trends in Texas Final report PRC 15-35 TRANSPORTATION Policy Research CENTER Oil and Gas Energy Developments and Changes in Crash Trends in Texas Texas A&M Transportation Institute PRC 15-35 F October 2015 Authors Cesar Quiroga, Ph.D., P.E. Ioannis Tsapakis, Ph.D. 2 Table of Contents List of Figures ................................................................................................................................ 4  List of Tables ................................................................................................................................. 5  Executive Summary ...................................................................................................................... 7  Chapter 1. Crash Data Collection and Processing ................................................................... 15  Introduction ............................................................................................................................... 15  Crash Data ................................................................................................................................. 16  General Trends .......................................................................................................................... 35  Chapter 2. Analysis and Trends ................................................................................................ 41  Oil and Gas Well Developments ............................................................................................... 41  Changes in the Number of Crashes and Injuries ....................................................................... 41  Changes in Crash Rates ............................................................................................................. 48  Correlations ............................................................................................................................... 57  Economic Impact of Crashes..................................................................................................... 63  Chapter 3. Findings .................................................................................................................... 66  References .................................................................................................................................... 74  Appendix ...................................................................................................................................... 75  3 List of Figures Figure 1. Change in the Number of Rural Fatal, Incapacitating, and Non-Incapacitating CMV Crashes vs. Change in the Number of Horizontal Wells. ............................................ 11  Figure 2. Overall Trends in Well Development, Rural CMV Crashes, and Crash Injury Costs from 2006–2009 to 2010–2013. ............................................................................................ 14  Figure 3. Counties Analyzed in the Eagle Ford Shale, Permian Basin, and Barnett Shale Regions. ................................................................................................................................. 16  Figure 4. Crash Index – All Crashes. ............................................................................................ 36  Figure 5. Crash Index – Rural Crashes. ........................................................................................ 37  Figure 6. Crash Index – Rural CMV Crashes. .............................................................................. 38  Figure 7. Top 10 Manners of Collision (2010–2013 Crash Data). ............................................... 39  Figure 8. Top 10 Vehicle Contributing Factors (2010–2013 Crash Data). .................................. 40  Figure 9. Changes in Crash Rate (per 100 Million VMTs) from 2006–2009 to 2010–2013 – All Crashes and Rural Crashes. ............................................................................................. 53  Figure 10. Changes in Crash Rate (per 100 Million VMTs) from 2006–2009 to 2010–2013 – CMV Crashes and Rural CMV Crashes. ............................................................................... 54  Figure 11. Changes in Crash Rate (per 100 Lane-Miles) from 2006–2009 to 2010–2013 – All Crashes and Rural Crashes. ............................................................................................. 55  Figure 12. Changes in Crash Rate (per 100 Lane-Miles) from 2006–2009 to 2010–2013 – CMV Crashes and Rural CMV Crashes. ............................................................................... 56  Figure 13. Change in the Number of Rural KAB CMV Crashes vs. Change in the Number of Horizontal Wells. ................................................................................................................... 59  Figure 14. Observed versus Predicted Number of Rural KAB CMV Crashes in 2013. ............... 61  4 List of Tables Table 1. Pearson Correlation Coefficients. ................................................................................... 10  Table 2. Changes in Economic and Comprehensive Costs for Injuries Resulting from Rural CMV Crashes......................................................................................................................... 12  Table 3. Number of Crashes. ........................................................................................................ 19  Table 4. Number of Rural Crashes. .............................................................................................. 20  Table 5. Number of CMV Crashes. .............................................................................................. 21  Table 6. Number of Rural CMV Crashes. .................................................................................... 22  Table 7. Number of Crashes on State Highways. ......................................................................... 23  Table 8. Number of Crashes on Rural State Highways. ............................................................... 24  Table 9. Number of CMV Crashes on State Highways. ............................................................... 25  Table 10. Number of CMV Crashes on Rural State Highways. ................................................... 26  Table 11. Number of Injuries. ....................................................................................................... 27  Table 12. Number of Injuries in Rural Crashes. ........................................................................... 28  Table 13. Number of Injuries in CMV Crashes. ........................................................................... 29  Table 14. Number of Injuries in Rural CMV Crashes. ................................................................. 30  Table 15. Number of Injuries in Crashes on State Highways. ...................................................... 31  Table 16. Number of Injuries in Crashes on Rural State Highways. ............................................ 32  Table 17. Number of Injuries in CMV Crashes on State Highways............................................. 33  Table 18. Number of Injuries in CMV Crashes on Rural State Highways. .................................. 34  Table 19. Changes in the Number of New Completed Wells. ...................................................... 41  Table 20. Changes in the Number of Crashes on All Highways. ................................................. 42  Table 21. Changes in the Number of Crashes on State Highways. .............................................. 43  Table 22. Changes in the Number of Injuries on All Highways. .................................................. 44  Table 23. Changes in the Number of Injuries on State Highways. ............................................... 45  Table 24. Number of Crashes over Four Years per 100 Million VMTs. ...................................... 50  Table 25. Number of Crashes over Four Years per 100 Lane-Miles. ........................................... 51  Table 26. Rudimentary Thresholds to Interpret Pearson Correlation Coefficients. ..................... 57  Table 27. Pearson Correlation Coefficients. ................................................................................. 57  Table 28. Pearson Correlation Coefficients – Change in the Number of Rural KAB CMV Crashes vs. Change in the Number of New Wells from 2006–2009 to 2010–2013. ............. 58  Table 29. Average Coefficient of Variation. ................................................................................ 63  Table 30. Injury Severity Levels and Estimated Comprehensive Costs (Adapted from 4). ......... 64  Table 31. Changes in Economic and Comprehensive Costs for Injuries Occurred in Rural CMV Crashes......................................................................................................................... 65  Table 32. Changes in the Number of Crashes on All Highways. ................................................. 68  Table 33. Changes in the Number of Injuries on All Highways. .................................................. 69  Table 34. Number of Crashes over Four Years per 100 Lane-Miles. ........................................... 70  Table 35. Changes in Economic and Comprehensive Costs for Injuries Occurred in Rural CMV Crashes......................................................................................................................... 72  5 Table 36. Summary Results of Statistical Analysis for Model 1. ................................................. 75  Table 37. Summary Results of Statistical Analysis for Model 2. ................................................. 75  Table 38. Summary Results of Statistical Analysis for Model 3. ................................................. 82  6 Executive Summary In fall 2014, the Texas Legislature asked the Texas A&M Transportation Institute (TTI) to update a study completed in late 2011 documenting locations and trends of oil and gas energy developments in the state. As part of the study, the Texas Legislature asked TTI to correlate oil and gas developments with changes in pavement condition data. TTI summarized the results of this analysis in a report published in March 2015. To complement the study, the Texas Legislature asked TTI to gather and process crash data at a level of spatial and temporal detail needed to document locations and trends of crashes in relation to oil and gas energy developments in the state. Location and attribute data about crashes and injuries (i.e., number of people who are injured in crashes) that the research team compiled included the following types of crashes:  All crashes.  Rural crashes (i.e., crashes that occur outside city limits).  Crashes in which commercial motor vehicles (CMVs) are involved.  Rural CMV crashes.  Crashes on state highways.  Crashes on rural state highways.  CMV crashes on state highways.  CMV crashes on rural state highways. With this information, the research team examined changes in the number of crashes and the number of injuries from 2006–2009 to 2010–2013. These date ranges were used for consistency with those in the original March 2015 report. The year 2009 was significant because this was when accelerated oil production started in the Eagle Ford Shale region and oil production in the Permian Basin region began to accelerate, making the end of 2009 suitable for use as a baseline for comparison purposes. The last year with reliable Railroad Commission data was 2013 (2014 data were still preliminary). In addition, the economic recession of 2008 caused significant volatility in the oil markets, which resulted in dramatic swings in prices, drilling, and production. In order to reduce the impact of these variations, the research team aggregated and compared data using two four-year blocks: 2006–2009 and 2010–2013. The total number of crashes decreased by 10 percent in the Barnett Shale region, increased by 1 percent in the Eagle Ford Shale region, and decreased by 4 percent in the Permian Basin region. As a reference, the number of crashes decreased by 7 percent in all other 175 counties in the state. However, these changes were not uniform either by crash location and type of vehicles 7 involved or by injury severity. There were also significant differences geographically within each region. In general:  Changes were more prominent for rural crashes. In the Barnett Shale region, the number of rural crashes decreased by 25 percent (compared to a 10 percent decrease overall in the region). In the Eagle Ford Shale region, the number of rural crashes increased by 4 percent (compared to a 1 percent increase overall in the region). In the Permian Basin region, the number of crashes increased by 11 percent (compared to a 4 percent decrease overall in the region).  Changes were even more prominent for crashes that involved CMVs and, in particular, for rural crashes that involved CMVs. For rural crashes that involved CMVs, there was a 34 percent decrease in the Barnett Shale region, a 61 percent increase in the Eagle Ford Shale region, and a 52 percent increase in the Permian Basin region. By comparison, there was a 9 percent decrease in all other 175 counties in the state.  For rural CMV crashes, changes in the relative number of crashes were larger as the severity of the injuries worsened. For example, in the Eagle Ford Shale region, there was a 77 percent increase in the number of fatal, incapacitating, and non-incapacitating injury crashes (compared to a 61 percent increase for all rural CMV crashes). For fatal crashes, the increase was 76 percent. In the Permian Basin region, there was a 57 percent increase in the number of fatal, incapacitating, and non-incapacitating injury crashes (compared to a 52 percent increase for all rural CMV crashes). For fatal crashes, the increase was 88 percent. The exception to this trend was the Barnett Shale region, where there was a 26 percent decrease in the number of fatal, incapacitating, and non-incapacitating injury crashes (compared to a 34 percent decrease for all crashes). For fatal crashes, the decrease was 37 percent.  Relative changes in the number of crashes on state highways were similar to those found for all highways. The changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. Overall, the percentage of crashes occurring on state highways increased. For all crashes, the increase was from 54 to 56 percent. For fatal, incapacitating, and non-incapacitating injury crashes, the increase was from 59 to 61 percent. For fatal crashes, the increase was from 74 to 76 percent. These percentages were higher for rural roads. For example, for rural CMV crashes, the percentage of crashes on state highways decreased slightly from 78 to 77 percent. For fatal, incapacitating, and non-incapacitating injury crashes, this percentage increased from 90 to 91 percent. For fatal crashes, it decreased slightly but stayed around 96 percent. 8 The research team calculated crash rates expressed both as the number of crashes per 100 million vehicle miles traveled (VMT) and number of crashes per 100 lane-miles. The results were similar with both approaches, although rates expressed as the number of crashes per 100 lanemiles were more stable particularly for roadway segments with low traffic volumes. In total, the crash rate decreased by 4 percent in the Barnett Shale region, increased by 7 percent in the Eagle Ford Shale region, and increased by 11 percent in the Permian Basin region. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. The changes were more prominent for rural crashes. The changes were even more prominent for crashes that involved CMVs and, in particular, for rural crashes that involved CMVs. In most cases, as the severity of the injuries worsened, the changes in the corresponding crash rate were more evident. The research team established correlations by comparing pairs of metrics representing historical data aggregated at the county level (Table 1). In the Barnett Shale region, there was a strong correlation between the number of new horizontal wells and the number of crashes (regardless of location or type of vehicles involved). In the Eagle Ford Shale region, there was a strong correlation between the number of new horizontal wells and the number of rural CMV crashes. In the Permian Basin region, there was a strong correlation between the number of new vertical wells and the number of rural CMV crashes. It is worth noting that In the Permian Basin, although the relative change in the number of horizontal wells was considerably higher than for vertical wells, in absolute terms the number of new vertical wells was much higher than the number of new horizontal wells. Judging from the trends in the Eagle Ford Shale and Barnett Shale regions, as the industry shifts from vertical drilling to horizontal drilling in the Permian Basin, the correlation between new horizontal wells and rural CMV crashes in that part of the state will likely increase. 9 Table 1. Pearson Correlation Coefficients. Number of Crashes Barnett Shale  Number of Rural Crashes Region Number of CMV‐Involved Crashes Number of Rural CMV‐Involved Crashes Number of Crashes Eagle Ford  Number of Rural Crashes Shale Region Number of CMV‐Involved Crashes Number of Rural CMV‐Involved Crashes Number of Crashes Permian  Number of Rural Crashes Basin Region Number of CMV‐Involved Crashes Number of Rural CMV‐Involved Crashes Number of Crashes Remaining  Number of Rural Crashes Counties Number of CMV‐Involved Crashes Number of Rural CMV‐Involved Crashes Number of New  Number of New  Horizontal Wells Vertical Wells 0.56 ‐0.15 0.62 ‐0.24 0.52 ‐0.19 0.63 ‐0.10 0.12 0.16 ‐0.07 ‐0.22 0.39 0.09 0.57 ‐0.10 ‐0.08 0.07 0.03 0.33 0.06 0.29 0.23 0.47 ‐0.04 ‐0.03 ‐0.03 0.00 ‐0.02 ‐0.02 0.00 0.06 In the Barnett Shale region, there was a very strong correlation between the change in the number of new horizontal wells and the change in the number of rural CMV crashes. In the Eagle Ford Shale region, the correlation between these two variables was also very strong. In the Permian Basin region, there was a strong correlation between the change in the number of new vertical wells and that of rural CMV crashes. The research team used this information to develop linear regression models (Figure 1) for county-level data from the Eagle Ford Shale, Barnett Shale, and Permian Basin regions that could be used for forecasting purposes in situations where other factors remain reasonably stable and there is a need for high-level estimates. These models suggest a generalized trend that could be used to estimate positive (or negative) changes in the number of rural CMV crashes in Texas as a function of the positive (or negative) change in the number of new horizontal wells. 10 Figure 1. Change in the Number of Rural Fatal, Incapacitating, and Non-Incapacitating CMV Crashes vs. Change in the Number of Horizontal Wells. The research team developed preliminary estimates of the change in the cost of injuries from 2006–2009 to 2010–2013 using standardized economic and comprehensive crash cost estimates from the National Safety Council (NSC) and comprehensive crash cost estimates from the U.S. Department of Transportation (DOT). Economic costs rely on calculable costs such as wage and productivity losses, medical expenses, administrative expenses, motor vehicle damage, and employers’ uninsured costs). Comprehensive costs include economic cost components and a measure of the value of lost quality of life, which makes comprehensive costs appropriate to analyze the anticipated benefit of future improvements (because they provide a measure of what people would be willing to pay for improved safety). In general, the U.S. DOT’s methodology for comprehensive cost estimates, which are based on a concept called the value of a statistical life (VSL), are considerably higher than those resulting from the NSC methodology. Table 2 summarizes the result of the analysis. Because the correlation between new completed wells and rural CMV crashes was stronger than for other types of crashes, the research team only included the number of injuries resulting from rural CMV crashes. Further, the research team only included the number of fatal, incapacitating, non-incapacitating, and possible injuries in the cost calculation. In the Barnett Shale region, there was a 35 percent decrease (i.e., $73 million in economic costs or $425 million in comprehensive costs) in NSC-based costs and a 30 percent decrease (i.e., $763 million) in VSL-based comprehensive costs. The cost reduction was the result of fewer rural CMV crashes and, correspondingly, fewer injuries. In the Eagle Ford Shale 11 region, there was a 52 percent increase (i.e., $139 million in economic costs or $801 million in comprehensive costs) in NSC-based costs and a 68 percent increase (i.e., $2 billion) in VSLbased comprehensive costs. In the Permian Basin region, there was a 103 percent increase (i.e., $176 million in economic costs or $1.03 billion in comprehensive costs) in NSC-based costs and a 97 percent increase (i.e., $2 billion) in VSL-based comprehensive costs. The huge increase in the cost of injuries resulting from rural CMV crashes in the Eagle Ford Shale and Permian Basin regions (covering 66 counties in total) was largely responsible for the net increase in the cost of injuries resulting from rural CMV crashes in the state from 2006–2009 to 2010–2013. As Table 2 shows, the net increase was 9 percent overall, even though there was a 35 percent reduction in the Barnett Shale region (covering 13 counties) and a 3 percent reduction in 175 other counties around the state. Table 2. Changes in Economic and Comprehensive Costs for Injuries Resulting from Rural CMV Crashes. Cost of Rural CMV Injuries (Million) Economic Cost (NSC) Comprehensive Cost (NSC) Region 2006‐09 2010‐13 Change Diff. 2006‐09 2010‐13 Change Diff. Barnett Shale $       212 $       138 $     (73) ‐35% $   1,224 $       799 $   (425) ‐35% Eagle Ford Shale $       269 $       408 $     139 52% $   1,548 $   2,349 $     801 52% Permian Basin $       171 $       348 $     176 103% $       981 $   2,011 $ 1,030 105% Other $   1,615 $   1,567 $     (47) ‐3% $   9,229 $   8,988 $   (241) ‐3% Grand Total $   2,266 $   2,461 $     194 9% $ 12,981 $ 14,146 $ 1,165 9% Comprehensive Cost (VSL) 2006‐09 2010‐13 Change Diff. $   2,510 $   1,747 $   (763) ‐30% $   2,931 $   4,927 $ 1,996 68% $   2,051 $   4,045 $ 1,994 97% $ 19,796 $ 19,205 $   (591) ‐3% $ 27,288 $ 29,924 $ 2,636 10% In practical terms, the research results mean the following (Figure 2):  The number of crashes and resulting injuries increased along with oil and gas well development activities, but the changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region.  The increases in the number of crashes and injuries were more prominent in rural areas where energy developments take place (i.e., Eagle Ford Shale and Permian Basin regions). The highest increase was in the case of rural CMV crashes. Overall, there was a strong correlation between rural CMV crashes and the number of new wells.  The percentage of crashes on state highways increased. As the severity of the injuries increased, the percentage of crashes on state highways also increased. For rural CMV crashes, the percentage of crashes on state highways increased from 81 to 83 percent. For fatal, incapacitating, and non-incapacitating injury crashes, this percentage increased from 89 to 90 percent. For fatal crashes, it increased slightly but stayed around 95 percent. 12  The cost of injuries resulting from rural CMV crashes in energy development regions increased significantly and was largely responsible for the net increase in the cost of injuries resulting from rural CMV crashes in the state from 2006–2009 to 2010–2013. In the Eagle Ford Shale region, the increase was $139 million in economic costs or $801 million–$2 billion in comprehensive costs. In the Permian Basin region, the increase was $176 million in economic costs or $1.03–2.0 billion in comprehensive costs. These costs are of the same order of magnitude as the impact of energy developments on the transportation infrastructure (estimated at $1 billion per year on state highways and an additional $1 billion on county and local roads). 13 Figure 2. Overall Trends in Well Development, Rural CMV Crashes, and Crash Injury Costs from 2006– 2009 to 2010–2013. 14 Chapter 1. Crash Data Collection and Processing Introduction In fall 2014, the Texas Legislature asked TTI to update a study completed in late 2011 documenting locations and trends of oil and gas energy developments in the state (1). As part of the study, the Texas Legislature asked TTI to correlate oil and gas developments with changes in pavement condition data. TTI summarized the results of this analysis in a report published in March 2015 (2). To complement the study, the Texas Legislature asked TTI to gather and process crash data at a level of spatial and temporal detail needed to document locations and trends of crashes in relation to oil and gas energy developments in the state. The research team completed the following activities to document trends and changes in crash rates and to correlate this information with changes in oil and gas drilling developments:  Request and process crash data from the Texas Department of Transportation (TxDOT).  Analyze and correlate trends with oil and gas drilling developments.  Prepare deliverables. Figure 3 shows the location of counties associated with the Barnett Shale region (13 counties), Eagle Ford Shale region (29 counties), and Permian Basin region (37 counties) that were used in the March 2015 report to document changes in oil and gas well developments in the state. Of specific interest was to examine changes with respect to a pre-determined baseline. The research team used the end of 2009 for baseline and comparison purposes because this was when well drilling and oil production in the Eagle Ford Shale region began in earnest and when oil production in the Permian Basin began to accelerate. The last year in which the research team received reliable data from the Railroad Commission was 2013 (2014 data are still preliminary). In addition, the economic recession of 2008 caused significant volatility in the oil markets, which resulted in dramatic swings in prices, drilling, and production. In order to reduce the impact of these variations, the research team aggregated and compared data using two four-year blocks: 2006–2009 and 2010–2013. 15 Barnett Shale Region Permian Basin Region Eagle Ford Shale Region Figure 3. Counties Analyzed in the Eagle Ford Shale, Permian Basin, and Barnett Shale Regions. Crash Data The research team gathered and processed data from TxDOT’s Crash Record Information System (CRIS) at a level of spatial and temporal detail needed to document locations and trends of crashes in relation to oil and gas energy developments in the state. Available data from CRIS covered the 2010–2014 period. The research team complemented this information with historical crash data from 2003–2009 that TTI had received from TxDOT before the introduction of CRIS. Consistent with protocols between TTI and TxDOT for the release of historical crash data, the research team worked through a designated official at the TTI Traffic Safety Center to download and furnish relevant crash data. The focus was on crash locations and basic crash attribute data such as severity, number and type of vehicles involved, manner of collision, contributing factors, and type of roadway. The crash data available to the research team included generic information about the vehicles involved in crashes (such as vehicle type and vehicle body type), but not specific vehicle identifiable information. The data included information about the number of people injured in crashes (and the corresponding injury level, e.g., fatal, incapacitating injury, 16 non-incapacitating injury, possible injury, no injury, or unknown), but did not include identifiable information about those people. Beginning with 2008, crash data records were sufficiently complete to conduct the analysis. Prior to 2008, there were gaps in certain data categories. Specifically, rural crashes, CMV crashes, and injuries (i.e., the number of people who were injured in crashes). The gaps were not uniform. For example, the number of rural crashes in CRIS was practically nonexistent from 2003–2005, but appeared to be reasonable for subsequent years. Likewise, the number of fatal CMV crashes in CRIS for 2006 and 2007 appeared to be reasonable, but the number of CMV crashes for other injury types during the same two years appeared to be much lower than what the historical trend would suggest (based on data from subsequent years). The number of CMV crashes in CRIS from 2003–2005 was also practically nonexistent. Because it was of interest to compare crash data for two four-year blocks (2006–2009 and 2010– 2013) to enable a comparison with oil and gas well developments during the same periods, it was necessary to estimate missing CMV crash and injury data for 2006 and 2007. Fortunately, the CRIS data contained reliable information about vehicles that were involved in crashes. This record appeared reasonably complete for all the years of interest. Based on this information, the process to estimate missing CMV crash and injury data was as follows:  Determine the number of trucks (i.e., the number of CMVs) that were involved in crashes in 2008. The research team classified the following vehicle types as trucks: truck and trailer, semi-trailer, mobile home, other truck combinations, farm tractor, road machinery, and other machinery.  Determine the ratio of number of crashes in 2008 to the number of CMVs in 2008. For accuracy, the research team determined a separate ratio for crashes according to severity (i.e., fatal, incapacitating injury, non-incapacitating injury, possible injury, no injury, or unknown) and according to region (i.e., Barnett Shale, Eagle Ford, Permian Basin, and other).  Estimate the number of CMV crashes in 2006 and 2007 by dividing the number of crashes in 2006 and 2007 by the corresponding ratio of number of crashes in 2008 to the number of CMVs in 2008. This process resulted in four estimated sets of CMV crashes for 2006 and 2007.  Determine the ratio of number of injuries in CMV-related crashes in 2008 to the number of CMVs in 2008. For accuracy, the research team determined a separate ratio for injuries according to severity (i.e., fatal, incapacitating injury, non-incapacitating injury, possible injury, no injury, or unknown) and according to region (i.e., Barnett Shale, Eagle Ford, Permian Basin, and other). 17  Estimate the number of injuries in 2006 and 2007 by dividing the number of injuries in 2006 and 2007 by the corresponding ratio of number of injuries in 2008 to the number of CMVs in 2008. This process resulted in four estimated sets of injuries for 2006 and 2007. Table 3 through Table 18 summarize the result of the crash data compilation, as follows:  Number of crashes: o Table 3: Number of crashes. o Table 4: Number of rural crashes. o Table 5: Number of CMV crashes. o Table 6: Number of rural CMV crashes. o Table 7: Number of crashes on state highways. o Table 8: Number of crashes on rural state highways. o Table 9: Number of CMV crashes on state highways. o Table 10: Number of CMV crashes on rural state highways.  Number of injuries: o Table 11: Number of injuries. o Table 12: Number of injuries in rural crashes. o Table 13: Number of injuries in CMV crashes. o Table 14: Number of injuries in rural CMV crashes. o Table 15: Number of injuries in crashes on state highways. o Table 16: Number of injuries in crashes on rural state highways. o Table 17: Number of injuries in CMV crashes on state highways. o Table 18: Number of injuries in CMV crashes on rural state highways. In these tables, a rural crash is a crash that occurs on a roadway outside city limits. An on-system crash is a crash that occurs on a state highway. A rural on-system crash is a crash that occurs on a state highway outside city limits. Years marked with a ‘*’ (i.e., 2006 and 2007) correspond to CMV crashes or injuries that were estimated following the procedure described above. 18 Table 3. Number of Crashes. Number of Crashes ‐ Statewide Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           3,118           3,101           3,126           2,821           2,781           2,803           3,037           3,054 Incapacit.         13,596         13,250         12,562         11,466         11,813         11,753         12,870         13,436 Non‐Incap. Possible Inj.         50,480      108,914         51,178      107,870         49,652         97,538         47,425         95,798         48,394         81,658         46,570         81,321         50,836         88,583         52,205         88,799 No Injury      243,094      263,334      257,155      251,891      234,050      228,491      247,674      272,579 Unknown         17,985         19,252         18,962         18,929         12,880         12,706         14,014         15,171 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               224               211               218               198               188               204               274               236 Incapacit. Non‐Incap. Possible Inj.               732           2,678           4,098               690           2,770           4,118               695           2,754           3,788               686           2,526           3,698               693           2,595           3,432               741           2,826           3,756               804           2,866           4,089               840           2,997           4,031 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               311               320               294               277               251               257               263               259 Incapacit. Non‐Incap. Possible Inj. No Injury Unknown Total           1,738           6,003         11,126         24,324           1,879         45,381           1,769           6,429         10,935         27,132           2,067         48,652      184,735           1,667           5,891           9,821         26,063           2,090         45,826           1,500           5,540           9,515         25,954           2,090         44,876           1,598           5,705           8,309         23,128           1,359         40,350           1,552           5,471           8,166         22,351           1,312         39,109      166,474           1,658           5,923           8,534         24,185           1,241         41,804           1,650           6,141           8,742         26,967           1,452         45,211 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               171               152               174               151               167               169               205               248 Incapacit. Non‐Incap. Possible Inj.               499           2,144           4,166               532           2,326           4,185               476           2,292           4,161               410           2,193           4,036               410           2,090           3,478               478           2,238           3,662               525           2,408           4,066               602           2,479           4,204 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal Incapacit. Non‐Incap. Possible Inj.           2,412         10,627         39,655         89,524           2,418         10,259         39,653         88,632           2,440           9,724         38,715         79,768           2,195           8,870         37,166         78,549           2,175           9,112         38,004         66,439           2,173           8,982         36,035         65,737           2,295           9,883         39,639         71,894           2,311         10,344         40,588         71,822 Total      437,187      457,985   1,762,497      438,995      428,330      391,576      383,644   1,637,478      417,014      445,244 Number of Crashes ‐ Eagle Ford Shale Region No Injury         12,642         13,533         13,660         13,044         12,737         13,051         13,820         14,201 Unknown               707               778               832               684               538               579               619               627 Total         21,081         22,100         85,964         21,947         20,836         20,183         21,157         86,744         22,472         22,932 Number of Crashes ‐ Barnett Shale Region Number of Crashes ‐ Permian Basin Region No Injury         11,710         12,945         12,149         12,198         11,099         10,821         12,198         13,059 Unknown               836               725           1,059           1,201               443               651               889               922 Total         19,526         20,865         80,891         20,311         20,189         17,687         18,019         77,511         20,291         21,514 Number of Crashes ‐ Remaining Counties 19 No Injury      194,418      209,724      205,283      200,695      187,086      182,268      197,471      218,352 Unknown         14,563         15,682         14,981         14,954         10,540         10,164         11,265         12,170 Total      351,199      366,368   1,410,907      350,911      342,429      313,356      305,359   1,306,749      332,447      355,587 Table 4. Number of Rural Crashes. Number of Rural Crashes ‐ Statewide Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,642           1,610           1,542           1,397           1,358           1,390           1,543           1,539 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               162               165               161               141               138               162               217               177 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               119               118               126                 96                 83                 70                 91                 81 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               120               110               103                 97                 96               115               139               168 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,241           1,217           1,152           1,063           1,041           1,043           1,096           1,113 Incapacit.           4,798           4,529           4,212           3,887           3,895           3,978           4,188           4,661 Non‐Incap. Possible Inj.         12,831         15,938         12,773         16,079         12,654         14,658         11,497         13,782         11,878         12,273         11,242         12,430         11,863         13,850         12,225         14,034 No Injury Unknown Total         51,326           2,744         89,279         55,654           2,968         93,613      356,636         52,876           3,192         89,134         51,124           2,923         84,610         51,719           2,263         83,386         50,202           1,936         81,178      349,182         54,614           2,209         88,267         61,377           2,515         96,351 Number of Rural Crashes ‐ Eagle Ford Shale Region Incapacit. Non‐Incap. Possible Inj. No Injury               447           1,230           1,058           4,084               392           1,228           1,068           4,308               378           1,173               918           4,003               383           1,029               936           3,549               363               964               885           3,752               423           1,105               946           3,854               477           1,176           1,182           4,575               530           1,216           1,170           4,814 Unknown               208               212               250               177               158               151               187               182 Total           7,189           7,373         27,660           6,883           6,215           6,260           6,641         28,804           7,814           8,089 Number of Rural Crashes ‐ Barnett Shale Region Incapacit.               405               383               378               292               243               281               271               260 Non‐Incap. Possible Inj.               880               877               883               898               924               867               742               658               699               565               682               519               695               542               709               482 No Injury           3,622           4,326           3,928           3,247           2,925           2,837           2,872           3,167 Unknown               181               214               233               175               125                 98               113               111 Total           6,084           6,822         24,572           6,456           5,210           4,640           4,487         18,521           4,584           4,810 Number of Rural Crashes ‐ Permian Basin Region Incapacit.               226               216               214               168               152               215               253               357 Non‐Incap. Possible Inj.               633               577               707               609               663               523               584               500               670               360               657               515               780               610               922               561 No Injury Unknown Total           2,202               100           3,858           2,607                 80           4,329         15,689           2,328               105           3,936           2,129                 88           3,566           2,104                 65           3,447           2,329                 49           3,880         17,426           2,813                 65           4,660           3,339                 92           5,439 Number of Rural Crashes ‐ Remaining Counties Incapacit. Non‐Incap. Possible Inj.           3,720         10,088         13,426           3,538           9,955         13,504           3,242           9,894         12,350           3,044           9,142         11,688           3,137           9,545         10,463           3,059           8,798         10,450           3,187           9,212         11,516           3,514           9,378         11,821 20 No Injury Unknown Total         41,418           2,255         72,148         44,413           2,462         75,089      288,715         42,617           2,604         71,859         42,199           2,483         69,619         42,938           1,915         69,039         41,182           1,638         66,170      284,431         44,354           1,844         71,209         50,057           2,130         78,013 Table 5. Number of CMV Crashes. Number of CMV Crashes ‐ Statewide Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               409               440               432               301               393               381               462               456 Incapacit. Non‐Incap. Possible Inj.               976           2,927           5,553               998           3,076           5,754               925           3,005           5,088               743           2,439           4,367               891           2,731           4,097               825           2,584           3,883               922           2,831           4,207           1,073           3,063           4,279 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 33                 38                 28                 43                 34                 73                 34                 46                 27                 82                 40                 86                 73               105                 64               127 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 38                 55                 49                 39                 40                 31                 33                 31 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 27                 39                 28                 52                 23                 48                 16                 34                 26                 38                 40                 63                 57                 83                 60               117 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               311               328               326               212               300               270               299               301 No Injury         18,624         21,120         20,021         16,900         16,653         16,451         17,972         20,402 Unknown               483               443               304               254               133               144               163               202 Total         28,972         31,831      115,582         29,775         25,004         24,898         24,268      105,198         26,557         29,475 Number of CMV Crashes ‐ Eagle Ford Shale Region Non‐Incap. Possible Inj.               209               197               220               228               177               213               160               169               202               209               227               263               275               320               333               368 No Injury Unknown Total           1,122                 40           1,639           1,322                 28           1,870           6,607           1,236                 17           1,750               923                 16           1,348           1,114                   7           1,641           1,290                 17           1,923           8,708           1,670                 17           2,460           1,778                 14           2,684 Number of CMV Crashes ‐ Barnett Shale Region Incapacit.               132               127               145               120               107               103               129               105 Non‐Incap. Possible Inj.               343               596               408               621               387               559               282               429               329               418               285               409               301               401               352               431 No Injury Unknown Total           2,149                 66           3,323           2,732                 47           3,989         14,119           2,675                 47           3,862           2,043                 32           2,945           2,065                 17           2,976           2,090                 13           2,931         12,367           2,156                 15           3,035           2,486                 20           3,425 Number of CMV Crashes ‐ Permian Basin Region Non‐Incap. Possible Inj.               132               149               170               202               165               194               149               140               156               136               193               179               217               223               283               221 No Injury Unknown Total               704                 19           1,069               912                   8           1,372           4,775               849                 14           1,293               692                 10           1,041               846                   6           1,208               908                 10           1,393           6,368           1,145                   4           1,729           1,346                 11           2,038 Number of CMV Crashes ‐ Remaining Counties Incapacit. Non‐Incap. Possible Inj.               766           2,244           4,611               776           2,279           4,703               659           2,276           4,122               543           1,848           3,629               664           2,044           3,334               573           1,879           3,032               605           2,038           3,263               724           2,095           3,259 21 No Injury         14,649         16,154         15,261         13,242         12,628         12,163         13,001         14,792 Unknown               360               360               226               196               103               104               127               157 Total         22,941         24,600         90,081         22,870         19,670         19,073         18,021         77,755         19,333         21,328 Table 6. Number of Rural CMV Crashes. Number of Rural CMV Crashes ‐ Statewide Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               250               263               262               171               223               235               300               319 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 27                 24                 25                 26                 22                 37                 64                 56 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 24                 24                 35                 17                 18                 11                 14                 20 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 21                 25                 20                 14                 21                 33                 45                 52 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               178               190               182               114               162               154               177               191 Incapacit. Non‐Incap. Possible Inj.               415           1,069           1,130               420           1,226           1,349               421           1,189           1,233               305               927           1,021               412           1,032               937               396           1,032               944               443           1,106           1,096               558           1,313           1,179 No Injury Unknown Total           5,437                 85           8,387           6,380               109           9,747         34,635           6,020                 50           9,175           4,865                 37           7,326           5,014                 22           7,640           5,257                 31           7,895         34,288           5,756                 23           8,724           6,636                 24         10,029 Number of Rural CMV Crashes ‐ Eagle Ford Shale Region Incapacit.                 22                 26                 54                 29                 55                 64                 85                 95 Non‐Incap. Possible Inj.               115                 77               123               103               106                 71                 85                 77               114                 87               163               111               186               173               232               217 No Injury Unknown Total               423                 16               679               528                 12               816           2,820               482                   2               740               368               ‐               585               439                   2               719               578                   6               959           4,542               823                   7           1,338               921                   5           1,526 Number of Rural CMV Crashes ‐ Barnett Shale Region Incapacit. Non‐Incap. Possible Inj.                 51                 87                 85                 41               109               102                 57               108               103                 27                 61                 54                 28                 73                 38                 34                 66                 41                 37                 63                 38                 22                 88                 47 No Injury Unknown Total               490                   5               742               638                 11               925           3,130               650                   3               956               347                   1               507               324               ‐               481               352                   2               506           2,061               364               ‐               516               380                   1               558 Number of Rural CMV Crashes ‐ Permian Basin Region Incapacit.                 24                 31                 31                 27                 25                 48                 62                 96 Non‐Incap. Possible Inj.               103                 74               125                 89               101                 83                 94                 51                 98                 56               135                 94               154               118               202               106 No Injury Unknown Total               348                   6               576               451                   1               722           2,464               426                   3               664               314                   2               502               408                   1               609               513                   1               824           3,743               663                   1           1,043               809                   2           1,267 Number of Rural CMV Crashes ‐ Remaining Counties Incapacit.               319               323               279               222               304               250               259               345 Non‐Incap. Possible Inj.               764               895               868           1,056               874               976               687               839               747               756               668               698               703               767               791               809 22 No Injury Unknown Total           4,177                 58           6,390           4,763                 85           7,284         26,221           4,462                 42           6,815           3,836                 34           5,732           3,843                 19           5,831           3,814                 22           5,606         23,942           3,906                 15           5,827           4,526                 16           6,678 Table 7. Number of Crashes on State Highways. Number of On‐System Crashes ‐ Statewide Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           2,277           2,285           2,267           2,037           2,068           2,077           2,244           2,303 Incapacit.           8,112           7,828           7,491           6,921           7,396           7,306           7,972           8,485 Non‐Incap. Possible Inj.         27,954         56,602         28,554         56,382         27,257         51,376         26,496         51,213         28,007         45,509         26,960         45,062         29,020         49,141         30,096         48,968 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               196               182               179               165               160               184               241               207 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal Incapacit. Non‐Incap. Possible Inj.               232           1,060           3,435           6,158               243           1,033           3,711           6,055               214           1,026           3,352           5,503               198               917           3,215           5,467               190               964           3,442           5,002               196               980           3,311           4,822               190           1,025           3,520           5,166               185           1,034           3,729           5,161 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               132               109               121               120               131               139               173               204 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,717           1,751           1,753           1,554           1,587           1,558           1,640           1,707 No Injury Unknown Total      128,073           4,371      227,389      139,734           4,686      239,469      921,379      135,906           4,192      228,489      135,191           4,174      226,032      129,438           2,864      215,282      125,877           2,740      210,022      897,671      136,608           3,060      228,045      151,109           3,361      244,322 Number of On‐System Crashes ‐ Eagle Ford Shale Region Incapacit. Non‐Incap. Possible Inj.               559           1,859           2,643               520           1,860           2,623               523           1,842           2,428               511           1,726           2,537               529           1,792           2,315               595           1,980           2,547               646           2,025           2,818               678           2,171           2,767 No Injury           7,914           8,313           8,459           8,125           8,245           8,563           9,286           9,809 Unknown               222               232               234               168               138               160               169               176 Total         13,393         13,730         54,020         13,665         13,232         13,179         14,029         58,201         15,185         15,808 Number of On‐System Crashes ‐ Barnett Shale Region No Injury         13,562         15,206         14,565         14,797         13,844         13,364         14,395         15,955 Unknown               445               493               467               435               277               241               241               294 Total         24,892         26,741      101,789         25,127         25,029         23,719         22,914         97,528         24,537         26,358 Number of On‐System Crashes ‐ Permian Basin Region Incapacit. Non‐Incap. Possible Inj.               295           1,158           2,070               315           1,299           2,075               299           1,241           2,027               254           1,142           1,933               272           1,271           1,780               344           1,350           2,030               362           1,474           2,326               449           1,630           2,400 No Injury           5,584           6,410           5,684           5,845           5,774           5,973           6,822           7,699 Unknown               149               106               172               163                 71                 90               128               148 Total           9,388         10,314         38,703           9,544           9,457           9,299           9,926         43,040         11,285         12,530 Number of On‐System Crashes ‐ Remaining Counties Incapacit.           6,198           5,960           5,643           5,239           5,631           5,387           5,939           6,324 Non‐Incap. Possible Inj.         21,502         45,731         21,684         45,629         20,822         41,418         20,413         41,276         21,502         36,412         20,319         35,663         22,001         38,831         22,566         38,640 23 Total No Injury Unknown      101,013           3,555      179,716      109,805           3,855      188,684      726,867      107,198           3,319      180,153      106,424           3,408      178,314      101,575           2,378      169,085         97,977           2,249      163,153      698,902      106,105           2,522      177,038      117,646           2,743      189,626 Table 8. Number of Crashes on Rural State Highways. Number of Rural On‐System Crashes ‐ Statewide Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,333           1,327           1,249           1,125           1,098           1,122           1,260           1,268 Incapacit. Non‐Incap. Possible Inj. No Injury           3,494           8,973         10,405         33,306           3,321           9,058         10,593         36,406           3,045           8,715           9,558         33,859           2,754           7,956           8,873         32,467           2,873           8,200           7,857         32,360           2,846           7,786           7,843         31,492           3,152           8,231           8,774         34,330           3,564           8,541           8,733         38,310 Unknown               969           1,086           1,106               955               692               584               729               812 Total         58,480         61,791      231,933         57,532         54,130         53,080         51,673      222,457         56,476         61,228 Number of Rural On‐System Crashes ‐ Eagle Ford Shale Region Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               145               153               140               126               121               150               196               160 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal                 92                 97               104                 76                 71                 59                 75                 63 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal                 98                 90                 81                 86                 78               103               125               148 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               998               987               924               837               828               810               864               897 Incapacit.               376               328               321               317               313               371               416               474 Non‐Incap. Possible Inj.           1,013               859               996               873               926               738               843               777               799               730               938               811           1,004               997           1,044               998 No Injury           3,228           3,371           3,163           2,811           3,014           3,195           3,793           4,065 Unknown               129               124               133                 84                 78                 85               112               111 Total           5,750           5,845         21,974           5,421           4,958           5,055           5,550         23,975           6,518           6,852 Number of Rural On‐System Crashes ‐ Barnett Shale Region Incapacit.               296               273               283               194               171               195               227               205 Non‐Incap. Possible Inj.               606               611               630               597               636               579               511               451               482               376               481               373               459               377               510               339 No Injury Unknown Total           2,418                 59           4,082           2,850                 80           4,527         16,179           2,551                 82           4,235           2,061                 42           3,335           1,935                 35           3,070           1,874                 29           3,011         12,523           1,922                 38           3,098           2,186                 41           3,344 Number of Rural On‐System Crashes ‐ Permian Basin Region Incapacit.               173               165               173               132               131               190               220               311 Non‐Incap. Possible Inj.               523               455               565               478               521               411               452               384               524               281               537               404               631               501               777               473 No Injury Unknown Total           1,695                 36           2,980           2,064                 30           3,392         12,046           1,704                 46           2,936           1,643                 41           2,738           1,633                 23           2,670           1,871                 20           3,125         14,114           2,302                 28           3,807           2,756                 47           4,512 Number of Rural On‐System Crashes ‐ Remaining Counties Incapacit. Non‐Incap. Possible Inj.           2,649           6,831           8,480           2,555           6,867           8,645           2,268           6,632           7,830           2,111           6,150           7,261           2,258           6,395           6,470           2,090           5,830           6,255           2,289           6,137           6,899           2,574           6,210           6,923 24 No Injury         25,965         28,121         26,441         25,952         25,778         24,552         26,313         29,303 Unknown               745               852               845               788               556               450               551               613 Total         45,668         48,027      181,734         44,940         43,099         42,285         39,987      171,845         43,053         46,520 Table 9. Number of CMV Crashes on State Highways. Number of CMV On‐System Crashes ‐ Statewide Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               373               394               385               276               348               356               414               421 Incapacit. Non‐Incap. Possible Inj.               786           2,185           3,625               804           2,341           3,849               737           2,283           3,503               580           1,902           3,076               746           2,157           2,943               691           2,063           2,776               764           2,290           3,086               920           2,524           3,269 No Injury         12,282         14,017         13,549         11,553         11,736         11,740         13,011         14,747 Unknown               215               198               131                 89                 50                 65                 70                 83 Total         19,465         21,603         79,132         20,588         17,476         17,980         17,691         77,270         19,635         21,964 Number of CMV On‐System Crashes ‐ Eagle Ford Shale Region Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 33                 32                 28                 39                 30                 69                 33                 40                 26                 75                 40                 78                 71               100                 62               119 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 36                 50                 46                 35                 35                 30                 28                 29 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 29                 33                 29                 45                 21                 45                 15                 28                 24                 36                 39                 56                 54                 74                 55               108 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               277               288               288               193               263               247               261               275 Non‐Incap. Possible Inj.               183               160               198               175               156               159               134               149               168               175               206               226               249               286               309               325 No Injury Unknown Total               809                 10           1,226               942                   9           1,391           4,979               882                   1           1,297               705                   4           1,065               838                   5           1,287           1,000                 10           1,560           7,186           1,351                   8           2,065           1,455                   4           2,274 Number of CMV On‐System Crashes ‐ Barnett Shale Region Incapacit.               112               101               114                 89                 83                 83               110                 87 Non‐Incap. Possible Inj.               259               437               324               432               296               413               213               328               257               334               234               304               236               310               291               339 No Injury Unknown Total           1,466                 36           2,346           1,840                 20           2,768         10,010           1,882                 23           2,774           1,444                 13           2,122           1,486                   7           2,202           1,539                   4           2,194           9,278           1,605                   4           2,293           1,834                   9           2,589 Number of CMV On‐System Crashes ‐ Permian Basin Region Non‐Incap. Possible Inj.               113               122               143               163               132               149               120               109               126               107               168               139               178               176               248               177 No Injury Unknown Total               511                   9               816               687                   1           1,068           3,645               636                   6               989               495                   5               772               645                   1               939               721                   5           1,128           5,189               918                   3           1,403           1,125                   6           1,719 Number of CMV On‐System Crashes ‐ Remaining Counties Incapacit. Non‐Incap. Possible Inj.               609           1,629           2,906               618           1,676           3,078               509           1,699           2,782               423           1,435           2,490               552           1,606           2,327               474           1,455           2,107               480           1,627           2,314               606           1,676           2,428 25 No Injury           9,496         10,548         10,149           8,909           8,767           8,480           9,137         10,333 Unknown               161               168               101                 67                 37                 46                 55                 64 Total         15,077         16,376         60,499         15,528         13,517         13,552         12,809         55,617         13,874         15,382 Table 10. Number of CMV Crashes on Rural State Highways. Number of Rural CMV On‐System Crashes ‐ Statewide Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               240               252               241               164               208               230               286               304 Incapacit.               366               380               383               267               377               357               397               523 Non‐Incap. Possible Inj.               930               891           1,074           1,091           1,032               986               832               830               911               767               911               760               981               901           1,178           1,001 No Injury Unknown Total           4,281                 49           6,757           5,086                 71           7,955         28,178           4,843                 29           7,514           3,846                 12           5,951           4,039                 12           6,314           4,291                 16           6,565         28,598           4,694                 16           7,275           5,425                 13           8,444 Number of Rural CMV On‐System Crashes ‐ Eagle Ford Shale Region Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 26                 24                 23                 26                 22                 37                 62                 55 Incapacit.                 19                 25                 52                 27                 52                 58                 81                 95 Non‐Incap. Possible Inj.               112                 67               122                 92               103                 59                 80                 74               105                 83               155               106               173               168               227               207 No Injury Unknown Total               370                   7               601               472                   6               740           2,556               435               ‐               672               336               ‐               543               409                   2               673               531                   4               891           4,259               759                   6           1,249               858                   4           1,446 Number of Rural CMV On‐System Crashes ‐ Barnett Shale Region Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 22                 24                 33                 16                 15                 11                 13                 19 Incapacit. Non‐Incap. Possible Inj.                 49                 71                 69                 35               102                 81                 51                 97                 80                 21                 53                 43                 28                 61                 33                 28                 60                 37                 37                 55                 32                 21                 80                 38 No Injury Unknown Total               394                   2               607               501                   6               750           2,525               510                   1               772               263               ‐               396               251               ‐               388               294                   1               431           1,745               305               ‐               442               325                   1               484 Number of Rural CMV On‐System Crashes ‐ Permian Basin Region Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 23                 26                 18                 14                 20                 32                 43                 47 Incapacit.                 20                 28                 30                 25                 24                 43                 56                 90 Non‐Incap. Possible Inj.                 94                 68               110                 76                 87                 74                 86                 46                 91                 50               119                 85               133               106               183                 97 No Injury Unknown Total               291                   2               498               408                   1               649           2,189               380                   2               591               280                   1               452               366                   1               552               458                   1               738           3,384               599                   1               938               738                   1           1,156 Number of Rural CMV On‐System Crashes ‐ Remaining Counties Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               169               178               167               108               151               150               168               183 Incapacit.               278               293               250               194               273               228               223               317 Non‐Incap. Possible Inj.               652               687               740               842               745               773               613               667               654               601               577               532               620               595               688               659 26 Total No Injury Unknown           3,227                 38           5,052           3,705                 58           5,816         20,907           3,518                 26           5,479           2,967                 11           4,560           3,013                   9           4,701           3,008                 10           4,505         19,210           3,031                   9           4,646           3,504                   7           5,358 Table 11. Number of Injuries. Number of Injuries ‐ Statewide Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           3,521           3,466           3,487           3,122           3,060           3,067           3,417           3,396 Incapacit.         17,554         17,070         15,901         14,569         15,275         14,787         16,196         16,807 Non‐Incap. Possible Inj.         71,621      181,886         71,995      176,948         68,940      158,689         65,648      154,502         68,212      134,417         65,386      132,509         71,067      143,678         72,622      143,143 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               271               247               258               226               219               235               329               264 Incapacit. Non‐Incap. Possible Inj.           1,030           3,987           7,055               956           4,149           6,715               917           4,088           6,209               887           3,621           6,090               953           3,802           5,612               981           4,156           6,233           1,104           4,285           6,882           1,137           4,480           6,731 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               332               350               314               298               276               285               278               286 Incapacit. Non‐Incap. Possible Inj.           2,163           8,356         18,280           2,213           8,848         17,797           2,061           7,988         15,708           1,840           7,582         14,932           2,021           7,888         13,651           1,923           7,578         13,294           2,047           7,923         13,243           2,056           8,358         13,583 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               199               175               192               172               190               193               250               281 Incapacit. Non‐Incap. Possible Inj.               654           3,053           6,918               680           3,275           6,883               628           3,182           6,684               501           2,993           6,559               550           2,982           5,774               607           3,070           5,892               696           3,352           6,626               776           3,568           6,842 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           2,719           2,694           2,723           2,426           2,375           2,354           2,560           2,565 Incapacit.         13,707         13,221         12,295         11,341         11,751         11,276         12,349         12,838 No Injury      842,887      888,654      860,395      844,813      795,341      772,926      841,395      897,798 Unknown      135,295      139,775      127,651      123,216         49,512         48,029         53,947         60,299 Total   1,252,764   1,297,908   4,991,605   1,235,063   1,205,870   1,065,817   1,036,704   4,426,286   1,129,700   1,194,065 Number of Injuries ‐ Eagle Ford Shale Region No Injury Unknown Total         41,736           5,920         59,999         43,775           6,316         62,158      241,484         44,203           5,720         61,395         42,084           5,024         57,932         42,313           1,540         54,439         43,101           1,596         56,302      230,497         45,204           1,687         59,491         45,933           1,720         60,265 Number of Injuries ‐ Barnett Shale Region No Injury         85,891         92,987         88,461         87,049         79,987         76,437         84,158         89,964 Unknown         14,446         15,394         14,389         13,625           5,432           5,118           5,235           5,845 Total      129,468      137,589      521,304      128,921      125,326      109,255      104,635      446,866      112,884      120,092 Number of Injuries ‐ Permian Basin Region No Injury Unknown Total         39,239           5,279         55,342         42,197           5,549         58,759      225,483         39,632           5,980         56,298         38,813           6,046         55,084         36,766           1,119         47,381         35,391           1,680         46,833      202,170         39,545           2,281         52,750         41,261           2,478         55,206 Number of Injuries ‐ Remaining Counties Non‐Incap. Possible Inj.         56,225      149,633         55,723      145,553         53,682      130,088         51,452      126,921         53,540      109,380         50,582      107,090         55,507      116,927         56,216      115,987 27 No Injury      676,021      709,695      688,099      676,867      636,275      617,997      672,488      720,640 Unknown      109,650      112,516      101,562         98,521         41,421         39,635         44,744         50,256 Total   1,007,955   1,039,402   4,003,334      988,449      967,528      854,742      828,934   3,546,753      904,575      958,502 Table 12. Number of Injuries in Rural Crashes. Number of Rural Injuries ‐ Statewide Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,908           1,855           1,759           1,586           1,531           1,563           1,785           1,766 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               204               198               198               163               167               188               269               201 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               125               134               136               104                 92                 81                 98                 93 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               144               131               111               113               110               137               174               198 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,435           1,392           1,314           1,206           1,162           1,157           1,244           1,274 Incapacit.           6,585           6,214           5,710           5,293           5,318           5,311           5,676           6,190 Non‐Incap. Possible Inj.         19,028         27,065         19,010         26,758         18,359         24,251         16,739         22,343         17,378         20,065         16,460         19,984         17,489         22,634         17,988         22,677 No Injury      149,280      157,998      149,636      143,096      147,016      142,575      157,466      173,598 Unknown         24,442         25,570         22,899         20,187           6,288           5,630           6,201           7,474 Total      228,308      237,405      897,571      222,614      209,244      197,596      191,523      830,063      211,251      229,693 Number of Rural Injuries ‐ Eagle Ford Shale Region Incapacit. Non‐Incap. Possible Inj.               649           1,906           1,942               586           1,921           1,823               525           1,796           1,601               511           1,550           1,602               541           1,502           1,450               577           1,685           1,509               684           1,820           2,037               748           1,875           2,028 No Injury         10,460         10,693         10,036           8,731           9,641           9,632         11,496         12,007 Unknown           1,754           1,851           1,684           1,266               356               335               463               468 Total         16,915         17,072         63,650         15,840         13,823         13,657         13,926         61,679         16,769         17,327 Number of Rural Injuries ‐ Barnett Shale Region Incapacit.               504               499               498               382               313               362               356               351 Non‐Incap. Possible Inj.           1,252           1,400           1,294           1,387           1,319           1,277           1,038           1,028               940               854               955               835               963               832           1,005               747 No Injury           9,595         11,047         10,446           7,877           7,041           6,842           7,473           7,664 Unknown           1,782           1,984           1,806           1,124               273               220               247               224 Total         14,658         16,345         58,038         15,482         11,553           9,513           9,295         38,861           9,969         10,084 Number of Rural Injuries ‐ Permian Basin Region Incapacit.               334               299               301               223               227               292               367               484 Non‐Incap. Possible Inj.               996           1,062           1,085           1,109               983               886               871               871           1,028               647               976               798           1,154           1,094           1,485               979 No Injury           5,777           6,573           5,826           5,126           5,211           5,924           7,227           8,560 Unknown           1,097           1,305           1,143               920               154               133               171               258 Total           9,410         10,502         37,286           9,250           8,124           7,377           8,260         37,788         10,187         11,964 Number of Rural Injuries ‐ Remaining Counties Incapacit.           5,098           4,830           4,386           4,177           4,237           4,080           4,269           4,607 Non‐Incap. Possible Inj.         14,874         22,661         14,710         22,439         14,261         20,487         13,280         18,842         13,908         17,114         12,844         16,842         13,552         18,671         13,623         18,923 28 No Injury      123,448      129,685      123,328      121,362      125,123      120,177      131,270      145,367 Unknown         19,809         20,430         18,266         16,877           5,505           4,942           5,320           6,524 Total      187,325      193,486      738,597      182,042      175,744      167,049      160,042      691,735      174,326      190,318 Table 13. Number of Injuries in CMV Crashes. Number of CMV Injuries ‐ Statewide Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit. Non‐Incap. Possible Inj.               497           1,304           4,241           9,572               532           1,333           4,448           9,913               525           1,235           4,341           8,765               352               966           3,453           7,176               453           1,164           4,065           7,171               422           1,052           3,738           6,653               528           1,193           4,119           7,232               534           1,418           4,404           7,249 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 44                 37                 45                 43                 33                 46                 86                 72 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 42                 61                 54                 45                 49                 33                 35                 33 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 30                 48                 31                 64                 25                 59                 18                 42                 34                 60                 47                 81                 74               128                 69               157 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               382               404               401               246               337               296               333               360 No Injury         69,644         78,876         74,778         65,273         66,079         63,438         69,283         74,396 Unknown         41,090         37,751         25,711         21,169           1,877           1,946           1,926           2,419 Total      126,347      132,854      472,946      115,355         98,389         80,809         77,249      332,759         84,281         90,420 Number of CMV Injuries ‐ Eagle Ford Shale Region Incapacit.                 55                 62               105                 62               101               115               139               169 Non‐Incap. Possible Inj.               334               371               352               429               283               401               232               277               342               359               364               428               444               551               474               643 No Injury Unknown Total           3,884           3,729           8,417           4,578           2,663           8,121         28,518           4,279           1,598           6,711           3,424           1,231           5,269           4,087                 68           4,990           4,528                 83           5,564         24,812           5,402               107           6,729           6,049               122           7,529 Number of CMV Injuries ‐ Barnett Shale Region Incapacit.               162               156               179               150               151               123               159               136 Non‐Incap. Possible Inj.               473               989               563           1,030               534               928               404               676               481               691               403               669               387               644               462               684 No Injury           7,980         10,144           9,933           7,615           7,607           7,219           8,059           8,815 Unknown           4,948           3,550           3,550           2,540               231               244               208               304 Total         14,595         15,504         56,707         15,178         11,430           9,210           8,691         37,827           9,492         10,434 Number of CMV Injuries ‐ Permian Basin Region Non‐Incap. Possible Inj.               169               232               218               316               212               303               200               261               224               223               267               289               306               408               421               368 No Injury Unknown Total           2,274           1,556           4,309           2,947               681           4,256         16,734           2,743           1,167           4,509           2,244               895           3,660           2,462                 31           3,034           2,782                 42           3,508         16,109           3,522                 41           4,479           3,980                 93           5,088 Number of CMV Injuries ‐ Remaining Counties Incapacit. Non‐Incap. Possible Inj.           1,037           3,265           7,979           1,050           3,316           8,138               892           3,312           7,133               712           2,617           5,962               852           3,018           5,898               733           2,704           5,267               767           2,982           5,629               956           3,047           5,554 29 No Injury         55,506         61,207         57,823         51,990         51,923         48,909         52,300         55,552 Unknown         30,857         30,857         19,396         16,503           1,547           1,577           1,570           1,900 Total         99,027      104,973      370,987         88,957         78,030         63,575         59,486      254,011         63,581         67,369 Table 14. Number of Injuries in Rural CMV Crashes. Number of Rural CMV Injuries ‐ Statewide Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               306               321               319               209               269               266               351               377 Incapacit. Non‐Incap. Possible Inj.               558           1,526           2,051               566           1,746           2,458               568           1,695           2,223               426           1,311           1,716               544           1,577           1,630               534           1,605           1,595               622           1,675           2,002               763           1,957           2,084 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 38                 32                 35                 37                 36                 79                 34                 42                 28                 73                 43                 85                 77               116                 63               131 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 26                 26                 38                 20                 24                 12                 15                 20 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 23                 30                 28                 40                 22                 40                 15                 35                 28                 43                 40                 64                 56                 98                 60               132 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               218               232               223               140               189               171               203               234 No Injury         18,708         21,894         20,661         17,405         18,194         18,162         20,081         22,263 Unknown         18,017         21,540           8,959           7,074               309               302               296               465 Total         41,166         48,526      152,258         34,425         28,141         22,523         22,464         97,923         25,027         27,909 Number of Rural CMV Injuries ‐ Eagle Ford Shale Region Non‐Incap. Possible Inj.               182               201               194               270               167               186               129               130               235               162               273               177               309               314               322               367 No Injury Unknown Total           1,332           6,176           7,961           1,664           4,632           6,833         19,625           1,519               772           2,759           1,163               574           2,072           1,550                 26           2,074           1,834                 29           2,441         11,395           2,345                 52           3,213           2,719                 65           3,667 Number of Rural CMV Injuries ‐ Barnett Shale Region Incapacit.                 63                 50                 70                 34                 36                 42                 45                 36 Non‐Incap. Possible Inj.               122               118               153               142               151               144                 87                 78               101                 65               112                 74                 75                 68               109                 71 No Injury Unknown Total           1,698           1,448           3,475           2,212           3,378           5,961         14,923           2,253               965           3,621           1,164               483           1,866               929                 18           1,173           1,124                 24           1,388           5,518           1,238                 11           1,452           1,246                 23           1,505 Number of Rural CMV Injuries ‐ Permian Basin Region Non‐Incap. Possible Inj.               133               121               161               145               130               136               128                 90               138                 85               191               147               216               239               313               182 No Injury               901           1,167           1,103               819           1,033           1,415           1,785           2,241 Unknown           1,328               221               664               511                 10                 14                 16                 34 Total           2,536           1,762           7,992           2,095           1,598           1,337           1,871           8,580           2,410           2,962 Number of Rural CMV Injuries ‐ Remaining Counties Incapacit. Non‐Incap. Possible Inj.               433           1,089           1,611               439           1,238           1,901               379           1,247           1,757               315               967           1,418               392           1,103           1,318               343           1,029           1,197               363           1,075           1,381               464           1,213           1,464 30 No Injury         14,778         16,851         15,786         14,259         14,682         13,789         14,713         16,057 Unknown           9,065         13,309           6,558           5,506               255               235               217               343 Total         27,194         33,970      109,719         25,950         22,605         17,939         16,764         72,430         17,952         19,775 Table 15. Number of Injuries in Crashes on State Highways. Number of On‐System Injuries ‐ Statewide Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           2,620           2,585           2,575           2,280           2,306           2,303           2,565           2,597 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               243               215               213               191               189               215               293               234 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               246               269               230               216               212               215               203               207 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               157               130               135               139               152               161               215               235 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,974           1,971           1,997           1,734           1,753           1,712           1,854           1,921 Incapacit.         10,840         10,499           9,873           9,097           9,950           9,518         10,414         10,985 Non‐Incap. Possible Inj.         40,891         95,881         41,387         93,962         39,145         85,086         37,962         83,911         40,651         76,503         39,030         74,782         41,844         81,168         43,250         80,495 No Injury      454,513      480,608      465,722      464,696      451,060      437,621      476,549      509,153 Unknown         61,576         64,219         56,885         53,994         22,282         21,508         24,159         27,173 Total      666,321      693,260   2,670,807      659,286      651,940      602,752      584,762   2,497,866      636,699      673,653 Number of On‐System Injuries ‐ Eagle Ford Shale Region Incapacit. Non‐Incap. Possible Inj.               817           2,867           4,667               748           2,891           4,385               703           2,831           4,126               686           2,537           4,306               755           2,717           3,895               814           2,995           4,243               913           3,130           4,782               949           3,304           4,769 No Injury         26,455         27,372         28,223         27,126         28,309         29,069         31,009         32,113 Unknown           3,250           3,330           2,938           2,486               721               794               872               894 Total         38,299         38,941      153,606         39,034         37,332         36,586         38,130      157,978         40,999         42,263 Number of On‐System Injuries ‐ Barnett Shale Region Incapacit. Non‐Incap. Possible Inj. No Injury Unknown Total           1,345           4,890         10,195         49,664           6,990         73,330           1,321           5,241         10,021         53,626           7,719         78,197      297,381           1,298           4,670           8,910         50,959           6,910         72,977           1,121           4,480           8,704         51,911           6,445         72,877           1,254           4,817           8,332         49,920           2,593         67,128           1,235           4,722           7,964         46,977           2,496         63,609      272,618           1,286           4,783           8,204         51,784           2,682         68,942           1,310           5,172           8,111         55,209           2,930         72,939 Number of On‐System Injuries ‐ Permian Basin Region Incapacit. Non‐Incap. Possible Inj.               391           1,747           3,524               417           1,903           3,492               423           1,778           3,316               327           1,638           3,171               387           1,886           3,103               448           1,952           3,331               510           2,169           3,950               601           2,443           4,028 No Injury         18,977         21,037         19,153         18,804         19,533         20,047         22,660         24,495 Unknown           2,083           2,290           2,162           2,007               403               635               854               963 Total         26,879         29,269      109,201         26,967         26,086         25,464         26,574      115,161         30,358         32,765 Number of On‐System Injuries ‐ Remaining Counties Incapacit.           8,287           8,013           7,449           6,963           7,554           7,021           7,705           8,125 Non‐Incap. Possible Inj.         31,387         77,495         31,352         76,064         29,866         68,734         29,307         67,730         31,231         61,173         29,361         59,244         31,762         64,232         32,331         63,587 31 No Injury      359,417      378,573      367,387      366,855      353,298      341,528      371,096      397,336 Unknown         49,253         50,880         44,875         43,056         18,565         17,583         19,751         22,386 Total      527,813      546,853   2,110,619      520,308      515,645      473,574      456,449   1,952,109      496,400      525,686 Table 16. Number of Injuries in Crashes on Rural State Highways. Number of Rural On‐System Injuries ‐ Statewide Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,573           1,546           1,451           1,293           1,256           1,279           1,483           1,477 Incapacit.           4,963           4,705           4,281           3,868           4,059           3,899           4,404           4,873 Non‐Incap. Possible Inj.         13,674         18,113         13,912         18,105         13,037         16,246         11,931         14,802         12,362         13,286         11,742         13,009         12,479         14,929         12,972         14,714 No Injury         97,888      103,908         96,403         91,030         91,464         89,583         98,449      107,378 Unknown         15,728         17,023         14,821         12,474           2,905           2,681           2,964           3,519 Total      151,939      159,199      592,775      146,239      135,398      125,332      122,193      527,166      134,708      144,933 Number of Rural On‐System Injuries ‐ Eagle Ford Shale Region Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               187               185               173               148               148               176               246               183 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal                 95               111               114                 84                 79                 69                 81                 72 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal               120               110                 88               102                 92               124               160               177 Year 2006 2007 2008 2009 2010 2011 2012 2013 Fatal           1,171           1,140           1,076               959               937               910               996           1,045 Incapacit. Non‐Incap. Possible Inj.               568           1,609           1,620               510           1,601           1,533               453           1,464           1,293               433           1,279           1,371               479           1,278           1,236               516           1,464           1,332               605           1,591           1,781               676           1,642           1,782 No Injury           8,439           8,593           8,257           7,243           8,028           8,341           9,830         10,405 Unknown           1,479           1,541           1,352           1,038               214               240               342               349 Total         13,902         13,963         52,369         12,992         11,512         11,383         12,069         52,884         14,395         15,037 Number of Rural On‐System Injuries ‐ Barnett Shale Region Incapacit.               380               373               379               262               224               257               301               286 Non‐Incap. Possible Inj.               885           1,004               951               968               940               904               754               731               666               588               708               597               658               616               757               555 No Injury           6,791           7,697           7,074           5,366           4,951           4,759           5,354           5,541 Unknown           1,267           1,485           1,272               735               112               103               127               112 Total         10,422         11,585         40,622         10,683           7,932           6,620           6,493         27,573           7,137           7,323 Number of Rural On‐System Injuries ‐ Permian Basin Region Incapacit.               251               228               257               184               199               260               328               429 Non‐Incap. Possible Inj.               838               839               881               886               778               701               697               690               815               517               810               629               958               925           1,269               836 No Injury           4,549           5,168           4,375           4,044           4,021           4,801           5,999           7,141 Unknown               874           1,084               906               739                 83                 79               113               179 Total           7,471           8,357         29,389           7,105           6,456           5,727           6,703         30,944           8,483         10,031 Number of Rural On‐System Injuries ‐ Remaining Counties Incapacit. Non‐Incap. Possible Inj.           3,764         10,342         14,650           3,594         10,479         14,718           3,192           9,855         13,348           2,989           9,201         12,010           3,157           9,603         10,945           2,866           8,760         10,451           3,170           9,272         11,607           3,482           9,304         11,541 32 No Injury         78,109         82,450         76,697         74,377         74,464         71,682         77,266         84,291 Unknown         12,108         12,913         11,291           9,962           2,496           2,259           2,382           2,879 Total      120,144      125,294      470,395      115,459      109,498      101,602         96,928      415,765      104,693      112,542 Table 17. Number of Injuries in CMV Crashes on State Highways. Number of CMV On‐System Injuries ‐ Statewide Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit. Non‐Incap. Possible Inj.               459           1,087           3,223           6,166               482           1,113           3,445           6,547               472           1,018           3,357           5,956               327               773           2,678           4,961               405               990           3,261           5,113               396               898           2,997           4,709               479           1,016           3,381           5,375               499           1,230           3,703           5,563 No Injury         43,337         49,377         47,722         41,495         42,887         41,937         46,080         49,946 Unknown         40,369         37,172         19,342         16,255           1,312           1,329           1,438           1,740 Total         94,641         98,136      337,132         77,867         66,489         53,968         52,266      226,684         57,769         62,681 Number of CMV On‐System Injuries ‐ Eagle Ford Shale Region Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 44                 38                 41                 42                 32                 46                 84                 70 Incapacit.                 46                 57               100                 56                 94               104               134               161 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 38                 53                 49                 41                 43                 32                 30                 31 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 31                 41                 31                 56                 23                 56                 17                 36                 32                 55                 46                 74                 71               118                 64               146 Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               345               359               359               227               298               272               294               334 Non‐Incap. Possible Inj.               304               283               328               310               259               281               191               249               294               307               335               378               414               500               439               544 No Injury           2,852           3,322           3,109           2,410           3,000           3,524           4,195           4,855 Unknown         11,704         10,472           1,232               996                 56                 55                 84                 97 Total         15,233         14,527         38,725           5,022           3,944           3,783           4,442         19,802           5,411           6,166 Number of CMV On‐System Injuries ‐ Barnett Shale Region Incapacit.               141               128               144               113               121               101               137               112 Non‐Incap. Possible Inj.               359               696               449               687               410               657               295               508               371               560               335               481               303               510               384               555 No Injury           4,947           6,208           6,349           5,178           5,356           4,934           5,459           6,255 Unknown           4,037           2,307           2,595           1,945               157               163               171               203 Total         10,217           9,832         38,333         10,204           8,080           6,608           6,046         26,804           6,610           7,540 Number of CMV On‐System Injuries ‐ Permian Basin Region Non‐Incap. Possible Inj.               147               193               186               259               172               236               162               172               188               175               235               224               263               347               375               291 No Injury           1,585           2,132           1,974           1,515           1,862           2,164           2,792           3,216 Unknown           1,409               157               939               731                 16                 28                 32                 72 Total           3,406           2,821         12,259           3,400           2,633           2,328           2,771         12,886           3,623           4,164 Number of CMV On‐System Injuries ‐ Remaining Counties Incapacit. Non‐Incap. Possible Inj.               859           2,413           4,995               872           2,482           5,291               718           2,516           4,782               568           2,030           4,032               720           2,408           4,071               619           2,092           3,626               627           2,401           4,018               811           2,505           4,173 33 No Injury         33,953         37,715         36,290         32,392         32,669         31,315         33,634         35,620 Unknown         23,220         24,237         14,576         12,583           1,083           1,083           1,151           1,368 Total         65,785         70,956      247,815         59,241         51,832         41,249         39,007      167,192         42,125         44,811 Table 18. Number of Injuries in CMV Crashes on Rural State Highways. Number of Rural CMV On‐System Injuries ‐ Statewide Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               299               312               298               202               252               261               337               362 Incapacit. Non‐Incap. Possible Inj.               501           1,353           1,612               523           1,559           1,983               527           1,499           1,777               383           1,179           1,373               500           1,417           1,368               489           1,416           1,294               570           1,517           1,675               723           1,788           1,776 No Injury         13,614         16,121         15,358         12,579         13,099         13,799         14,784         16,705 Unknown         12,694         19,620           7,772           6,060               246               244               238               388 Total         30,073         40,118      119,198         27,231         21,776         16,882         17,503         75,248         19,121         21,742 Number of Rural CMV On‐System Injuries ‐ Eagle Ford Shale Region Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 39                 28                 35                 36                 34                 77                 34                 40                 28                 70                 43                 76                 75               112                 62               131 Non‐Incap. Possible Inj.               179               151               193               208               164               134               119               126               218               157               263               172               295               308               317               353 No Injury Unknown Total           1,156           2,132           3,684           1,475           1,827           3,776         11,854           1,359               707           2,475           1,053               547           1,919           1,355                 26           1,854           1,714                 24           2,292         10,472           2,057                 45           2,892           2,510                 61           3,434 Number of Rural CMV On‐System Injuries ‐ Barnett Shale Region Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal                 24                 27                 36                 19                 20                 12                 14                 19 Incapacit.                 61                 44                 64                 28                 36                 36                 45                 35 Non‐Incap. Possible Inj.               100               101               143               120               136               118                 78                 55                 85                 59               106                 55                 67                 60               100                 56 No Injury Unknown Total           1,266           1,616           3,169           1,610           4,848           6,792         14,214           1,639               808           2,801               877               395           1,452               722                 10               932               816                 20           1,045           4,326               963                 10           1,159               959                 21           1,190 Number of Rural CMV On‐System Injuries ‐ Permian Basin Region Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal Incapacit.                 25                 26                 29                 36                 20                 39                 15                 33                 27                 42                 39                 59                 54                 91                 55               124 Non‐Incap. Possible Inj.               123               113               143               126               113               122               119                 82               130                 77               170               133               192               223               287               167 No Injury               720           1,010               942               749               902           1,284           1,635           2,098 Unknown               609               305               609               473                   9                 12                 16                 32 Total           1,615           1,649           6,580           1,845           1,471           1,187           1,697           7,858           2,211           2,763 Number of Rural CMV On‐System Injuries ‐ Remaining Counties Year 2006* 2007* 2008 2009 2010 2011 2012 2013 Fatal               211               221               208               134               177               167               194               226 Incapacit.               386               406               347               282               352               318               322               433 Non‐Incap. Possible Inj.               951           1,247           1,079           1,529           1,086           1,403               863           1,110               984           1,075               877               934               963           1,084           1,084           1,200 34 No Injury         10,473         12,026         11,418           9,900         10,120           9,985         10,129         11,138 Unknown           8,338         12,641           5,648           4,645               201               188               167               274 Total         21,605         27,902         86,551         20,110         16,934         12,909         12,469         52,592         12,859         14,355 General Trends With the data compiled in Table 3 through Table 18, as well as basic crash attribute data such as severity, number and type of vehicles involved, manner of collision, contributing factors and type of roadway, the research team prepared a series of figures and charts to develop a high-level understanding of crash locations and trends. This section includes a small sample of charts and figures to illustrate general highlights. Additional materials could be prepared as needed. Figure 4 shows the annual relative variation in the number of crashes using year 2006 as the reference point. In the figure, an index value of 1.0 means that the number of crashes for any given year was the same as the number of crashes in 2006. The figure shows aggregated changes for the Barnet Shale region (13 counties), Eagle Ford Shale region (29 counties), and Permian Basin region (37 counties), as well as all other 175 counties in the state. Figure 4 shows values for all crashes; fatal, incapacitating injuries, and non-incapacitating (KAB) injury crashes; and fatal crashes. Overall, the figure shows that both crashes in energy sector counties and nonenergy sector counties decreased from 2006 through 2009, then increased from 2010 through 2013. Crashes in the Eagle Ford Shale and Permian Basin regions increased faster than in other regions in the state. The relative growth in crashes in those two regions increased faster as the severity of the crashes increased. The decrease in the number of fatal crashes in the Eagle Ford Shale region in 2013 was an interesting observation, but the research team could not identify a clear reason as to why this happened. It is possible that enhanced safety campaigns by the Department of Public Safety in the region had an impact. However, more research would be necessary to isolate and quantify this effect. In any case, even with this decrease, the number of fatal crashes in the Eagle Ford Shale region in 2013 was slightly higher than in 2006. Figure 5 shows a similar figure for rural crashes. As in the previous figure, Figure 5 shows all rural crashes, rural KAB crashes, and rural fatal crashes. The trends are somewhat similar as in Figure 4, but the differences among regions are more clearly defined. In particular, the number of KAB crashes and fatal crashes in rural areas in the Eagle Ford Shale and Permian Basin regions began to increase in 2010 and 2011, and by 2012 they were higher than in 2006. As described in Chapter 2, there was a very strong correlation between the change in the number of new wells and rural CMV crashes. Figure 6 shows all rural CMV crashes, rural KAB CMV crashes, and rural fatal CMV crashes. In this case, the trends in Figure 5 were even more clearly defined. In particular, not only did rural CMV crashes in the Eagle Ford Shale and Permian Basin regions begin to increase in 2010, they were already higher that year than in 2006. Notice also that, despite the decrease in rural fatal CMV crashes in the Eagle Ford Shale region in 2013, the number of rural CMV crashes in that region was still more than twice the number in 2006. 35 All crashes Crash Index ‐ All Crashes 1.2 1.0 0.8 0.6 0.4 0.2 0.0 2006 2007 Eagle Ford Shale 2008 2009 Barnett Shale 2010 2011 Permian Basin 2012 2013 Other Counties All KAB crashes Crash Index ‐ All KAB Crashes 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 2006 2007 Eagle Ford Shale 2008 2009 Barnett Shale 2010 2011 Permian Basin 2012 2013 Other Counties All fatal crashes Crash Index ‐ All Fatal Crashes 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 2006 2007 Eagle Ford Shale 2008 2009 Barnett Shale Figure 4. Crash Index – All Crashes. 36 2010 Permian Basin 2011 2012 Other Counties 2013 Rural crashes Crash Index ‐ Rural Crashes 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 2006 2007 Eagle Ford Shale 2008 2009 Barnett Shale 2010 2011 Permian Basin 2012 2013 Other Counties Rural KAB crashes Crash Index ‐ Rural KAB Crashes 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 2006 2007 Eagle Ford Shale 2008 2009 Barnett Shale 2010 2011 Permian Basin 2012 2013 Other Counties Rural fatal crashes Crash Index ‐ Rural Fatal Crashes 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 2006 2007 Eagle Ford Shale 2008 2009 Barnett Shale 2010 Permian Basin Figure 5. Crash Index – Rural Crashes. 37 2011 2012 Other Counties 2013 Rural CMV crashes Crash Index ‐ Rural CMV Crashes 2.5 2.0 1.5 1.0 0.5 0.0 2006 2007 2008 2009 Eagle Ford Shale Barnett Shale 2006 2008 2010 2011 Permian Basin 2012 2013 Other Counties Rural KAB CMV crashes Crash Index ‐ Rural KAB CMV Crashes 2.5 2.0 1.5 1.0 0.5 Rural fatal CMV crashes Crash Index ‐ Rural Fatal CMV Crashes 0.0 2007 2009 Eagle Ford Shale Barnett Shale 2006 2008 2010 2011 Permian Basin 2012 2013 Other Counties 3.0 2.5 2.0 1.5 1.0 0.5 0.0 2007 Eagle Ford Shale 2009 Barnett Shale 2010 Permian Basin Figure 6. Crash Index – Rural CMV Crashes. 38 2011 2012 Other Counties 2013 CRIS contains data regarding manners of collision and crash contributing factors for individual crashes, according to information included in the crash reports. This information is based on police officers’ interpretation of what happened at the crash sites, not based on forensic or engineering analyses. Despite these caveats, a high-level review of the data offers insight as to potential crash causes. Figure 7 shows the top 10 manners of collision in CRIS. There were some differences among regions, but for the most part, the trends were similar throughout the state. By far, the most common manner of collision was a single motor vehicle going straight. The second and third most common manners of collision were two vehicles going straight either in opposite or same directions. Readers should note that a vehicle going straight does not mean that the road alignment was straight. It just means the vehicle was not making a turn. A more detailed analysis would be necessary to clarify the impact of roadway characteristics, such as horizontal and vertical alignments, curvature, cross section characteristics, visibility restrictions, and pavement conditions. 70% 60% 50% 40% 30% 20% 10% 0% Barnett Shale Eagle Ford Shale Permian Basin Other 1. OMV VEHICLE GOING STRAIGHT 2. SD BOTH GOING STRAIGHT‐REAR END 3. OD BOTH GOING STRAIGHT 4. ANGLE ‐ BOTH GOING STRAIGHT 5. SD BOTH GOING STRAIGHT‐SIDESWIPE 6. SD ONE STRAIGHT‐ONE STOPPED 7. SD ONE STRAIGHT‐ONE LEFT TURN 8. OD ONE STRAIGHT‐ONE LEFT TURN 9. ANGLE ‐ ONE STRAIGHT‐ONE LEFT TURN 10. SD ONE STRAIGHT‐ONE RIGHT TURN Figure 7. Top 10 Manners of Collision (2010–2013 Crash Data). Figure 8 shows the top 10 vehicle contributing factors in CRIS. There were some differences among regions, but for the most part, the trends were similar throughout the state. By far, the most common vehicle contributing factor category was “not applicable,” which is an indication 39 of the police officer’s inability to identify what vehicle factor may have contributed to the crash based on information that was readily available at the crash site. In many cases, it is possible that the interaction between the roadway environment and the vehicle was a contribution factor to a crash. However, this information is normally not included in crash reports. Most other vehicle contributing factors in Figure 8 are factors that could be considered driver factors (i.e., factors that might suggest lack of control by the driver as a contributing factor to the crash). 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Barnett Shale Eagle Ford Shale Permian Basin Other 1. NOT APPLICABLE 2. UNSAFE SPEED 3. FAILED TO CONTROL SPEED 4. DRIVER INATTENTION 5. FAULTY EVASIVE ACTION 6. FATIGUED OR ASLEEP 7. FAILED TO DRIVE IN SINGLE LANE 8. OTHER (EXPLAIN IN NARRATIVE) 9. FAILED TO YIELD ROW ‐ STOP SIGN 10. CHANGED LANE WHEN UNSAFE Figure 8. Top 10 Vehicle Contributing Factors (2010–2013 Crash Data). 40 Chapter 2. Analysis and Trends Oil and Gas Well Developments The March 2015 report provides information on the methodology to extract and analyze oil and gas energy developments in the state (2). In summary, the research team aggregated and compared data using two four-year blocks: 2006–2009 and 2010–2013. Table 19 summarizes changes in the number of new completed wells in the Barnett Shale, Eagle Ford Shale, and Permian Basin regions from 2006–2009 to 2010–2013. The table also shows the total number of wells completed in other areas and throughout the state. The total number of new wells increased from 51,393 during the four-year period from 2006– 2009 to 55,398 from 2010–2013 (i.e., the total number of wells drilled increased by 8 percent). This growth was not uniform. For example, in the Barnett Shale region, the number of new horizontal wells decreased by 48 percent and the number of new vertical wells decreased by 53 percent. In the Eagle Ford Shale region, the number of new horizontal wells increased by 941 percent but the number of new vertical wells decreased by 20 percent. In the Permian Basin region, the number of new horizontal wells increased by 240 percent and the number of new vertical wells increased by 49 percent. Table 19. Changes in the Number of New Completed Wells. Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total Karnes County Number of Horizontal Wells 2006‐09 2010‐13 Diff.       8,663       4,490 ‐48%           854       8,886 941%           951       3,230 240%       1,761       3,356 91%     12,229     19,962 63%             28       1,312 4586% Number of Vertical Wells 2006‐09 2010‐13 Diff.       1,482           698 ‐53%       4,595       3,689 ‐20%     14,381     21,396 49%     18,706       9,653 ‐48%     39,164     35,436 ‐10%             38             50 32% Total Number of Wells 2006‐09 2010‐13 Diff.  10,145     5,188 ‐49%     5,449  12,575 131%  15,332  24,626 61%  20,467  13,009 ‐36%  51,393  55,398 8%           66     1,362 1964% Changes in the Number of Crashes and Injuries With the data compiled in Table 3 through Table 18, the research team examined changes in the number of crashes and the number of injuries from 2006–2009 to 2010–2013. Table 20 through Table 23 summarize the results, as follows:  Table 20 shows changes in the number of crashes on all highways.  Table 21 shows changes in the number of crashes on state highways.  Table 22 shows changes in the number of injuries on all highways.  Table 23 shows changes in the number of injuries on state highways. 41 Table 20. Changes in the Number of Crashes on All Highways. Note: Green dots correspond to decreases in the number of crashes (desirable trend). Red dots correspond to increases in the number of crashes (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09        184,735          85,964          80,891     1,410,907    1,762,497 Number of Crashes (Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown) All Rural CMV‐Involved Rural & CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.        166,474 ‐10%       24,572       18,521 ‐25%     14,119       12,367 ‐12%      3,130      2,061 ‐34%          86,744 1%       27,660       28,804 4%       6,607         8,708 32%      2,820      4,542 61%          77,511 ‐4%       15,689       17,426 11%       4,775         6,368 33%      2,464      3,743 52%     1,306,749 ‐7%    288,715    284,431 ‐1%     90,081       77,755 ‐14%    26,221    23,942 ‐9%    1,637,478 ‐7%    356,636    349,182 ‐2%  115,582    105,198 ‐9%   34,635   34,288 ‐1% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09          31,739          14,382          11,520        204,134       261,775 All 2010‐13          30,728          15,264          12,019        201,541       259,552 Number of Fatal, Incapacitating, Non‐Incapacitating Crashes Rural CMV‐Involved Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. ‐3%         5,346         4,165 ‐22%       2,124         1,846 ‐13% 6%         6,889         6,948 1%       1,096         1,641 50% 4%         3,841         4,524 18%          883         1,333 51% ‐1%       57,296       54,123 ‐6%     12,568       11,792 ‐6% ‐1%      73,372      69,760 ‐5%    16,671      16,612 0% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.          642          474 ‐26%          662      1,173 77%          617          971 57%      4,998      4,751 ‐5%      6,919      7,369 7% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09            1,202                851                648            9,465         12,166 All 2010‐13            1,030                902                789            8,954         11,675 Diff. ‐14% 6% 22% ‐5% ‐4% 2006‐09            459            629            430         4,673        6,191 Number of Fatal Crashes Rural CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff.            325 ‐29%          181            135 ‐25%            694 10%          129            204 58%            518 20%            94            183 94%         4,293 ‐8%       1,177         1,170 ‐1%        5,830 ‐6%      1,582        1,692 7% 42 Rural & CMV‐Involved 2006‐09 2010‐13 Diff.          101            63 ‐37%          102          179 76%            80          151 88%          663          684 3%         946      1,077 14% Table 21. Changes in the Number of Crashes on State Highways. Note: Green dots correspond to decreases in the number of crashes (desirable trend). Red dots correspond to increases in the number of crashes (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total Number of Crashes (Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown) All Rural CMV‐Involved Rural & CMV‐Involved 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.   101,789     54,020     38,703   726,867  921,379     97,528     58,201     43,040   698,902  897,671 ‐4% 8% 11% ‐4% ‐3%     16,179     21,974     12,046   181,734  231,933     12,523     23,975     14,114   171,845  222,457 ‐23% 9% 17% ‐5% ‐4%   10,010     4,979     3,645   60,499  79,132     9,278     7,186     5,189   55,617  77,270 ‐7% 44% 42% ‐8% ‐2%        2,525        2,556        2,189     20,907     28,178        1,745        4,259        3,384     19,210     28,598 ‐31% 67% 55% ‐8% 1% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09     18,636     10,122        6,485   114,236  149,479 All 2010‐13     18,766     11,208        7,799   116,161  153,934 Number of Fatal, Incapacitating, Non‐Incapacitating Crashes Rural CMV‐Involved Rural & CMV‐Involved Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 1%        3,798        2,998 ‐21%     1,675     1,503 ‐10%           575           428 ‐26% 11%        5,684        5,986 5%         974     1,503 54%           639        1,122 76% 20%        3,059        3,775 23%         752     1,166 55%           561           881 57% 2%     39,809     37,182 ‐7%     9,645     9,522 ‐1%        4,387        4,232 ‐4% 3%     52,350     49,941 ‐5%  13,046  13,694 5%       6,162       6,663 8% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09           887           722           482        6,775       8,866 All 2010‐13           761           792           647        6,492       8,692 Diff. ‐14% 10% 34% ‐4% ‐2% 2006‐09           369           564           355        3,746       5,034 Number of Fatal Crashes Rural CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff.           268 ‐27%         167         122 ‐27%           627 11%         123         199 62%           454 28%           93         172 84%        3,399 ‐9%     1,046     1,046 0%       4,748 ‐6%     1,429     1,539 8% 43 Rural & CMV‐Involved 2006‐09 2010‐13 Diff.             96             58 ‐39%             99           176 77%             80           142 77%           622           652 5%          897       1,028 15% Table 22. Changes in the Number of Injuries on All Highways. Note: Green dots correspond to decreases in the number of injuries (desirable trend). Red dots correspond to increases in the number of injuries (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09       521,304       241,484       225,483   4,003,334  4,991,605 Number of Injuries (Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown) All Rural CMV‐Involved Rural & CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.       446,866 ‐14%         58,038     38,861 ‐33%     56,707     37,827 ‐33%     14,923        5,518 ‐63%       230,497 ‐5%         63,650     61,679 ‐3%     28,518     24,812 ‐13%     19,625     11,395 ‐42%       202,170 ‐10%         37,286     37,788 1%     16,734     16,109 ‐4%        7,992        8,580 7%   3,546,753 ‐11%       738,597   691,735 ‐6%   370,987   254,011 ‐32%   109,719     72,430 ‐34%  4,426,286 ‐11%      897,571  830,063 ‐8%  472,946  332,759 ‐30%  152,258     97,923 ‐36% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09         42,345         20,637         15,704       278,208      356,894 All 2010‐13         40,919         21,945         16,515       273,913      353,292 Number of Fatal, Incapacitating, Non‐Incapacitating Injuries Rural CMV‐Involved Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. ‐3%           7,285        5,609 ‐23%        2,823        2,452 ‐13% 6%         10,207     10,257 0%        1,654        2,385 44% 5%           5,591        6,632 19%        1,115        1,868 67% ‐2%         80,963     75,957 ‐6%     17,635     16,385 ‐7% ‐1%      104,046     98,455 ‐5%     23,228     23,090 ‐1% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.           840           627 ‐25%        1,006        1,755 75%           786        1,379 76%        6,920        6,779 ‐2%       9,552     10,540 10% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09           1,294           1,002              738         10,562        13,596 All 2010‐13           1,125           1,047              914           9,854        12,940 Diff. ‐13% 4% 24% ‐7% ‐5% 2006‐09              499              763              499           5,347          7,108 Number of Fatal Injuries Rural CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff.           364 ‐27%           201           150 ‐26%           825 8%           169           237 40%           619 24%           103           224 117%        4,837 ‐10%        1,433        1,326 ‐7%       6,645 ‐7%       1,907       1,937 2% 44 Rural & CMV‐Involved 2006‐09 2010‐13 Diff.           111             71 ‐36%           143           211 47%             88           184 109%           813           797 ‐2%       1,155       1,263 9% Table 23. Changes in the Number of Injuries on State Highways. Note: Green dots correspond to decreases in the number of injuries (desirable trend). Red dots correspond to increases in the number of injuries (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09       297,381       153,606       109,201   2,110,619  2,670,807 Number of Injuries (Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown) All Rural CMV‐Involved Rural & CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.       272,618 ‐8%     40,622     27,573 ‐32%     38,333     26,804 ‐30%         14,214           4,326 ‐70%       157,978 3%     52,369     52,884 1%     38,725     19,802 ‐49%         11,854         10,472 ‐12%       115,161 5%     29,389     30,944 5%     12,259     12,886 5%           6,580           7,858 19%   1,952,109 ‐8%   470,395   415,765 ‐12%   247,815   167,192 ‐33%         86,551         52,592 ‐39%  2,497,866 ‐6%  592,775  527,166 ‐11%  337,132  226,684 ‐33%      119,198        75,248 ‐37% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09         25,327         14,942           9,185       160,300      209,754 All 2010‐13         25,416         16,508         11,159       162,330      215,413 Diff. 0% 10% 21% 1% 3% Number of Fatal, Incapacitating, Non‐Incapacitating Injuries Rural CMV‐Involved 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.        5,328        4,158 ‐22%        2,220        2,000 ‐10%        8,610        9,004 5%        1,506        2,207 47%        4,534        5,621 24%           959        1,667 74%     57,762     53,502 ‐7%     13,748     13,381 ‐3%     76,234     72,285 ‐5%     18,434     19,255 4% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.              761              575 ‐24%              979           1,690 73%              720           1,270 76%           6,175           6,097 ‐1%          8,635          9,632 12% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09              961              862              561           7,676        10,060 All 2010‐13              837              931              763           7,240          9,771 Diff. ‐13% 8% 36% ‐6% ‐3% 2006‐09           404           693           420        4,346       5,863 Number of Fatal Injuries Rural CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff.           301 ‐25%           181           136 ‐25%           753 9%           165           232 41%           553 32%           103           213 107%        3,888 ‐11%        1,290        1,198 ‐7%       5,495 ‐6%       1,739       1,779 2% 45 Rural & CMV‐Involved 2006‐09 2010‐13 Diff.              106                 65 ‐39%              142              208 46%                 89              175 97%              774              764 ‐1%          1,111          1,212 9% Table 20 shows changes in the number of crashes on all highways from 2006–2009 to 2010– 2013. In total, the number of crashes decreased by 10 percent in the Barnett Shale region, increased by 1 percent in the Eagle Ford Shale region, and decreased by 4 percent in the Permian Basin region. As a reference, the number of crashes decreased by 7 percent in all other 175 counties in the state. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. For example:  Changes were more prominent for rural crashes. In the Barnett Shale region, the number of rural crashes decreased by 25 percent (compared to a 10 percent decrease overall in the region). In the Eagle Ford Shale region, the number of rural crashes increased by 4 percent (compared to a 1 percent increase overall in the region). In the Permian Basin region, the number of crashes increased by 11 percent (compared to a 4 percent decrease overall in the region). The changes were even more noticeable for crashes that involved CMVs and, in particular, for rural crashes that involved CMVs. Overall, for rural crashes that involved CMVs, there was a 34 percent decrease in the Barnett Shale region, a 61 percent increase in the Eagle Ford Shale region, and a 52 percent increase in the Permian Basin region. By comparison, there was a 9 percent decrease in all other 175 counties in the state.  In most cases, as the severity of the injuries worsened, the changes in the corresponding number of crashes were more evident. For example, for rural crashes that involved CMVs in the Barnett Shale region, there was a 26 percent decrease in the number of fatal, incapacitating, and non-incapacitating injury crashes (compared to a 34 percent decrease for all crashes). For fatal crashes, the decrease was 37 percent. In the Eagle Ford Shale region, there was a 77 percent increase in the number of fatal, incapacitating, and nonincapacitating injury crashes (compared to a 61 percent increase for all crashes). For fatal crashes, the increase was 76 percent. In the Permian Basin region, there was a 57 percent increase in the number of fatal, incapacitating, and non-incapacitating injury crashes (compared to a 52 percent increase for all crashes). For fatal crashes, the increase was 88 percent. Table 21 shows changes in the number of crashes on state highways from 2006–2009 to 2010– 2013. In total, the number of crashes decreased by 4 percent in the Barnett Shale region, increased by 8 percent in the Eagle Ford Shale region, and increased by 11 percent in the Permian Basin region. As a reference, the number of crashes decreased by 3 percent in all other 175 counties in the state. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. The relative changes in the number of crashes on state highways (Table 21) were similar to those found for all highways (Table 20). 46 From 2006–2009 to 2010–2013, the percentage of crashes occurring on state highways increased by about 2 percent. For all crashes, this percentage increased from 52.2 to 54.8 percent. For fatal, incapacitating, and non-incapacitating injury crashes, it increased from 57.1 to 59.3 percent. For fatal crashes, it increased from 72.8 to 74.4 percent. These percentages were higher for rural roads. For example, for CMV crashes on rural roadways, the percentage of crashes on state highways increased from 81.4 to 83.4 percent. For fatal, incapacitating, and non-incapacitating injury crashes, this percentage increased from 89.1 to 90.4 percent. For fatal crashes, it increased from 94.8 to 95.4 percent. Table 22 shows changes in the number of injuries in crashes on all highways from 2006–2009 to 2010–2013. In total, the number of injuries decreased by 14 percent in the Barnett Shale region, decreased by 5 percent in the Eagle Ford Shale region, and decreased by 10 percent in the Permian Basin region. As a reference, the number of injuries decreased by 11 percent in all other 175 counties in the state. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. For example:  The changes were more prominent in the case of rural crashes. In the Barnett Shale region, the number of injuries in rural crashes decreased by 33 percent (compared to a 14 percent decrease overall in the region). In the Eagle Ford Shale region, the number of injuries in rural crashes decreased by 3 percent (compared to a 5 percent decrease overall in the region). In the Permian Basin region, the number of injuries in crashes increased by 1 percent (compared to a 10 percent decrease overall in the region). The changes were even more noticeable for crashes that involved CMVs and, particularly, for rural crashes that involved CMVs. Overall, for rural crashes that involved CMVs, there was a 63 percent decrease in the Barnett Shale region, a 41 percent decrease in the Eagle Ford Shale region, and a 7 percent increase in the Permian Basin region. By comparison, there was a 34 percent decrease in all other 175 counties in the state.  In most cases, as the severity of the injuries worsened, the changes in the corresponding number of crashes were more evident. For example, for rural crashes that involved CMVs in the Barnett Shale region, there was a 25 percent decrease in the number of injuries resulting from fatal, incapacitating, and non-incapacitating injury crashes (compared to a 63 percent decrease for all crashes). For fatal crashes, the decrease was 36 percent. In the Eagle Ford Shale region, there was a 75 percent increase in the number of injuries resulting from fatal, incapacitating, and non-incapacitating injury crashes (compared to a 42 percent decrease for all crashes). For fatal crashes, the increase was 47 percent. In the Permian Basin region, there was a 76 percent increase in the number of fatal, 47 incapacitating, and non-incapacitating injury crashes (compared to a 7 percent increase for all crashes). For fatal crashes, the increase was 109 percent. Table 23 shows changes in the number of injuries from crashes on state highways from 2006– 2009 to 2010–2013. In total, the number of crashes decreased by 8 percent in the Barnett Shale region, increased by 3 percent in the Eagle Ford Shale region, and increased by 5 percent in the Permian Basin region. As a reference, the number of crashes decreased by 8 percent in all other 175 counties in the state. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. The relative changes in the number of injuries from crashes on state highways (Table 23) were similar to those found for all highways (Table 22). From 2006–2009 to 2010–2013, the percentage of injuries from crashes occurring on state highways increased by about 2 percent. For all crashes, this percentage increased from 53.5 to 56.4 percent. For fatal, incapacitating, and non-incapacitating injury crashes, it increased from 58.8 to 61.0 percent. For fatal crashes, it increased from 74.0 to 75.5 percent. These percentages were higher for rural roads. For example, for rural CMV crashes, the percentage of injuries from crashes on state highways decreased from 78.3 to 76.8 percent. For fatal, incapacitating, and non-incapacitating injury crashes, this percentage increased from 90.4 to 91.4 percent. For fatal crashes, it decreased from 96.2 to 96.0 percent. Changes in Crash Rates As part of the 0-6498 research project completed in 2011, the researchers conducted an evaluation of crash rates in the north and northwest regions of the state, with a focus on areas within the jurisdiction of the TxDOT Abilene, Lubbock, and Fort Worth Districts (1). For the analysis, the researchers calculated crash rates by dividing the number of crashes found by the number of VMT based on traffic information in the TxDOT Pavement Management Information System (PMIS). Because of concerns about the reliability of average annual daily traffic (AADT) data in the PMIS database (as well as the low frequency of crashes and relatively low traffic volumes that characterize most rural highways), the researchers also used distance (more specifically 100 miles) to estimate crash rates. Although this approach did not explicitly consider exposure, at least it eliminated the issue of traffic volume uncertainty. Overall, the 0-6498 analysis indicated higher crash rates along corridors where energy developments took place than crash rates along control corridors. The trends also indicated higher crash rates along corridors with higher traffic volumes. In the Lubbock District, crash rates along energy development corridors were similar to those along control corridors. However, traffic volumes were low in general. In the Abilene District (which exhibited higher traffic volumes than those in the Lubbock District), crash rates along energy development corridors were higher than (a) crash rates along control corridors in the same district and (b) crash rates along energy development corridors in the Lubbock District. In the Fort Worth District (which exhibited higher traffic volumes than those in the Abilene District), crash rates 48 along energy development corridors were higher than (a) crash rates along control corridors in the same district and (b) crash rates along energy development corridors in the Abilene District. For the 2006–2013 crash data compiled as described in the previous chapter, the research team calculated crash rates expressed both as the number of crashes per 100 million VMTs and number of crashes per 100 lane-miles. Table 24 shows the changes in the number of crashes over four years per 100 million VMTs from 2006–2009 to 2010–2013. Table 25 shows the changes in the number of crashes over four years per 100 lane-miles from 2006–2009 to 2010–2013. Table 24 shows changes in the number of crashes over four years per 100 million VMTs from 2006–2009 to 2010–2013. In total, the crash rate decreased by 2 percent in the Barnett Shale region, increased by 4 percent in the Eagle Ford Shale region, and increased by 7 percent in the Permian Basin region. As a reference, the crash rate decreased by 2 percent in all other 175 counties in the state. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. For example:  The changes were more prominent for rural crashes. In the Barnett Shale region, the rural crash rate decreased by 18 percent (compared to a 2 percent decrease overall in the region). In the Eagle Ford Shale region, the rural crash rate increased by 3 percent (compared to a 4 percent increase overall in the region). In the Permian Basin region, the rural crash rate increased by 13 percent (compared to a 7 percent increase overall in the region). The changes were even more noticeable for crashes that involved CMVs and, particularly, for rural crashes that involved CMVs. For rural crashes that involved CMVs, there was a 26 percent decrease in the Barnett Shale region, a 57 percent increase in the Eagle Ford Shale region, and a 49 percent increase in the Permian Basin region. By comparison, there was a 6 percent decrease in all other 175 counties in the state.  In most cases, as the severity of the injuries worsened, the changes in the corresponding crash rate were more evident. For example, for rural crashes that involved CMVs in the Barnett Shale region, there was a 21 percent decrease in the crash rate for fatal, incapacitating, and non-incapacitating injury crashes (compared to a 26 percent decrease for all crashes). For fatal crashes, the decrease was 36 percent. In the Eagle Ford Shale region, there was a 65 percent increase in the crash rate for fatal, incapacitating, and nonincapacitating injury crashes (compared to a 57 percent increase for all crashes). For fatal crashes, the increase was 67 percent. In the Permian Basin region, there was a 51 percent increase in the crash rate for fatal, incapacitating, and non-incapacitating injury crashes (compared to a 49 percent increase for all crashes). For fatal crashes, the increase was 70 percent. 49 Table 24. Number of Crashes over Four Years per 100 Million VMTs. Note: Green dots correspond to decreases in crash rates (desirable trend). Red dots correspond to increases in crash rates (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09       121.8       115.1       126.5       130.6 Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown Crash Rates All Rural CMV‐Involved Rural & CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.       119.7 ‐2%          95.1          78.4 ‐18%       12.0       11.4 ‐5%          14.8          10.9 ‐26%       119.5 4%          69.6          71.4 3%       10.6       14.8 39%            8.1          12.7 57%       135.9 7%          59.2          66.7 13%       11.9       16.4 37%          10.8          16.0 49%       127.9 ‐2%          88.7          85.9 ‐3%       10.9       10.2 ‐6%          10.2            9.6 ‐6% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09          22.3          21.6          21.2          20.5 All 2010‐13          23.0          23.0          24.6          21.3 Diff. 3% 7% 16% 4% Fatal, Incapacitating, Non‐Incapacitating Crash Rates Rural CMV‐Involved 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.          22.3          18.8 ‐16%          2.0          1.8 ‐8%          18.0          17.8 ‐1%          2.1          3.1 49%          15.0          17.8 19%          2.5          3.7 50%          19.4          18.6 ‐4%          1.7          1.7 1% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.            3.4            2.7 ‐21%            2.0            3.3 65%            2.8            4.2 51%            2.1            2.1 ‐1% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09            1.1            1.5            1.6            1.2 All 2010‐13            0.9            1.6            2.0            1.2 Diff. ‐12% 6% 30% ‐2% 2006‐09            2.2            1.8            1.7            1.8 Rural 2010‐13            1.7            1.9            2.1            1.7 50 Fatal Crash Rates CMV‐Involved Diff. 2006‐09 2010‐13 Diff. ‐23%          0.2          0.1 ‐25% 4%          0.3          0.4 56% 23%          0.3          0.5 78% ‐7%          0.2          0.2 2% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.            0.6            0.4 ‐36%            0.3            0.5 67%            0.4            0.7 70%            0.3            0.3 7% Table 25. Number of Crashes over Four Years per 100 Lane-Miles. Note: Green dots correspond to decreases in crash rates (desirable trend). Red dots correspond to increases in crash rates (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09       777.9       261.2       169.2       532.2 Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown Crash Rates All Rural CMV‐Involved Rural & CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.       743.7 ‐4%       230.2       177.9 ‐23%       76.5       70.8 ‐8%          35.9          24.8 ‐31%       278.4 7%       123.4       133.2 8%       24.1       34.4 43%          14.4          23.7 65%       187.5 11%          59.8          69.9 17%       15.9       22.6 42%          10.9          16.8 54%       505.5 ‐5%       184.7       173.2 ‐6%       44.3       40.2 ‐9%          21.2          19.4 ‐9% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09       142.4          48.9          28.3          83.6 All 2010‐13       143.1          53.6          34.0          84.0 Diff. 0% 10% 20% 0% Fatal, Incapacitating, Non‐Incapacitating Crash Rates Rural CMV‐Involved 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.          54.0          42.6 ‐21%       12.8       11.5 ‐10%          31.9          33.3 4%          4.7          7.2 53%          15.2          18.7 23%          3.3          5.1 54%          40.5          37.5 ‐7%          7.1          6.9 ‐2% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.            8.2            6.1 ‐26%            3.6            6.2 74%            2.8            4.4 57%            4.5            4.3 ‐4% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09            6.8            3.5            2.1            5.0 All 2010‐13            5.8            3.8            2.8            4.7 Diff. ‐14% 9% 34% ‐5% 2006‐09            5.2            3.2            1.8            3.8 Rural 2010‐13            3.8            3.5            2.2            3.4 51 Fatal Crash Rates CMV‐Involved Diff. 2006‐09 2010‐13 Diff. ‐27%          1.3          0.9 ‐27% 10%          0.6          1.0 60% 28%          0.4          0.7 83% ‐10%          0.8          0.8 ‐1% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.            1.4            0.8 ‐40%            0.6            1.0 75%            0.4            0.7 76%            0.6            0.7 4% Table 25 shows changes in the number of crashes over four years per 100 lane-miles from 2006– 2009 to 2010–2013. In total, the crash rate decreased by 4 percent in the Barnett Shale region, increased by 7 percent in the Eagle Ford Shale region, and increased by 11 percent in the Permian Basin region. As a reference, the crash rate decreased by 5 percent in all other 175 counties in the state. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences within each region. For example:  The changes were more pronounced for rural crashes. In the Barnett Shale region, the rural crash rate decreased by 23 percent (compared to a 4 percent decrease overall in the region). In the Eagle Ford Shale region, the rural crash rate increased by 8 percent (compared to a 7 percent increase overall in the region). In the Permian Basin region, the rural crash rate increased by 17 percent (compared to an 11 percent increase overall in the region). The changes were even more pronounced for crashes that involved CMVs and, particularly, for rural crashes that involved CMVs. For rural crashes that involved CMVs, there was a 31 percent decrease in the Barnett Shale region, a 65 percent increase in the Eagle Ford Shale region, and a 54 percent increase in the Permian Basin region. By comparison, there was a 9 percent decrease in all other 175 counties in the state.  In most cases, as the severity of the injuries worsened, the changes in the corresponding crash rate were more pronounced. For example, for rural crashes that involved CMVs in the Barnett Shale region, there was a 26 percent decrease in the crash rate for fatal, incapacitating, and non-incapacitating injury crashes (compared to a 31 percent decrease for all crashes). For fatal crashes, the decrease was 40 percent. In the Eagle Ford Shale region, there was a 74 percent increase in the crash rate for fatal, incapacitating, and nonincapacitating injury crashes (compared to a 65 percent increase for all crashes). For fatal crashes, the increase was 75 percent. In the Permian Basin region, there was a 57 percent increase in the crash rate for fatal, incapacitating, and non-incapacitating injury crashes (compared to a 54 percent increase for all crashes). For fatal crashes, the increase was 76 percent. Although the changes in Table 24 and Table 25 are similar, it is worth noting that the basis for the analysis was slightly different, making a comparison between the tables difficult. The reason is that Table 24 relied on AADT data, which were only available for state highways (both rural and urban). By comparison, Table 25 relied on lane-miles, which were only available for rural roads (both state highways and county roads). It is also worth noting that the uncertainty in crash rates (when expressed as the number of crashes per 100 million VMT) increases as traffic volumes decrease, and becomes particularly evident in the case of highway segments with very low AADT values. Figure 9 and Figure 10 illustrate this situation. The uncertainty in crash rates is lower when expressing crash rates per 100 lane-miles (see Figure 11 and Figure 12). 52 All crashes Rural crashes Figure 9. Changes in Crash Rate (per 100 Million VMTs) from 2006–2009 to 2010–2013 – All Crashes and Rural Crashes. 53 CMV crashes Rural CMV crashes Figure 10. Changes in Crash Rate (per 100 Million VMTs) from 2006–2009 to 2010–2013 – CMV Crashes and Rural CMV Crashes. 54 All crashes Rural crashes Figure 11. Changes in Crash Rate (per 100 Lane-Miles) from 2006–2009 to 2010–2013 – All Crashes and Rural Crashes. 55 CMV crashes Rural CMV crashes Figure 12. Changes in Crash Rate (per 100 Lane-Miles) from 2006–2009 to 2010–2013 – CMV Crashes and Rural CMV Crashes. 56 Correlations The research team used the Pearson product-moment correlation coefficient to compare pairs of metrics using historical data aggregated at the county level. As a reference, a Pearson coefficient of −1 between any two variables indicates total negative correlation, 0 indicates no correlation, and +1 indicates total positive correlation. As the absolute value of the Pearson coefficient increases, the linear correlation between the two variables increases. Table 26 provides a rudimentary way to interpret Pearson correlation coefficients. The literature contains a variety of threshold alternatives and statistical procedures to measure the significance of the Pearson coefficient. Table 26. Rudimentary Thresholds to Interpret Pearson Correlation Coefficients. From/To (Positive)  +0.70 to +1.00  +0.40 to +0.69  +0.30 to +0.39  +0.20 to +0.29  0.00 to +0.19  From/To (Negative)  −0.70 to −1.00  −0.40 to −0.69  −0.30 to −0.39  −0.20 to −0.29  0.00 to −0.19  Correlation Strength  Very strong  Strong  Moderate  Weak  No or negligible  Table 27 summarizes the results of the analysis, using eight years of data from 2006 through 2013. In the Barnett Shale region, there was a strong correlation between the number of new horizontal wells and the number of crashes (regardless of location or type of vehicles involved). The correlation between the number of new vertical wells and the number of crashes was weak to negligible. Table 27. Pearson Correlation Coefficients. Number of New  Number of New  Horizontal Wells Vertical Wells 0.56 ‐0.15 0.62 ‐0.24 0.52 ‐0.19 0.63 ‐0.10 0.12 0.16 ‐0.07 ‐0.22 0.39 0.09 0.57 ‐0.10 ‐0.08 0.07 0.03 0.33 0.06 0.29 0.23 0.47 ‐0.04 ‐0.03 ‐0.03 0.00 ‐0.02 ‐0.02 0.00 0.06 Number of Crashes Barnett Shale  Number of Rural Crashes Region Number of CMV Crashes Number of Rural CMV Crashes Number of Crashes Eagle Ford  Number of Rural Crashes Shale Region Number of CMV Crashes Number of Rural CMV Crashes Number of Crashes Permian  Number of Rural Crashes Basin Region Number of CMV Crashes Number of Rural CMV Crashes Number of Crashes Remaining  Number of Rural Crashes 175 Counties Number of CMV Crashes Number of Rural CMV Crashes 57 In the Eagle Ford Shale region, there was a strong correlation between the number of new horizontal wells and the number of rural CMV crashes. There was also a moderate correlation between the number of horizontal wells and the number of CMV crashes. All other correlations were weak to negligible. In the Permian Basin region, there was a strong correlation between the number of new vertical wells and the number of rural CMV crashes. There was also a moderate correlation between the number of new vertical wells and the number of rural crashes and the number of CMV crashes. All other correlations were weak to negligible. Based on these results, the research team focused further on rural CMV-related crashes, more specifically, to establish potential correlations between changes in the number of rural CMV crashes from 2006–2009 to 2010–2013 at the county level and the corresponding changes in the number of new horizontal and vertical wells. Because of some uncertainties regarding the number of crashes resulting in unknown and possible injuries, the research team decided to focus on fatal, incapacitating, and non-incapacitating injury crashes. Table 28 shows the corresponding Pearson correlation coefficients. Table 28. Pearson Correlation Coefficients – Change in the Number of Rural KAB CMV Crashes vs. Change in the Number of New Wells from 2006–2009 to 2010–2013. Change in Number of  New Horizontal Wells Barnett Shale Region Eagle Ford Shale Region Permian Basin Region Remaining 175 Counties Change in  Numbe of  Rural KAB  CMV Crashes 0.88 0.77 0.22 ‐0.04 Change in Number of  New Vertical Wells ‐0.25 0.03 0.72 0.31 In the Barnett Shale region, there was a very strong correlation between the change in the number of new horizontal wells and that of rural CMV crashes. In the Eagle Ford Shale region, the correlation between these two variables was also very strong. In the Permian Basin region, there was a strong correlation between the change in the number of new vertical wells and that of rural CMV crashes. Other correlations were weak to negligible. Figure 13 shows a scatter plot of the change in the number of rural KAB CMV crashes versus the change in the number of new horizontal wells. Each point on the scatter plot corresponds to one county. Figure 13 also shows linear regression lines for county-level data from the Eagle Ford Shale, Barnett Shale, and Permian Basin regions. These regression models could be used for forecasting purposes in situations where other factors remain reasonably stable, and there is a need for high-level estimates. 58 Figure 13. Change in the Number of Rural KAB CMV Crashes vs. Change in the Number of Horizontal Wells. Overall, the predictive power of the regression models is much higher for the Eagle Ford Shale and Barnett Shale regions than for the Permian Basin region. Nevertheless, the similarity between the trends for the Eagle Ford Shale and the Barnett Shale regions in Figure 13 suggests a generalized trend that could be used to estimate positive (or negative) changes in the number of rural KAB CMV crashes in Texas as a function of the positive (or negative) change in the number of new completed horizontal wells. For example, for a county with 1,000 new horizontal wells over four years, the anticipated increase in the number of rural KAB CMV crashes could be 40. Similarly, for a county with 2,000 fewer new horizontal wells over four years, the anticipated decrease in the number of rural KAB CMV crashes could be about 70. The research team also developed statistical models to account for differences at the individual county level. The initial dataset included 2,032 records (one record per year per county) from 2006–2013. The research team used 2006–2012 data for model calibration and 2013 data for model validation. Initially, the research team explored several combinations of independent variables and variable selection methods, and used Bayesian analysis to improve model accuracy. After noticing that the corresponding improvement was negligible (compared to simpler non-Bayesian models), the research team developed three models that met an expectation of simplicity for potential implementation: 59 Model 1 (ordinary least squares regression model): 0.0281 0.0155 7.34 10 Model 2 (ordinary least squares regression model): 0.0349 0.0156 Model 3 (analysis of covariance (ANCOVA) model): 10.6 0.00428 0.00219 2.26 log or 10.6 0.00428 0.00219 2.26 3 8 where = County of interest (1 through 254) = Number of rural KAB CMV crashes on state and county roads per year in county = / = Mean value of = Number of new horizontal wells per year in county = Number of new vertical wells per year in county = VMT on state roads per year in county = Coefficient for county = Random error for county The appendix provides a list of all the coefficients and other relevant data related to the statistical analysis conducted to develop Models 1,2, and 3. 60 All independent variables used in the models are statistically significant at the 95 percent confidence interval. In all three models, the beta coefficient for was larger than the beta coefficient for , suggesting that an increase in the number of new horizontal wells is likely to yield more crashes, compared to the same increase in the number of new vertical wells. This is consistent with field observations, which point to a significantly larger number of trucks needed to develop horizontal wells compared to the number of trucks needed to develop vertical wells. Figure 14 shows a comparison between the observed number of rural KAB CMV crashes in 2013 and the predicted values using the three models. Overall, the predictive power (R2) of Model 1 was similar to that of Model 2, but Model 3 outperformed the other two models. Nevertheless, F-test and two-sample t-test results showed that the means of the predicted values and the variances of the predicted values were not statistically different. Figure 14. Observed versus Predicted Number of Rural KAB CMV Crashes in 2013. 61 To understand the similarities and differences between models further, the research team classified counties into three groups based on the 33rd and the 66th percentiles of the observed number of rural KAB CMV crashes in 2013: low (0–3), medium (4–9), and high (>10). Then, the research team estimated the average coefficient of variation (ACV) for every group within each region, as follows: 1 , , 100 where = Number of counties in Texas (254) , = Predicted number of crashes in county i , = Observed number of crashes in county i = Average of , and , Table 29 shows ACV values for all three models. In general, ACVs were higher for groups of counties with the lowest number of rural KAB CMV crashes, and decreased as the number of rural KAB CMV crashes increased. With a few exceptions, there were not significant differences between the models within each group, indicating that any of the models would perform adequately in most cases. 62 Table 29. Average Coefficient of Variation. Number of Rural  Number of  KAB CMV Crashes  Counties per County (2013) Low:          0‐3 4 Barnett Shale Medium: 4‐9 5  High:        >10  4 Low:          0‐3 0 Eagle Ford  Medium: 4‐9 14 Shale  High:        >10  15 Low:          0‐3 13 Permian  Medium: 4‐9 11 Basin  High:        >10  13 Low:          0‐3 70 Remaining  Medium: 4‐9 68 175 Counties  High:        >10  37 ACV  (Model 1) Region Low:         0‐3 Medium: 4‐9   High:        >10  ACV  (Model 2) ACV  (Model 3) 53.5% 14.9% 44.0% ‐ 30.1% 27.5% 77.2% 27.8% 38.7% 48.8% 34.0% 36.1% 117.1% 17.9% 39.0% ‐ 33.0% 28.5% 76.2% 36.3% 46.7% 55.2% 30.5% 26.2% 29.3% 18.8% 40.7% ‐ 27.0% 27.2% 76.7% 27.0% 40.3% 51.3% 29.5% 29.8% 53.3% 40.1% 35.2% 61.2% 30.8% 31.3% 54.1% 28.3% 31.8% Economic Impact of Crashes The research team developed preliminary estimates of the change in the cost of injuries from 2006–2009 to 2010–2013 using standardized economic and comprehensive crash cost estimates from NSC and comprehensive crash cost estimates from U.S. DOT (3, 4). Economic costs rely on calculable costs such as wage and productivity losses, medical expenses, administrative expenses, motor vehicle damage, and employers’ uninsured costs. Comprehensive costs include economic cost components and a measure of the value of lost quality of life, which makes comprehensive costs appropriate to analyze the anticipated benefit of future improvements (because they provide a measure of what people would be willing to pay for improved safety). In general, the U.S. DOT’s methodology for comprehensive cost estimates, which are based on a concept called the value of a statistical life, are considerably higher than those resulting from the NSC methodology. Economic costs in 2013 dollars, according to the NSC methodology, are as follows:  Death: $1,500,000  Incapacitating injury (i.e., Class A): $74,900  Non-incapacitating injury (i.e., Class B): $24,000 63  Possible injury (i.e., Class C): $13,600 Comprehensive costs include both the economic cost components above and a measure of the value of lost quality of life, and are therefore appropriate to use as a reference to analyze the anticipated benefit of future improvements (because they provide a measure of what people would be willing to pay for improved safety). In 2013 dollars, comprehensive costs of motor vehicle crashes are as follows:  Death $4,628,000  Incapacitating injury $235,400  Non-incapacitating evident injury $60,000  Possible injury $28,600 These dollar amounts do not include property damage-only crashes. According to the U.S. DOT methodology, VSL was $9.1 million in 2012 dollars. For subsequent years, the methodology recommends adjusting this value using a 1.07 percent annual growth rate. In 2014 dollars, adjusted VSL was $9,295,782. The guidance also includes six injury severity levels, each one with a factor to estimate the corresponding comprehensive cost, as shown in Table 30. Table 30. Injury Severity Levels and Estimated Comprehensive Costs (Adapted from 4). Severity Level  Fraction of  VSL  Minor  Moderate   Serious  Severe  Critical  Un‐survivable  0.003  0.047  0.105  0.266  0.593  1.000  Comprehensive  Cost Adjusted  for 2014  $27,887  $436,902  $976,057  $2,472,678  $5,512,399  $9,295,782  Corresponding Type  of Injury Used in the  Analysis  Estimated  Comprehensive  Cost  Possible Injury  $232,395  Non‐Incapacitating  $976,057  Incapacitating  $3,992,538  Fatal  $9,295,782  The severity levels in Table 30 provide a measure of survivability using the abbreviated injury scale, which is different from the system of classification of injuries in CRIS. This is a common problem reflecting different practices between police agencies (which normally complete crash reports) and health providers (which care for the injured). Information on how to map injury levels from one system to the other is not available. For simplicity, the research team assumed that possible injuries in CRIS captured a large percentage of minor and moderate injuries, nonincapacitating injuries in CRIS captured serious injuries, incapacitating injuries in CRIS included severe and critical injuries, and fatal injuries in CRIS corresponded to un-survivable injuries. 64 Table 30 shows the result of this mapping process, along with the corresponding comprehensive costs for each injury level in CRIS. Table 31 summarizes the result of the analysis. Because the correlation between new completed wells and rural CMV crashes was stronger than for other types of crashes, the research team only included the number of injuries resulting from rural CMV crashes. Further, the research team only included the number of fatal, incapacitating, non-incapacitating, and possible injuries in the cost calculation. In the Barnett Shale region, there was a 35 percent decrease (i.e., $73 million in economic costs or $425 million in comprehensive costs) in NSC-based costs and a 30 percent decrease (i.e., $763 million) in VSL-based comprehensive costs. The cost reduction was the result of fewer rural CMV crashes and, correspondingly, fewer injuries. In the Eagle Ford Shale region, there was a 52 percent increase (i.e., $139 million in economic costs or $801 million in comprehensive costs) in NSC-based costs and a 68 percent increase (i.e., $2 billion) in VSLbased comprehensive costs. In the Permian Basin region, there was a 103 percent increase (i.e., $176 million in economic costs or $1.03 billion in comprehensive costs) in NSC-based costs and a 97 percent increase (i.e., $2 billion) in VSL-based comprehensive costs. Table 31. Changes in Economic and Comprehensive Costs for Injuries Occurred in Rural CMV Crashes. Cost of Rural CMV Injuries (Million) Economic Cost (NSC) Comprehensive Cost (NSC) Region 2006‐09 2010‐13 Change Diff. 2006‐09 2010‐13 Change Diff. Barnett Shale $       212 $       138 $     (73) ‐35% $   1,224 $       799 $   (425) ‐35% Eagle Ford Shale $       269 $       408 $     139 52% $   1,548 $   2,349 $     801 52% Permian Basin $       171 $       348 $     176 103% $       981 $   2,011 $ 1,030 105% Other $   1,615 $   1,567 $     (47) ‐3% $   9,229 $   8,988 $   (241) ‐3% Grand Total $   2,266 $   2,461 $     194 9% $ 12,981 $ 14,146 $ 1,165 9% Comprehensive Cost (VSL) 2006‐09 2010‐13 Change Diff. $   2,510 $   1,747 $   (763) ‐30% $   2,931 $   4,927 $ 1,996 68% 97% $   2,051 $   4,045 $ 1,994 $ 19,796 $ 19,205 $   (591) ‐3% $ 27,288 $ 29,924 $ 2,636 10% The huge increase in the cost of injuries resulting from rural CMV crashes in the Eagle Ford Shale and Permian Basin regions (covering 66 counties in total) was largely responsible for the net increase in the cost of injuries resulting from rural CMV crashes in the state from 2006–2009 to 2010–2013. As Table 31 shows, the net increase was 9 percent overall, even though there was a 35 percent reduction in the Barnett Shale region (covering 13 counties) and a 3 percent reduction in 175 other counties around the state. 65 Chapter 3. Findings In fall 2014, the Texas Legislature asked TTI to update a study completed in late 2011 documenting locations and trends of oil and gas energy developments in the state (1). As part of the study, the Texas Legislature asked TTI to correlate oil and gas developments with changes in pavement condition data. TTI summarized the results of this analysis in a report published in March 2015 (2). To complement the study, the Texas Legislature asked TTI to gather and process crash data at a level of spatial and temporal detail needed to document locations and trends of crashes in relation to oil and gas energy developments in the state. To accomplish this goal, the research team gathered and processed data from TxDOT’s Crash Record Information System. Available data from CRIS covered the 2010–2014 period. The research team complemented this information with historical crash data from 2003–2009 that TTI had received from TxDOT before the introduction of CRIS. Location and attribute data about crashes and injuries (i.e., number of people who are injured in crashes) that the research team compiled included the following types of crashes:  All crashes.  Rural crashes (i.e., crashes that occur outside city limits).  CMV crashes (i.e., crashes in which CMVs are involved).  Rural CMV crashes.  Crashes on state highways.  Crashes on rural state highways.  CMV crashes on state highways.  CMV crashes on rural state highways. With this information, the research team examined changes in the number of crashes and the number of injuries from 2006–2009 to 2010–2013. These date ranges were used for consistency with those in the original March 2015 report. The year 2009 was significant because this was when accelerated oil production started in the Eagle Ford Shale region and oil production in the Permian Basin region began to accelerate, making the end of 2009 suitable for use as a baseline for comparison purposes. The last year with reliable Railroad Commission data was 2013 (2014 data were still preliminary). In addition, the economic recession of 2008 caused significant volatility in the oil markets, which resulted in dramatic swings in prices, drilling, and production. In order to reduce the impact of these variations, the research team aggregated and compared data using two four-year blocks: 2006–2009 and 2010–2013. 66 This chapter replicates tables from Chapter 2, which illustrate the changes. In particular, Table 32 shows changes in the number of crashes on all highways, and Table 33 shows changes in the number of injuries on all highways. Table 32 shows changes in the number of crashes on all highways from 2006–2009 to 2010–2013. In total, the number of crashes decreased by 10 percent in the Barnett Shale region, increased by 1 percent in the Eagle Ford Shale region, and decreased by 4 percent in the Permian Basin region. As a reference, the number of crashes decreased by 7 percent in all other 175 counties in the state. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. The changes were more prominent for rural crashes. The changes were even more noticeable for crashes that involved CMVs and, particularly, for rural crashes that involved CMVs. In most cases, as the severity of the injuries worsened, the changes in the corresponding number of crashes were more evident. Relative changes in the number of crashes on state highways were similar to those found for all highways. The changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. Overall, the percentage of crashes occurring on state highways increased. For all crashes, the increase was from 54 to 56 percent. For fatal, incapacitating, and non-incapacitating injury crashes, the increase was from 59 to 61 percent. For fatal crashes, the increase was from 74 to 76 percent. These percentages were higher for rural roads. For example, for rural CMV crashes, the percentage of crashes on state highways decreased slightly from 78 to 77 percent. For fatal, incapacitating, and non-incapacitating injury crashes, this percentage increased from 90 to 91 percent. For fatal crashes, it decreased slightly but stayed around 96 percent. The research team calculated crash rates expressed both as the number of crashes per 100 million VMT and number of crashes per 100 lane-miles. The results were similar with both approaches, although rates expressed as number of crashes per 100 lane-miles were more stable particularly for roadway segments with low traffic volumes. Table 34 shows the changes in the number of crashes over four years per 100 lane-miles from 2006–2009 to 2010–2013. In total, the crash rate decreased by 4 percent in the Barnett Shale region, increased by 7 percent in the Eagle Ford Shale region, and increased by 11 percent in the Permian Basin region. These changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region. The changes were more prominent for rural crashes. The changes were even more evident for crashes that involved CMVs and, in particular, for rural CMV crashes. In most cases, as the severity of the injuries worsened, the changes in the corresponding crash rate were more evident. 67 Table 32. Changes in the Number of Crashes on All Highways. Note: Green dots correspond to decreases in the number of crashes (desirable trend). Red dots correspond to increases in the number of crashes (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09        184,735          85,964          80,891     1,410,907    1,762,497 Number of Crashes (Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown) All Rural CMV‐Involved Rural & CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.        166,474 ‐10%       24,572       18,521 ‐25%     14,119       12,367 ‐12%      3,130      2,061 ‐34%          86,744 1%       27,660       28,804 4%       6,607         8,708 32%      2,820      4,542 61%          77,511 ‐4%       15,689       17,426 11%       4,775         6,368 33%      2,464      3,743 52%     1,306,749 ‐7%    288,715    284,431 ‐1%     90,081       77,755 ‐14%    26,221    23,942 ‐9%    1,637,478 ‐7%    356,636    349,182 ‐2%  115,582    105,198 ‐9%   34,635   34,288 ‐1% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09          31,739          14,382          11,520        204,134       261,775 All 2010‐13          30,728          15,264          12,019        201,541       259,552 Number of Fatal, Incapacitating, Non‐Incapacitating Crashes Rural CMV‐Involved Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. ‐3%         5,346         4,165 ‐22%       2,124         1,846 ‐13% 6%         6,889         6,948 1%       1,096         1,641 50% 4%         3,841         4,524 18%          883         1,333 51% ‐1%       57,296       54,123 ‐6%     12,568       11,792 ‐6% ‐1%      73,372      69,760 ‐5%    16,671      16,612 0% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.          642          474 ‐26%          662      1,173 77%          617          971 57%      4,998      4,751 ‐5%      6,919      7,369 7% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09            1,202                851                648            9,465         12,166 All 2010‐13            1,030                902                789            8,954         11,675 Diff. ‐14% 6% 22% ‐5% ‐4% 2006‐09            459            629            430         4,673        6,191 Number of Fatal Crashes Rural CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff.            325 ‐29%          181            135 ‐25%            694 10%          129            204 58%            518 20%            94            183 94%         4,293 ‐8%       1,177         1,170 ‐1%        5,830 ‐6%      1,582        1,692 7% 68 Rural & CMV‐Involved 2006‐09 2010‐13 Diff.          101            63 ‐37%          102          179 76%            80          151 88%          663          684 3%         946      1,077 14% Table 33. Changes in the Number of Injuries on All Highways. Note: Green dots correspond to decreases in the number of injuries (desirable trend). Red dots correspond to increases in the number of injuries (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09       521,304       241,484       225,483   4,003,334  4,991,605 Number of Injuries (Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown) All Rural CMV‐Involved Rural & CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.       446,866 ‐14%         58,038     38,861 ‐33%     56,707     37,827 ‐33%     14,923        5,518 ‐63%       230,497 ‐5%         63,650     61,679 ‐3%     28,518     24,812 ‐13%     19,625     11,395 ‐42%       202,170 ‐10%         37,286     37,788 1%     16,734     16,109 ‐4%        7,992        8,580 7%   3,546,753 ‐11%       738,597   691,735 ‐6%   370,987   254,011 ‐32%   109,719     72,430 ‐34%  4,426,286 ‐11%      897,571  830,063 ‐8%  472,946  332,759 ‐30%  152,258     97,923 ‐36% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09         42,345         20,637         15,704       278,208      356,894 All 2010‐13         40,919         21,945         16,515       273,913      353,292 Number of Fatal, Incapacitating, Non‐Incapacitating Injuries Rural CMV‐Involved Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. ‐3%           7,285        5,609 ‐23%        2,823        2,452 ‐13% 6%         10,207     10,257 0%        1,654        2,385 44% 5%           5,591        6,632 19%        1,115        1,868 67% ‐2%         80,963     75,957 ‐6%     17,635     16,385 ‐7% ‐1%      104,046     98,455 ‐5%     23,228     23,090 ‐1% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.           840           627 ‐25%        1,006        1,755 75%           786        1,379 76%        6,920        6,779 ‐2%       9,552     10,540 10% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other Grand Total 2006‐09           1,294           1,002              738         10,562        13,596 All 2010‐13           1,125           1,047              914           9,854        12,940 Diff. ‐13% 4% 24% ‐7% ‐5% 2006‐09              499              763              499           5,347          7,108 Number of Fatal Injuries Rural CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff.           364 ‐27%           201           150 ‐26%           825 8%           169           237 40%           619 24%           103           224 117%        4,837 ‐10%        1,433        1,326 ‐7%       6,645 ‐7%       1,907       1,937 2% 69 Rural & CMV‐Involved 2006‐09 2010‐13 Diff.           111             71 ‐36%           143           211 47%             88           184 109%           813           797 ‐2%       1,155       1,263 9% Table 34. Number of Crashes over Four Years per 100 Lane-Miles. Note: Green dots correspond to decreases in crash rates (desirable trend). Red dots correspond to increases in crash rates (undesirable trend). (a) All crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09       777.9       261.2       169.2       532.2 Fatal, Incapacitating, Non‐Incapacitating, Possible Injury, No‐Injury, Unknown Crash Rates All Rural CMV‐Involved Rural & CMV‐Involved 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.       743.7 ‐4%       230.2       177.9 ‐23%       76.5       70.8 ‐8%          35.9          24.8 ‐31%       278.4 7%       123.4       133.2 8%       24.1       34.4 43%          14.4          23.7 65%       187.5 11%          59.8          69.9 17%       15.9       22.6 42%          10.9          16.8 54%       505.5 ‐5%       184.7       173.2 ‐6%       44.3       40.2 ‐9%          21.2          19.4 ‐9% (b) Fatal, incapacitating, and non-incapacitating injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09       142.4          48.9          28.3          83.6 All 2010‐13       143.1          53.6          34.0          84.0 Diff. 0% 10% 20% 0% Fatal, Incapacitating, Non‐Incapacitating Crash Rates Rural CMV‐Involved 2006‐09 2010‐13 Diff. 2006‐09 2010‐13 Diff.          54.0          42.6 ‐21%       12.8       11.5 ‐10%          31.9          33.3 4%          4.7          7.2 53%          15.2          18.7 23%          3.3          5.1 54%          40.5          37.5 ‐7%          7.1          6.9 ‐2% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.            8.2            6.1 ‐26%            3.6            6.2 74%            2.8            4.4 57%            4.5            4.3 ‐4% (c) Fatal injury crashes Region Barnett Shale Eagle Ford Shale Permian Basin Other 2006‐09            6.8            3.5            2.1            5.0 All 2010‐13            5.8            3.8            2.8            4.7 Diff. ‐14% 9% 34% ‐5% 2006‐09            5.2            3.2            1.8            3.8 Rural 2010‐13            3.8            3.5            2.2            3.4 70 Fatal Crash Rates CMV‐Involved Diff. 2006‐09 2010‐13 Diff. ‐27%          1.3          0.9 ‐27% 10%          0.6          1.0 60% 28%          0.4          0.7 83% ‐10%          0.8          0.8 ‐1% Rural & CMV‐Involved 2006‐09 2010‐13 Diff.            1.4            0.8 ‐40%            0.6            1.0 75%            0.4            0.7 76%            0.6            0.7 4% The research team established correlations by comparing pairs of metrics representing historical data aggregated at the county level. In the Barnett Shale region, there was a strong correlation between the number of new horizontal wells and the number of crashes (regardless of location or type of vehicles involved). In the Eagle Ford Shale region, there was a strong correlation between the number of new horizontal wells and the number of rural CMV crashes. In the Permian Basin region, there was a strong correlation between the number of new vertical wells and the number of rural CMV crashes. It is worth noting that in the Permian Basin, although the relative change in the number of horizontal wells was considerably higher than for vertical wells, in absolute terms the number of new vertical wells was much higher than the number of new horizontal wells. Judging from the trends in the Eagle Ford Shale and Barnett Shale regions, as the industry shifts from vertical drilling to horizontal drilling in the Permian Basin, the correlation between new horizontal wells and rural CMV crashes in that part of the state will likely increase. In the Barnett Shale region, there was a very strong correlation between the change in the number of new horizontal wells and the change in the number of rural CMV crashes. In the Eagle Ford Shale region, the correlation between these two variables was also very strong. In the Permian Basin region, there was a strong correlation between the change in the number of new vertical wells and that of rural CMV crashes. The research team used this information to develop regression models for county-level data from the Eagle Ford Shale, Barnett Shale, and Permian Basin regions that could be used for forecasting purposes in situations where other factors remain reasonably stable and there is a need for high-level estimates. These models suggest a generalized trend that could be used to estimate positive (or negative) changes in the number of rural CMV crashes in Texas as a function of the positive (or negative) change in the number of new horizontal wells. The research team also developed statistical models to account for differences at the individual county level. The result of the modeling effort was three models: two least squares regression models and an analysis of covariance model with a transformed independent variable. The models were calibrated using data from 2006–2012 and validated using data from 2013. The research team developed preliminary estimates of the change in the cost of injuries from 2006–2009 to 2010–2013 using standardized economic and comprehensive crash cost estimates from NSC and comprehensive crash cost estimates from U.S. DOT. Because the correlation between new completed wells and rural CMV crashes was stronger than for other types of crashes, the research team only included the number of injuries in rural CMV crashes. Further, the research team only included the number of fatal, incapacitating, non-incapacitating, and possible injuries in the calculation. Table 35 summarizes the result of the analysis. In the Barnett Shale region, there was a 35 percent decrease (i.e., $73 million in economic costs or $425 million in comprehensive costs) in NSC-based costs and a 30 percent decrease (i.e., $763 million) in VSL-based comprehensive costs. The cost reduction was the result of fewer rural CMV crashes and, correspondingly, fewer 71 injuries. In the Eagle Ford Shale region, there was a 52 percent increase (i.e., $139 million in economic costs or $801 million in comprehensive costs) in NSC-based costs and a 68 percent increase (i.e., $2 billion) in VSL-based comprehensive costs. In the Permian Basin region, there was a 103 percent increase (i.e., $176 million in economic costs or $1.03 billion in comprehensive costs) in NSC-based costs and a 97 percent increase (i.e., $2 billion) in VSLbased comprehensive costs. Table 35. Changes in Economic and Comprehensive Costs for Injuries Occurred in Rural CMV Crashes. Cost of Rural CMV Injuries (Million) Economic Cost (NSC) Comprehensive Cost (NSC) Region 2006‐09 2010‐13 Change Diff. 2006‐09 2010‐13 Change Diff. Barnett Shale $       212 $       138 $     (73) ‐35% $   1,224 $       799 $   (425) ‐35% Eagle Ford Shale $       269 $       408 $     139 52% $   1,548 $   2,349 $     801 52% Permian Basin $       171 $       348 $     176 103% $       981 $   2,011 $ 1,030 105% Other $   1,615 $   1,567 $     (47) ‐3% $   9,229 $   8,988 $   (241) ‐3% Grand Total $   2,266 $   2,461 $     194 9% $ 12,981 $ 14,146 $ 1,165 9% Comprehensive (VSL) 2006‐09 2010‐13 Change Diff. $   2,510 $   1,747 $   (763) ‐30% $   2,931 $   4,927 $ 1,996 68% 97% $   2,051 $   4,045 $ 1,994 $ 19,796 $ 19,205 $   (591) ‐3% $ 27,288 $ 29,924 $ 2,636 10% The huge increase in the cost of injuries resulting from rural CMV crashes in the Eagle Ford Shale and Permian Basin regions (covering 66 counties in total) was largely responsible for the net increase in the cost of injuries resulting from rural CMV crashes in the state from 2006–2009 to 2010–2013. As Table 35 shows, the net increase was 9 percent overall, even though there was a 35 percent reduction in the Barnett Shale region (covering 13 counties) and a 3 percent reduction in 175 other counties around the state. In practical terms, the research results mean the following:  The number of crashes and resulting injuries increased along with oil and gas well development activities, but the changes were not uniform either by crash location and type of vehicles involved or by injury severity. There were also significant differences geographically within each region.  The increases in the number of crashes and injuries were more prominent in rural areas where energy developments take place (i.e., Eagle Ford Shale and Permian Basin regions). The highest increase was in the case of rural CMV crashes. Overall, there was a strong correlation between rural CMV crashes and the number of new wells.  The percentage of crashes on state highways increased. As the severity of the injuries increased, the percentage of crashes on state highways also increased. For rural CMV crashes, the percentage of crashes on state highways increased from 81 to 83 percent. For fatal, incapacitating, and non-incapacitating injury crashes, this percentage increased from 89 to 90 percent. For fatal crashes, it increased slightly but stayed around 95 percent. 72  The cost of injuries resulting from rural CMV crashes in energy development regions increased significantly and was largely responsible for the net increase in the cost of injuries resulting from rural CMV crashes in the state from 2006–2009 to 2010–2013. In the Eagle Ford Shale region, the increase was $139 million in economic costs or $801 million–$2 billion in comprehensive costs. In the Permian Basin region, the increase was $176 million in economic costs or $1.03–2.0 billion in comprehensive costs. These costs are of the same order of magnitude as the impact of energy developments on the transportation infrastructure (estimated at $1 billion per year on state highways and an additional $1 billion on county and local roads). 73 References 1. Quiroga, C., Fernando, E. and Oh, J. Energy Developments and the Transportation Infrastructure in Texas: Impacts and Strategies. Report FHWA/TX-12/0-6498-1, Texas Transportation Institute, Project 0-6498, March 2012. 2. Quiroga, C., Tsapakis, I., Kraus, E., Le, J., and Wimsatt, A. Oil and Gas Energy Developments and Changes in Pavement Conditions in Texas. Final Report. Report PRC 14-35F. Transportation Policy Center, Texas A&M Transportation, College Station, Texas, March 2015. 3. Estimating the Cost of Unintentional Injuries, 2013. National Safety Council, Itasca, Illinois, 2015. 4. Treatment of the Value of Preventing Fatalities and Injuries in Preparing Economic Analyses. Office of the Secretary of Transportation, U.S. Department of Transportation, 2013. 74 Appendix This appendix provides relevant data related to the statistical analysis conducted to develop Model 1 (Table 36), Model 2 (Table 37), and Model 3 (Table 38). Table 36. Summary Results of Statistical Analysis for Model 1. Parameter          0.0281  0.0155  7.34E‐06  Standard  Error  t  Significance 0.002  0.001  7.51E‐08  18.007  14.643  97.768  1.02E‐66  6.40E‐46  1.12E‐74  95% Confidence  Interval  Lower  Upper  Bound  Bound  0.0250  0.0135  7.19E‐06  0.0311  0.0176  7.49E‐06  Partial  Eta  Squared  0.154  0.108  0.843  Table 37. Summary Results of Statistical Analysis for Model 2. Parameter        0.0349  0.0156  Standard  Error  t  Significance 0.00232  0.00172  15.01  9.07  1.26E‐47  3.67E‐19  95% Confidence  Interval  Lower  Upper  Bound  Bound  .0303  .0122  .0395  .0190  Partial Eta  Squared  0.129  0.0512  Parameter Estimates for Counties in the Barnett Shale Region  Cooke  Denton  Erath  Hill  Hood  Jack  Johnson  Montague  Palo Pinto  Parker  Somervell  Tarrant  Wise  3.60  6.58  7.45  9.50  5.24  2.20  7.74  5.66  5.87  11.7  3.81  ‐8.24  12.0  1.30  1.34  1.29  1.29  1.32  1.30  1.71  1.31  1.29  1.35  1.29  1.74  1.36  2.78  4.91  5.76  7.34  3.98  1.69  4.54  4.32  4.54  8.61  2.95  ‐4.75  8.86  5.51E‐03  9.88E‐07  9.98E‐09  3.42E‐13  7.19E‐05  9.05E‐02  6.20E‐06  1.70E‐05  6.11E‐06  1.74E‐17  3.21E‐03  2.24E‐06  2.20E‐18  1.06  3.95  4.92  6.96  2.66  ‐0.347  4.39  3.09  3.34  9.01  1.28  ‐11.7  9.35  6.15  9.20  9.99  12.0  7.82  4.74  11.1  8.24  8.41  14.3  6.35  ‐4.84  14.7  5.05E‐03  1.56E‐02  2.14E‐02  3.42E‐02  1.03E‐02  1.88E‐03  1.33E‐02  1.21E‐02  1.34E‐02  4.65E‐02  5.69E‐03  1.46E‐02  4.90E‐02  Parameter Estimates for Counties in the Eagle Ford Shale Region  Atascosa  Bastrop  Bee  Brazos  Burleson  8.39  11.3  2.79  4.71  4.39  1.29  1.29  1.29  1.29  1.29  6.48  8.77  2.16  3.64  3.40  1.23E‐10  4.80E‐18  3.11E‐02  2.82E‐04  7.01E‐04  75 5.85  8.79  0.254  2.17  1.85  10.9  13.9  5.33  7.24  6.92  2.69E‐02  4.81E‐02  3.05E‐03  8.63E‐03  7.52E‐03  Parameter  Caldwell  DeWitt  Dimmit  Duval  Fayette  Frio  Goliad  Gonzales  Grimes  Guadalupe  Jim Wells  Karnes  Lampasas  Lavaca  Lee  Live Oak  Maverick  McMullen  Robertson  Victoria  Washington  Webb  Wilson  Zavala    3.76  1.74  ‐0.0300  3.37  8.31  3.26  2.88  6.90  9.38  10.1  7.37  3.81  3.29  3.83  6.20  6.21  ‐2.57  4.47  13.5  8.59  8.05  6.40  3.25  3.07  Standard  Error  t  Significance 1.29  1.30  1.35  1.29  1.29  1.29  1.29  1.30  1.29  1.29  1.29  1.32  1.32  1.29  1.29  1.29  1.43  1.30  1.30  1.29  1.29  1.38  1.29  1.29  2.91  1.33  ‐0.021  2.60  6.43  2.52  2.23  5.30  7.26  7.81  5.71  2.88  2.48  2.96  4.80  4.80  ‐1.80  3.45  10.4  6.64  6.23  4.64  2.52  2.38  3.67E‐03  1.82E‐01  9.83E‐01  9.40E‐03  1.69E‐10  1.17E‐02  2.59E‐02  1.32E‐07  6.25E‐13  1.07E‐14  1.39E‐08  4.05E‐03  1.31E‐02  3.15E‐03  1.72E‐06  1.75E‐06  7.19E‐02  5.83E‐04  1.23E‐24  4.41E‐11  5.88E‐10  3.72E‐06  1.20E‐02  1.75E‐02  95% Confidence  Interval  Lower  Upper  Bound  Bound  1.23  ‐.819  ‐2.67  0.827  5.78  0.727  0.346  4.35  6.84  7.55  4.84  1.21  0.693  1.29  3.67  3.67  ‐5.37  1.92  11.0  6.05  5.52  3.70  0.715  0.539  6.30  4.30  2.61  5.90  10.8  5.80  5.42  9.46  11.9  12.6  9.90  6.41  5.89  6.37  8.74  8.75  0.23  7.01  16.1  11.1  10.6  9.10  5.78  5.61  Partial Eta  Squared  5.53E‐03  1.17E‐03  2.95E‐07  4.42E‐03  2.65E‐02  4.17E‐03  3.25E‐03  1.81E‐02  3.35E‐02  3.85E‐02  2.09E‐02  5.41E‐03  4.04E‐03  5.71E‐03  1.49E‐02  1.49E‐02  2.13E‐03  7.74E‐03  6.67E‐02  2.81E‐02  2.49E‐02  1.40E‐02  4.14E‐03  3.70E‐03  Parameter Estimates for Counties in the Permian Basin Region  Andrews  Borden  Cochran  Coke  Crane  Crockett  Crosby  Dawson  Dickens  Ector  Gaines  Garza  Glasscock  Hockley  Howard  ‐2.07  ‐0.29  0.870  1.87  ‐0.390  3.68  ‐0.06  0.48  0.89  11.1  ‐0.31  2.82  0.08  4.83  7.11  1.71  1.29  1.29  1.29  1.31  1.33  1.29  1.29  1.29  1.43  1.32  1.29  1.37  1.30  1.31  ‐1.21  ‐0.223  0.672  1.45  ‐0.295  2.77  ‐0.043  0.373  0.691  7.76  ‐0.234  2.18  0.058  3.73  5.41  2.26E‐01  8.24E‐01  5.01E‐01  1.48E‐01  7.68E‐01  5.65E‐03  9.65E‐01  7.10E‐01  4.90E‐01  1.53E‐14  8.15E‐01  2.92E‐02  9.54E‐01  2.00E‐04  7.22E‐08  76 ‐5.41  ‐2.82  ‐1.67  ‐0.663  ‐2.95  1.08  ‐2.60  ‐2.06  ‐1.64  8.29  ‐2.91  0.286  ‐2.61  2.29  4.53  1.28  2.25  3.40  4.41  2.18  6.29  2.48  3.02  3.43  13.9  2.29  5.36  2.76  7.37  9.68  9.63E‐04  3.26E‐05  2.97E‐04  1.38E‐03  5.73E‐05  5.02E‐03  1.24E‐06  9.12E‐05  3.14E‐04  3.81E‐02  3.59E‐05  3.12E‐03  2.20E‐06  9.05E‐03  1.89E‐02  Parameter  Irion  Kent  Loving  Lubbock  Lynn  Martin  Midland  Mitchell  Pecos  Reagan  Reeves  Schleicher  Scurry  Sterling  Sutton  Terrell  Terry  Tom Green  Upton  Ward  Winkler  Yoakum    ‐0.27  0.21  0.01  6.91  2.19  6.08  13.1  4.12  5.19  ‐1.10  5.26  0.86  6.16  0.30  3.87  1.02  1.44  8.31  ‐2.73  0.57  0.41  0.59  Standard  Error  t  Significance 1.30  1.29  1.29  1.29  1.29  1.30  1.41  1.33  1.33  1.38  1.30  1.29  1.30  1.29  1.30  1.29  1.29  1.29  1.47  1.34  1.30  1.31  ‐0.212  0.161  0.006  5.35  1.70  4.66  9.32  3.10  3.91  ‐0.801  4.03  0.664  4.74  0.235  2.98  0.786  1.11  6.43  ‐1.86  0.421  0.317  0.452  8.33E‐01  8.72E‐01  9.95E‐01  1.03E‐07  8.95E‐02  3.48E‐06  3.95E‐20  1.94E‐03  9.69E‐05  4.23E‐01  5.76E‐05  5.07E‐01  2.28E‐06  8.14E‐01  2.94E‐03  4.32E‐01  2.67E‐01  1.75E‐10  6.35E‐02  6.74E‐01  7.51E‐01  6.51E‐01  95% Confidence  Interval  Lower  Upper  Bound  Bound  ‐2.82  ‐2.33  ‐2.53  4.37  ‐0.339  3.52  10.3  1.52  2.58  ‐3.80  2.70  ‐1.68  3.62  ‐2.24  1.32  ‐1.52  ‐1.10  5.77  ‐5.61  ‐2.07  ‐2.13  ‐1.98  Partial Eta  Squared  2.27  2.74  2.55  9.44  4.73  8.63  15.9  6.72  7.79  1.60  7.81  3.39  8.71  2.84  6.41  3.55  3.97  10.8  0.15  3.20  2.96  3.17  2.94E‐05  1.71E‐05  2.24E‐08  1.84E‐02  1.89E‐03  1.41E‐02  5.40E‐02  6.29E‐03  9.94E‐03  4.22E‐04  1.06E‐02  2.89E‐04  1.46E‐02  3.62E‐05  5.80E‐03  4.06E‐04  8.09E‐04  2.64E‐02  2.26E‐03  1.16E‐04  6.62E‐05  1.34E‐04  10.9  11.6  4.57  4.20  5.11  9.18  5.82  4.53  4.89  11.5  24.2  4.39  4.30  13.8  16.3  3.67  2.82  2.70E‐02  3.10E‐02  1.63E‐03  1.06E‐03  2.60E‐03  1.71E‐02  4.23E‐03  1.56E‐03  2.18E‐03  3.09E‐02  1.56E‐01  1.36E‐03  1.23E‐03  4.76E‐02  6.97E‐02  5.12E‐04  3.16E‐05  Parameter Estimates for the Remaining 175 Counties  Anderson  Angelina  Aransas  Archer  Armstrong  Austin  Bailey  Bandera  Baylor  Bell  Bexar  Blanco  Bosque  Bowie  Brazoria  Brewster  Briscoe  8.40  9.02  2.03  1.65  2.57  6.64  3.29  1.99  2.36  9.00  21.7  1.86  1.77  11.3  13.8  1.14  0.28  1.29  1.29  1.29  1.30  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  6.50  6.98  1.57  1.27  1.99  5.14  2.54  1.54  1.82  6.97  16.8  1.44  1.37  8.73  10.7  0.883  0.219  1.07E‐10  4.43E‐12  1.16E‐01  2.04E‐01  4.67E‐02  3.10E‐07  1.11E‐02  1.23E‐01  6.83E‐02  4.79E‐12  3.59E‐58  1.51E‐01  1.71E‐01  6.79E‐18  1.05E‐25  3.77E‐01  8.26E‐01  77 5.87  6.48  ‐0.500  ‐0.899  0.038  4.11  0.752  ‐0.540  ‐0.177  6.47  19.2  ‐0.677  ‐0.766  8.74  11.3  ‐1.39  ‐2.25  Parameter  Brooks  Brown  Burnet  Calhoun  Callahan  Cameron  Camp  Carson  Cass  Castro  Chambers  Cherokee  Childress  Clay  Coleman  Collin  Collingsworth  Colorado  Comal  Comanche  Concho  Coryell  Cottle  Culberson  Dallam  Dallas  Deaf Smith  Delta  Donley  Eastland  Edwards  El Paso  Ellis  Falls  Fannin  Fisher  Floyd  Foard  Fort Bend  Franklin    4.57  2.27  5.00  1.46  6.64  3.70  2.27  6.34  9.67  4.14  21.6  8.40  1.28  5.23  0.91  6.86  0.85  11.6  7.00  3.10  1.21  5.13  0.34  3.41  4.14  2.39  6.00  0.71  5.86  9.15  0.01  8.43  9.26  5.98  4.42  0.40  0.70  ‐0.100  13.5  2.62  Standard  Error  t  Significance 1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  3.53  1.76  3.87  1.13  5.14  2.87  1.76  4.91  7.49  3.21  16.7  6.50  0.990  4.04  0.708  5.31  0.657  8.99  5.42  2.40  0.935  3.97  0.263  2.64  3.20  1.85  4.64  0.553  4.53  7.08  0.010  6.52  7.17  4.63  3.42  0.309  0.542  ‐0.076  10.4  2.03  4.23E‐04  7.88E‐02  1.13E‐04  2.58E‐01  3.07E‐07  4.20E‐03  7.85E‐02  1.01E‐06  1.17E‐13  1.37E‐03  6.54E‐58  1.08E‐10  3.22E‐01  5.50E‐05  4.79E‐01  1.27E‐07  5.12E‐01  7.19E‐19  6.96E‐08  1.64E‐02  3.50E‐01  7.49E‐05  7.93E‐01  8.32E‐03  1.38E‐03  6.48E‐02  3.70E‐06  5.80E‐01  6.23E‐06  2.20E‐12  9.92E‐01  9.24E‐11  1.17E‐12  3.98E‐06  6.31E‐04  7.58E‐01  5.88E‐01  9.39E‐01  1.02E‐24  4.25E‐02  78 95% Confidence  Interval  Lower  Upper  Bound  Bound  2.03  ‐0.262  2.47  ‐1.07  4.11  1.17  ‐0.260  3.81  7.14  1.61  19.1  5.87  ‐1.25  2.69  ‐1.62  4.32  ‐1.69  9.08  4.47  0.568  ‐1.33  2.60  ‐2.19  0.879  1.60  ‐0.147  3.47  ‐1.82  3.32  6.61  ‐2.52  5.89  6.73  3.45  1.89  ‐2.14  ‐1.83  ‐2.63  11.0  0.088  7.10  4.81  7.53  4.00  9.18  6.24  4.81  8.88  12.2  6.68  24.2  10.9  3.81  7.76  3.45  9.39  3.38  14.2  9.53  5.64  3.74  7.66  2.87  5.95  6.67  4.92  8.53  3.25  8.39  11.7  2.55  11.0  11.8  8.51  6.96  2.93  3.23  2.44  16.0  5.16  Partial Eta  Squared  8.14E‐03  2.03E‐03  9.75E‐03  8.41E‐04  1.71E‐02  5.37E‐03  2.03E‐03  1.56E‐02  3.55E‐02  6.71E‐03  1.56E‐01  2.70E‐02  6.44E‐04  1.06E‐02  3.29E‐04  1.82E‐02  2.83E‐04  5.04E‐02  1.89E‐02  3.78E‐03  5.74E‐04  1.03E‐02  4.53E‐05  4.57E‐03  6.70E‐03  2.24E‐03  1.40E‐02  2.01E‐04  1.33E‐02  3.19E‐02  6.32E‐08  2.72E‐02  3.27E‐02  1.39E‐02  7.65E‐03  6.26E‐05  1.93E‐04  3.79E‐06  6.69E‐02  2.70E‐03  Parameter  Freestone  Galveston  Gillespie  Gray  Grayson  Gregg  Hale  Hall  Hamilton  Hansford  Hardeman  Hardin  Harris  Harrison  Hartley  Haskell  Hays  Hemphill  Henderson  Hidalgo  Hopkins  Houston  Hudspeth  Hunt  Hutchinson  Jackson  Jasper  Jeff Davis  Jefferson  Jim Hogg  Jones  Kaufman  Kendall  Kenedy  Kerr  Kimble  King  Kinney  Kleberg  Knox    10.5  4.01  3.86  7.50  5.21  20.7  3.98  2.00  1.12  1.54  1.98  8.99  109.0  18.6  5.48  0.96  7.43  2.76  6.12  14.2  9.42  5.84  12.4  11.4  1.30  4.18  6.73  1.70  11.7  1.13  3.15  8.71  4.86  1.52  5.86  3.27  1.45  1.29  2.84  0.65  Standard  Error  t  Significance 1.32  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.32  1.29  1.29  1.29  1.32  1.29  1.30  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  7.96  3.10  2.99  5.80  4.03  16.0  3.08  1.55  0.870  1.19  1.53  6.95  84.4  14.1  4.24  0.746  5.75  2.09  4.74  10.9  7.29  4.52  9.61  8.85  1.01  3.24  5.21  1.32  9.08  0.875  2.44  6.74  3.76  1.18  4.53  2.53  1.12  1.00  2.19  0.499  3.30E‐15  1.95E‐03  2.87E‐03  7.82E‐09  5.84E‐05  3.11E‐53  2.10E‐03  1.22E‐01  3.84E‐01  2.34E‐01  1.26E‐01  5.49E‐12  8.32E‐420  9.09E‐43  2.36E‐05  4.56E‐01  1.07E‐08  3.70E‐02  2.36E‐06  1.00E‐26  5.01E‐13  6.58E‐06  2.93E‐21  2.43E‐18  3.14E‐01  1.24E‐03  2.11E‐07  1.87E‐01  3.24E‐19  3.82E‐01  1.50E‐02  2.18E‐11  1.76E‐04  2.40E‐01  6.23E‐06  1.13E‐02  2.63E‐01  3.20E‐01  2.83E‐02  6.18E‐01  79 95% Confidence  Interval  Lower  Upper  Bound  Bound  7.92  1.48  1.32  4.97  2.67  18.1  1.45  ‐0.534  ‐1.41  ‐0.996  ‐0.558  6.45  106  16.1  2.95  ‐1.57  4.89  0.166  3.59  11.7  6.88  3.31  9.88  8.89  ‐1.23  1.65  4.20  ‐0.829  9.21  ‐1.40  0.613  6.18  2.32  ‐1.01  3.32  0.741  ‐1.09  ‐1.25  .301  ‐1.89  13.1  6.54  6.39  10.0  7.74  23.2  6.51  4.53  3.66  4.07  4.51  11.5  112  21.2  8.01  3.50  9.96  5.35  8.65  16.8  11.9  8.38  14.9  14.0  3.84  6.72  9.27  4.24  14.3  3.66  5.69  11.2  7.39  4.06  8.39  5.81  3.98  3.82  5.37  3.18  Partial Eta  Squared  4.00E‐02  6.29E‐03  5.82E‐03  2.17E‐02  1.06E‐02  1.44E‐01  6.20E‐03  1.57E‐03  4.97E‐04  9.31E‐04  1.54E‐03  3.07E‐02  8.24E‐01  1.16E‐01  1.17E‐02  3.66E‐04  2.13E‐02  2.85E‐03  1.45E‐02  7.25E‐02  3.37E‐02  1.33E‐02  5.72E‐02  4.89E‐02  6.67E‐04  6.83E‐03  1.75E‐02  1.14E‐03  5.14E‐02  5.03E‐04  3.88E‐03  2.90E‐02  9.20E‐03  9.09E‐04  1.33E‐02  4.20E‐03  8.23E‐04  6.50E‐04  3.15E‐03  1.64E‐04  Parameter  La Salle  Lamar  Lamb  Leon  Liberty  Limestone  Lipscomb  Llano  Madison  Marion  Mason  Matagorda  McCulloch  McLennan  Medina  Menard  Milam  Mills  Montgomery  Moore  Morris  Motley  Nacogdoches  Navarro  Newton  Nolan  Nueces  Ochiltree  Oldham  Orange  Panola  Parmer  Polk  Potter  Presidio  Rains  Randall  Real  Red River  Refugio    3.71  2.50  2.11  14.1  13.2  4.83  ‐1.44  3.14  5.13  2.42  ‐0.15  5.02  1.14  11.3  6.79  1.83  8.23  1.85  39.5  4.62  4.29  0.28  10.7  7.87  4.60  10.6  7.52  0.24  8.62  8.00  20.8  5.00  16.0  10.2  0.57  0.28  5.29  1.56  4.55  3.27  Standard  Error  t  Significance 1.29  1.29  1.29  1.29  1.29  1.30  1.31  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.30  1.29  1.29  1.29  1.29  1.29  1.31  1.29  1.29  1.29  1.29  1.30  1.29  1.29  1.35  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.31  2.88  1.94  1.63  10.9  10.2  3.72  ‐1.10  2.43  3.97  1.87  ‐0.118  3.89  0.885  8.76  5.26  1.42  6.31  1.44  30.6  3.57  3.32  0.219  8.17  6.09  3.56  8.16  5.81  0.185  6.67  6.20  15.4  3.87  12.4  7.87  0.442  0.218  4.09  1.21  3.53  2.50  4.09E‐03  5.31E‐02  1.03E‐01  1.66E‐26  1.20E‐23  2.06E‐04  2.72E‐01  1.51E‐02  7.44E‐05  6.15E‐02  9.06E‐01  1.07E‐04  3.76E‐01  5.13E‐18  1.69E‐07  1.57E‐01  3.64E‐10  1.51E‐01  2.81E‐160  3.68E‐04  9.29E‐04  8.26E‐01  6.37E‐16  1.39E‐09  3.78E‐04  6.92E‐16  7.48E‐09  8.53E‐01  3.53E‐11  7.43E‐10  4.00E‐50  1.13E‐04  1.15E‐33  6.55E‐15  6.58E‐01  8.28E‐01  4.50E‐05  2.28E‐01  4.34E‐04  1.25E‐02  80 95% Confidence  Interval  Lower  Upper  Bound  Bound  1.18  ‐0.034  ‐0.424  11.5  10.6  2.28  ‐4.00  0.609  2.60  ‐0.116  ‐2.69  2.48  ‐1.39  8.79  4.26  ‐0.706  5.67  ‐0.679  36.9  2.08  1.75  ‐2.25  8.12  5.34  2.07  8.02  4.98  ‐2.30  6.08  5.47  18.1  2.47  13.5  7.64  ‐1.96  ‐2.25  2.75  ‐0.977  2.02  0.706  6.25  5.03  4.64  16.6  15.7  7.38  1.13  5.68  7.67  4.95  2.38  7.55  3.68  13.9  9.33  4.36  10.8  4.39  42.0  7.15  6.82  2.82  13.3  10.4  7.14  13.1  10.1  2.78  11.2  10.5  23.4  7.53  18.5  12.7  3.11  2.82  7.82  4.09  7.09  5.83  Partial Eta  Squared  5.40E‐03  2.46E‐03  1.75E‐03  7.19E‐02  6.39E‐02  9.01E‐03  7.93E‐04  3.87E‐03  1.03E‐02  2.30E‐03  9.22E‐06  9.82E‐03  5.14E‐04  4.80E‐02  1.78E‐02  1.31E‐03  2.55E‐02  1.35E‐03  3.80E‐01  8.31E‐03  7.18E‐03  3.16E‐05  4.20E‐02  2.38E‐02  8.27E‐03  4.19E‐02  2.17E‐02  2.25E‐05  2.84E‐02  2.46E‐02  1.36E‐01  9.75E‐03  9.16E‐02  3.91E‐02  1.29E‐04  3.11E‐05  1.09E‐02  9.54E‐04  8.10E‐03  4.10E‐03  Parameter  Roberts  Rockwall  Runnels  Rusk  Sabine  San Augustine  San Jacinto  San Patricio  San Saba  Shackelford  Shelby  Sherman  Smith  Starr  Stephens  Stonewall  Swisher  Taylor  Throckmorton  Titus  Travis  Trinity  Tyler  Upshur  Uvalde  Val Verde  Van Zandt  Walker  Waller  Wharton  Wheeler  Wichita  Wilbarger  Willacy  Williamson  Wood  Young  Zapata    0.56  0.57  3.13  16.8  3.20  2.18  8.80  7.94  0.43  0.71  10.9  3.38  33.0  3.47  1.23  0.19  2.14  8.88  0.24  6.10  22.7  1.97  5.30  6.54  4.43  1.49  8.15  9.68  7.54  13.8  4.14  2.50  3.66  1.90  13.1  5.20  2.71  0.67  Standard  Error  t  Significance 1.30  1.29  1.29  1.33  1.29  1.29  1.29  1.29  1.29  1.30  1.30  1.29  1.29  1.30  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.29  1.30  1.32  1.31  1.29  1.29  1.29  1.29  1.30  1.31  0.434  0.442  2.42  12.6  2.48  1.68  6.81  6.14  0.332  0.552  8.44  2.62  25.6  2.67  0.952  0.147  1.66  6.88  0.188  4.72  17.6  1.53  4.10  5.06  3.43  1.16  6.31  7.49  5.84  10.6  3.14  1.90  2.83  1.47  10.2  4.02  2.09  0.511  6.64E‐01  6.58E‐01  1.57E‐02  1.68E‐34  1.34E‐02  9.31E‐02  1.38E‐11  1.03E‐09  7.40E‐01  5.81E‐01  7.34E‐17  8.90E‐03  3.69E‐120  7.64E‐03  3.41E‐01  8.83E‐01  9.73E‐02  9.00E‐12  8.51E‐01  2.56E‐06  3.72E‐63  1.27E‐01  4.38E‐05  4.70E‐07  6.27E‐04  2.48E‐01  3.71E‐10  1.12E‐13  6.48E‐09  1.42E‐25  1.75E‐03  5.72E‐02  4.77E‐03  1.41E‐01  1.48E‐23  6.04E‐05  3.65E‐02  6.09E‐01  81 95% Confidence  Interval  Lower  Upper  Bound  Bound  ‐1.98  ‐1.96  0.590  14.1  0.664  ‐0.364  6.27  5.40  ‐2.11  ‐1.83  8.40  0.850  30.5  0.923  ‐1.31  ‐2.35  ‐0.391  6.35  ‐2.29  3.56  20.2  ‐0.562  2.76  4.00  1.89  ‐1.04  5.61  7.15  5.01  11.3  1.55  ‐0.076  1.12  ‐0.632  10.6  2.66  0.171  ‐1.90  3.11  3.11  5.66  19.4  5.73  4.71  11.3  10.5  2.96  3.25  13.5  5.92  35.6  6.02  3.77  2.73  4.68  11.4  2.78  8.63  25.2  4.51  7.83  9.07  6.96  4.03  10.7  12.2  10.1  16.3  6.72  5.07  6.19  4.44  15.7  7.73  5.25  3.25  Partial Eta  Squared  1.24E‐04  1.29E‐04  3.83E‐03  9.39E‐02  4.01E‐03  1.85E‐03  2.96E‐02  2.42E‐02  7.23E‐05  2.00E‐04  4.47E‐02  4.49E‐03  3.00E‐01  4.67E‐03  5.95E‐04  1.42E‐05  1.80E‐03  3.01E‐02  2.31E‐05  1.44E‐02  1.69E‐01  1.53E‐03  1.09E‐02  1.65E‐02  7.66E‐03  8.77E‐04  2.55E‐02  3.56E‐02  2.19E‐02  6.93E‐02  6.42E‐03  2.37E‐03  5.22E‐03  1.42E‐03  6.36E‐02  1.05E‐02  2.87E‐03  1.72E‐04  Table 38. Summary Results of Statistical Analysis for Model 3. Parameter  Intercept      log(VMT)    ‐10.6  0.00428  0.00219  2.26  Standard Error  t Ratio  Prob> t   0.490  0.00031  0.00025  0.0863  ‐21.6  13.8  8.73  26.2  <0.0001  <0.0001  <0.0001  <0.0001  Parameter Estimates for Counties in the Barnett Shale Region  Cooke  Denton  Erath  Hill  Hood  Jack  Johnson  Montague  Palo Pinto  Parker  Somervell  Tarrant  Wise  ‐0.336  ‐0.045  0.120  ‐0.116  0.018  ‐0.178  ‐0.274  ‐0.053  ‐0.083  0.110  0.440  ‐1.368  0.169  0.201  0.204  0.202  0.205  0.202  0.202  0.228  0.202  0.202  0.207  0.203  0.231  0.208  ‐1.670  ‐0.220  0.590  ‐0.570  0.090  ‐0.880  ‐1.200  ‐0.260  ‐0.410  0.530  2.170  ‐5.920  0.810  0.095  0.826  0.553  0.571  0.930  0.378  0.231  0.795  0.680  0.596  0.030  <0.0001  0.417  Parameter Estimates for Counties in the Eagle Ford Shale Region  Atascosa  Bastrop  Bee  Brazos  Burleson  Caldwell  DeWitt  Dimmit  Duval  Fayette  Frio  Goliad  Gonzales  Grimes  Guadalupe  Jim Wells  Karnes  Lampasas  Lavaca  Lee  Live Oak  ‐0.216  0.059  ‐0.452  ‐0.388  ‐0.160  ‐0.269  ‐0.444  ‐0.357  ‐0.087  ‐0.168  ‐0.617  0.094  ‐0.290  0.432  ‐0.041  ‐0.233  0.046  ‐0.465  ‐0.035  0.177  ‐0.409  0.204  0.205  0.201  0.202  0.201  0.201  0.202  0.207  0.201  0.204  0.202  0.202  0.204  0.202  0.204  0.203  0.205  0.201  0.201  0.201  0.203  82 ‐1.060  0.290  ‐2.250  ‐1.920  ‐0.790  ‐1.340  ‐2.190  ‐1.730  ‐0.430  ‐0.820  ‐3.050  0.470  ‐1.420  2.140  ‐0.200  ‐1.150  0.220  ‐2.310  ‐0.170  0.880  ‐2.010  0.288  0.773  0.025  0.055  0.427  0.182  0.028  0.084  0.665  0.411  0.002  0.641  0.155  0.032  0.839  0.251  0.824  0.021  0.862  0.380  0.045  Parameter  Maverick  McMullen  Robertson  Victoria  Washington  Webb  Wilson  Zavala    ‐1.213  0.761  0.773  ‐0.071  ‐0.019  ‐0.015  ‐0.617  0.231  Standard Error  t Ratio  Prob> t   0.214  0.204  0.202  0.204  0.202  0.207  0.202  0.202  ‐5.670  3.730  3.840  ‐0.350  ‐0.090  ‐0.070  ‐3.060  1.140  <0.0001  0.000  0.000  0.728  0.927  0.941  0.002  0.252  Parameter Estimates for Counties in the Permian Basin Region  Andrews  Borden  Cochran  Coke  Crane  Crockett  Crosby  Dawson  Dickens  Ector  Gaines  Garza  Glasscock  Hockley  Howard  Irion  Kent  Loving  Lubbock  Lynn  Martin  Midland  Mitchell  Pecos  Reagan  Reeves  Schleicher  Scurry  Sterling  Sutton  Terrell  Terry  Tom Green  ‐0.679  0.447  0.401  0.195  ‐0.191  0.029  ‐0.305  ‐0.726  0.384  0.311  ‐0.689  ‐0.119  0.113  0.048  0.232  0.209  0.662  1.576  ‐0.248  ‐0.332  0.369  0.520  ‐0.049  ‐0.126  0.284  0.003  0.086  0.342  ‐0.223  ‐0.139  0.498  ‐0.355  0.079  0.239  0.212  0.208  0.204  0.203  0.203  0.204  0.201  0.208  0.216  0.203  0.201  0.210  0.201  0.203  0.206  0.213  0.223  0.203  0.201  0.202  0.212  0.203  0.203  0.210  0.202  0.205  0.201  0.203  0.201  0.208  0.201  0.202  83 ‐2.840  2.110  1.930  0.960  ‐0.940  0.140  ‐1.500  ‐3.610  1.850  1.440  ‐3.390  ‐0.590  0.540  0.240  1.140  1.010  3.100  7.060  ‐1.230  ‐1.650  1.820  2.450  ‐0.240  ‐0.620  1.350  0.010  0.420  1.700  ‐1.090  ‐0.690  2.390  ‐1.760  0.390  0.005  0.035  0.054  0.337  0.349  0.888  0.135  0.000  0.065  0.149  0.001  0.556  0.592  0.810  0.253  0.311  0.002  <0.0001  0.220  0.099  0.068  0.014  0.811  0.537  0.177  0.989  0.674  0.089  0.274  0.489  0.017  0.078  0.696  Parameter  Upton  Ward  Winkler  Yoakum    ‐0.123  ‐0.374  0.083  0.044  Standard Error  t Ratio  Prob> t   0.219  0.204  0.204  0.204  ‐0.560  ‐1.830  0.410  0.210  0.575  0.068  0.684  0.830  Parameter Estimates for the Remaining 175 Counties  Anderson  Angelina  Aransas  Archer  Armstrong  Austin  Bailey  Bandera  Baylor  Bell  Bexar  Blanco  Bosque  Bowie  Brazoria  Brewster  Briscoe  Brooks  Brown  Burnet  Calhoun  Callahan  Cameron  Camp  Carson  Cass  Castro  Chambers  Cherokee  Childress  Clay  Coleman  Collin  Collingsworth  Colorado  Comal  Comanche  0.197  0.131  ‐0.157  ‐0.271  ‐0.112  ‐0.197  0.430  ‐0.446  0.454  ‐0.387  0.650  ‐0.599  ‐0.620  0.008  ‐0.038  ‐0.200  0.667  ‐0.202  ‐0.560  ‐0.421  ‐0.458  ‐0.037  ‐1.000  0.101  ‐0.139  0.625  0.372  0.836  0.114  ‐0.561  ‐0.353  ‐0.449  0.051  0.331  0.148  ‐0.609  ‐0.117  0.202  0.202  0.202  0.202  0.202  0.202  0.203  0.201  0.204  0.206  0.209  0.201  0.201  0.205  0.207  0.203  0.213  0.201  0.201  0.202  0.202  0.202  0.204  0.203  0.202  0.201  0.202  0.207  0.202  0.201  0.201  0.202  0.201  0.208  0.204  0.204  0.201  84 0.970  0.650  ‐0.780  ‐1.340  ‐0.560  ‐0.980  2.110  ‐2.210  2.230  ‐1.880  3.110  ‐2.980  ‐3.080  0.040  ‐0.190  ‐0.990  3.130  ‐1.000  ‐2.780  ‐2.090  ‐2.270  ‐0.190  ‐4.910  0.500  ‐0.690  3.110  1.840  4.040  0.560  ‐2.790  ‐1.750  ‐2.230  0.250  1.590  0.720  ‐2.990  ‐0.580  0.330  0.516  0.436  0.180  0.577  0.329  0.035  0.027  0.026  0.060  0.002  0.003  0.002  0.969  0.853  0.325  0.002  0.316  0.005  0.037  0.023  0.853  <0.0001  0.619  0.490  0.002  0.066  <0.0001  0.574  0.005  0.080  0.026  0.799  0.112  0.470  0.003  0.562  Parameter  Concho  Coryell  Cottle  Culberson  Dallam  Dallas  Deaf Smith  Delta  Donley  Eastland  Edwards  El Paso  Ellis  Falls  Fannin  Fisher  Floyd  Foard  Fort Bend  Franklin  Freestone  Galveston  Gillespie  Gray  Grayson  Gregg  Hale  Hall  Hamilton  Hansford  Hardeman  Hardin  Harris  Harrison  Hartley  Haskell  Hays  Hemphill  Henderson  Hidalgo  Hopkins  Houston    ‐0.237  ‐0.096  0.423  ‐0.368  0.348  0.566  0.404  ‐0.089  0.053  0.244  0.191  ‐0.461  ‐0.222  0.116  ‐0.276  0.061  ‐0.220  0.463  ‐0.109  ‐0.223  ‐0.072  ‐0.600  ‐0.419  0.160  ‐0.461  1.080  ‐0.483  0.065  ‐0.394  0.417  ‐0.272  0.073  4.118  0.306  0.604  ‐0.073  ‐0.213  0.511  ‐0.273  ‐0.140  ‐0.045  0.216  Standard Error  t Ratio  Prob> t   0.203  0.201  0.210  0.201  0.202  0.206  0.201  0.204  0.201  0.202  0.208  0.205  0.204  0.201  0.201  0.205  0.204  0.213  0.208  0.201  0.205  0.202  0.201  0.201  0.202  0.203  0.201  0.203  0.202  0.206  0.202  0.202  0.231  0.207  0.202  0.203  0.203  0.204  0.202  0.208  0.203  0.201  ‐1.170  ‐0.470  2.010  ‐1.830  1.720  2.750  2.000  ‐0.440  0.260  1.210  0.920  ‐2.250  ‐1.090  0.580  ‐1.370  0.300  ‐1.080  2.180  ‐0.520  ‐1.110  ‐0.350  ‐2.970  ‐2.080  0.800  ‐2.280  5.320  ‐2.400  0.320  ‐1.950  2.030  ‐1.350  0.360  17.810  1.480  3.000  ‐0.360  ‐1.050  2.500  ‐1.350  ‐0.670  ‐0.220  1.080  0.243  0.635  0.044  0.068  0.085  0.006  0.045  0.663  0.794  0.227  0.360  0.025  0.276  0.564  0.171  0.765  0.281  0.030  0.600  0.267  0.726  0.003  0.038  0.427  0.023  <0.0001  0.017  0.749  0.051  0.043  0.178  0.717  <0.0001  0.140  0.003  0.720  0.293  0.012  0.177  0.503  0.826  0.282  85 Parameter  Hudspeth  Hunt  Hutchinson  Jackson  Jasper  Jeff Davis  Jefferson  Jim Hogg  Jones  Kaufman  Kendall  Kenedy  Kerr  Kimble  King  Kinney  Kleberg  Knox  La Salle  Lamar  Lamb  Leon  Liberty  Limestone  Lipscomb  Llano  Madison  Marion  Mason  Matagorda  McCulloch  McLennan  Medina  Menard  Milam  Mills  Montgomery  Moore  Morris  Motley  Nacogdoches  Navarro    0.417  ‐0.136  ‐0.102  ‐0.536  ‐0.322  ‐0.021  0.016  ‐0.202  ‐0.002  ‐0.553  ‐0.375  ‐0.783  ‐0.114  ‐0.279  0.648  ‐0.084  ‐0.574  0.167  ‐0.424  ‐0.388  ‐0.080  0.487  0.484  ‐0.117  0.298  ‐0.155  ‐0.308  ‐0.239  ‐0.351  ‐0.245  ‐0.384  ‐0.098  ‐0.243  0.454  0.144  ‐0.152  1.558  0.257  0.050  0.526  0.172  ‐0.234  Standard Error  t Ratio  Prob> t   0.203  0.205  0.202  0.202  0.203  0.203  0.205  0.203  0.201  0.206  0.202  0.201  0.202  0.201  0.209  0.204  0.201  0.206  0.205  0.201  0.201  0.204  0.203  0.201  0.208  0.201  0.202  0.202  0.204  0.201  0.202  0.206  0.202  0.205  0.202  0.202  0.215  0.201  0.201  0.212  0.203  0.203  2.050  ‐0.660  ‐0.510  ‐2.660  ‐1.590  ‐0.100  0.080  ‐1.000  ‐0.010  ‐2.690  ‐1.860  ‐3.890  ‐0.570  ‐1.390  3.090  ‐0.420  ‐2.850  0.810  ‐2.060  ‐1.930  ‐0.400  2.390  2.380  ‐0.580  1.430  ‐0.770  ‐1.520  ‐1.180  ‐1.720  ‐1.220  ‐1.900  ‐0.480  ‐1.200  2.210  0.710  ‐0.750  7.260  1.280  0.250  2.480  0.850  ‐1.160  0.040  0.508  0.613  0.008  0.112  0.917  0.938  0.319  0.994  0.007  0.063  0.000  0.572  0.166  0.002  0.678  0.004  0.418  0.039  0.054  0.691  0.017  0.017  0.561  0.153  0.440  0.128  0.237  0.085  0.224  0.058  0.633  0.229  0.027  0.477  0.454  <0.0001  0.201  0.803  0.013  0.396  0.248  86 Parameter  Newton  Nolan  Nueces  Ochiltree  Oldham  Orange  Panola  Parmer  Polk  Potter  Presidio  Rains  Randall  Real  Red River  Refugio  Roberts  Rockwall  Runnels  Rusk  Sabine  San Augustine  San Jacinto  San Patricio  San Saba  Shackelford  Shelby  Sherman  Smith  Starr  Stephens  Stonewall  Swisher  Taylor  Throckmorton  Titus  Travis  Trinity  Tyler  Upshur  Uvalde  Val Verde    0.161  0.416  ‐0.208  ‐0.045  0.430  0.066  1.108  0.294  0.753  0.229  ‐0.114  ‐0.518  ‐0.308  0.494  0.117  ‐0.521  0.628  ‐0.519  ‐0.013  0.702  0.136  0.078  0.272  ‐0.330  ‐0.119  0.000  0.668  0.531  1.433  ‐0.617  ‐0.019  0.460  ‐0.222  0.322  0.518  ‐0.262  0.820  ‐0.319  0.110  ‐0.177  ‐0.298  ‐0.248  Standard Error  t Ratio  Prob> t   0.201  0.201  0.203  0.203  0.201  0.202  0.205  0.201  0.204  0.203  0.205  0.202  0.202  0.207  0.201  0.202  0.208  0.203  0.202  0.205  0.202  0.202  0.201  0.204  0.205  0.204  0.201  0.203  0.210  0.202  0.203  0.209  0.201  0.202  0.212  0.202  0.209  0.202  0.201  0.202  0.201  0.202  0.800  2.060  ‐1.030  ‐0.220  2.130  0.330  5.410  1.460  3.700  1.130  ‐0.560  ‐2.560  ‐1.530  2.390  0.580  ‐2.580  3.020  ‐2.560  ‐0.070  3.430  0.680  0.390  1.350  ‐1.620  ‐0.580  0.000  3.320  2.620  6.830  ‐3.060  ‐0.090  2.200  ‐1.100  1.600  2.450  ‐1.300  3.920  ‐1.580  0.550  ‐0.880  ‐1.480  ‐1.230  0.424  0.039  0.304  0.824  0.033  0.745  <0.0001  0.145  0.000  0.259  0.578  0.011  0.127  0.017  0.563  0.010  0.003  0.011  0.947  0.001  0.499  0.699  0.177  0.106  0.564  1.000  0.001  0.009  <0.0001  0.002  0.925  0.028  0.270  0.111  0.015  0.195  <0.0001  0.114  0.583  0.381  0.139  0.220  87 Parameter  Van Zandt  Walker  Waller  Wharton  Wheeler  Wichita  Wilbarger  Willacy  Williamson  Wood  Young  Zapata    ‐0.423  ‐0.271  ‐0.351  0.341  ‐0.136  ‐0.543  ‐0.231  ‐0.388  0.134  ‐0.210  0.043  ‐0.534  Standard Error  t Ratio  Prob> t   0.205  0.204  0.204  0.203  0.202  0.202  0.201  0.201  0.206  0.202  0.202  0.202  ‐2.060  ‐1.320  ‐1.720  1.680  ‐0.670  ‐2.690  ‐1.150  ‐1.930  0.650  ‐1.040  0.220  ‐2.640  0.040  0.186  0.085  0.093  0.500  0.007  0.250  0.054  0.517  0.297  0.829  0.008  88