- ?ff?5+ Hcme Help 1' Ftescurces 1' Ecntact Us regulahonsgov Vaur Hence in Federal Decisicn-Malung Advanced Search Ccmment submitted by I:Iiirla E. Cellier, Benesch Friedlander Ceplan 8 Areneff LLP en behalf of Murray Energy Ccrperatien The is a an the Enyircnmental Frctecticn Agency Ftule: Hepeal cf Earbcn Fellutlcn Emisaicn Guidelines Existing Staticnary Ecurces: Electric Utility Generating Units; Flecpenlng cf the Ecmment Ferlcd Inci'r- pass-1 9 see Fer related Clpen Dccitet Fclder El ?umber; ij-eet ?Share Email Cemment Ddeument Infermaticn Please see attached Cernmants ei Murray Energy C?l?p?l?ali?n. Appendix A through Appendix will he sent under separate ceyer. Thahlt ydu. ?Hts Fastest Apr??. acts sin; Net Assigned Attachments Shaw Mcre Details at? |[Zrzimment Attachment: BEFORE THE UNITED STATES ENVIRONMENTAL PROTECTION AGENCY REPEAL OF CARBON POLLUTION EMISSION GillDELINES FOR EXISTING STATIONARY SOURCES: ELECTRIC UTILITY GENERATING UNITS ) ) ) ) DOCKET ID NO. EPA-HQ-OAR-2017-0355 COMMENTS OF MURRAY ENERGY CORPORATION I. INTRODUCTION Murray Energy Corporation ("Murray Energy") enthusiastically applauds President Donald J. Trump's Energy Independence Executive Order 13783, issued March 28, 2017 ("Executive Order"), which directed the United States Environmental Protection Agency (the "U.S. EPA") to "suspend, revise, or rescind" the Obama Administration's so-called and illegal Clean Power Plan ("CPP") and, further, affirms the "national interest to promote clean and safe development of our Nation's vast energy resources, while at the same time avoiding regulatory burdens that unnecessarily encumber energy production, constrain economic growth, and prevent job creation". Executive Order 13783, Section 1(a). Specifically, the Executive Order directs all executive department~ and agencies, including the U.S. EPA, to "immediately review existing regulations that potentially burden the development or use of domestically produced energy resources and appropriately suspend, revise, or rescind those that unduly burden the development of domestic energy resources beyond the degree necessary to protect the public interest or otherwise comply with the law." /d. Section 1(c). Indeed, the Executive Order specifically directs the U.S. EPA to review and initiate reconsideration proceedings to "suspend, revise, or rescind" the Obama CPP, "as appropriate and consistent with law." /d. Section 4(a)-(c). We strongly support the full and complete repeal of the Obama CPP. 11093257 v2 We have been examining these issues for many years, as Murray Energy was the very first party to file a lawsuit challenging the Obama CPP in the case before the federal D.C. Circuit styled Murray Energy v. U.S. Environmental Protection Agency, D.C. Circuit Case No. 14-1112. After two (2) years of expensive litigation, Murray Energy was joined by twenty-nine (29) states. The cases were consolidated in the case styled West Virginia v. U.S. Environmental Protection Agency, D.C. Circuit Case No. 15-1363. On February 9, 2016, the U.S. Supreme Court stayed implementation of the CPP, pending further judicial review. This was the first time in American history that the U.S. Supreme Court has intervened to stay, or temporarily block, an agency's regulation before a lower court heard legal challenges to it. Accordingly, we were extremely pleased that President Trump issued Executive Order 13783, as it saved us many years of litigation, in which we would have ultimately prevailed, and it helped to protect the jobs and family livelihoods of our coal miners, and low-cost, reliable electricity for all Americans. On October 16,2017, the U.S. EPA, pursuant to Executive Order 13783, issued its proposed action (the "CPP Repeal Rule") to repeal the Obama CPP and rescind the documents in the CPP docket titled "Legal Memorandum For Proposed Carbon Pollution Emissions Guidelines for Existing Electric Utility Generating Units" (in the docket for the proposed rule) and "Legal Memorandum for Accompanying Clean Power Plan for Certain Issues," (a supplementary document in the docket for the final rule) (collectively, the "Legal Memorandum"). See Docket ld No. EPA-HQ-OAR-2017-0355; 82 Fed. Reg. 48035. In this action, the U.S. EPA noted that the EPA promulgated the CPP under section 111 of the Clean Air Act (the "CAA''), 42 U.S.C. 7411. The U.S. EPA asserted that section 111(b) authorizes the EPA to issue nationally applicable new source performance standards limiting air pollution from "new sources" in source categories that cause or contribute to air pollution that 2 11093257 v2 may reasonably be anticipated to endanger public health or welfare. In 2015, U.S. EPA issued such a rule for COz emissions from certain new fossil fuel-fired power plants in light of the U.S. EPA's assessment "that [greenhouse gases] endanger public health, now and in the future". See Standards of Performance for Greenhouse Gas Emissions from New, Modified, and Reconstructed Stationary Sources: Electric Generating Units, 80 Fed. Reg. 64510,64518 (October 23, 2015; see also Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean Air Act, 74 Fed. Reg. 66496 (December 15, 2009) (the "2009 Endangerment Finding"). 82 Fed. Reg. at 48037. The U.S. EPA, in this action, does not base its proposed repeal of the CPP on rescission of the 2009 Endangerment Finding but on consideration of the statutory text, context and legislative history of CAA section 111. 82 Fed. Reg. at 48043. Murray Energy enthusiastically supports the U.S. EPA's proposed actions. The U.S. EPA's proposed actions are correct for several fundamental reasons. First, the CAA does not authorize the CPP's wholesale transformation of the U.S. electricity grid. Utility Air Regulatory Group v. EPA, 134 S. Ct. 2427 (2014). Thus, the agency's proposal is in line with President Donald J. Trump's directives in Executive Order 13783 to review regulations that burden the development of domestic resources such as coal and ensure the environmental regulations comply with law. See Executive Order 13783 at Section 2. Second, the CPP is categorically foreclosed by the CAA's exclusion for regulating facilities under section 111(d) that are already regulated under section 112. Third, the CAA is a program of cooperative federalism, which expressly provides the States-not the U.S. EPA-with the right under section 111(d) to "establish" and "apply" performance standards to existing power plants and to "take into consideration, among other factors, the remaining useful life of the existing source to which [a] 3 11093257 v2 standard [of performance] applies." 42 U.S.C. § 741l(d)(1). The U.S. Constitution preserves the sovereignty of the States by barring the federal government from compelling them to implement federal policies. The CPP violates this sovereignty by mandating that the States implement U.S. EPA's decarbonization of the U.S. power system and violates the exclusive jurisdiction of the Federal Energy Regulatory Commission (PERC) under the Federal Power Act. Finally, the CAA only provides the U.S. EPA with authority to regulate under section 111 after the agency has made two findings-section 111 's "endangerment" and "significant contribution" findings. But the U.S. EPA has not, and cannot, make such findings to support the CPP as a section 111(d) regulation. For the same reasons, as well as the fact that they do not accurately reflect the Agency's current legal positions, the Legal Memoranda should be repealed, not just to the extent they conflict, but in their entirety. These memoranda are inaccurate and do not reflect a proper interpretation of the issues that they describe. There is also no way to clearly distinguish those issues that would remain if they are repealed only to the extent they contradict the Agency's Proposed Rule. Finally, EPA requests comment on the impact its Proposed Rule will have in several key policy areas, including whether further problems may exist with the "speak clearly" doctrine, PERC jurisdiction, and traditional state authority to regulate electric power, if EPA adopts the Proposed Rule. Murray Energy supports EPA's focus on preserving the traditional role of the states and PERC in regulating the electric power sector and in not overstepping its jurisdictional bounds under either the Clean Air Act or the Federal Power Act. While the Proposed Rule takes significant steps toward alleviating specific identified violations of Constitutional, statutory, and regulatory policy, however, adopting the Proposed Rule does not cure all defects with the Clean 4 11093257 v2 Power Plan and would not, on its own, resolve all conflicts a replacement rule would face with these same issues. 1 II. MURRAY ENERGY'S VITAL INTEREST IN THIS ACTION For more than eight (8) years, Murray Energy has warned of the regulatory rampage that was being illegally waged by the Obama Administration and its supporters against the U.S. coal industry, and particularly of the destruction wrought by the Clean Power Plan. In President Trump, we finally have a President who has vowed to preserve coal jobs and low-cost, reliable and fuel-secure electricity for all Americans, including retirees on fixed incomes, single mothers, and manufacturers who depend on low cost electricity to produce their products. This is why the American people elected President Trump. By issuing Executive Order 13783, President Trump has followed through on his promises, and, by repealing the Obama Clean Power Plan, the U.S. EPA, under the direction of Administrator Pruitt, will take one more step towards fulfilling the promises made to the American people. Murray Energy has a vital interest in the repeal of the Obama Clean Power Plan. Murray Energy was established in 1988 when Mr. Robert E. Murray mortgaged virtually everything he owned and purchased a single coal mine in Southern Ohio. Thirty years later, Murray Energy is the largest underground coal mining company in the U.S. Moreover, Murray Energy is the largest employer of coal workers in the U.S. in the underground mining industry, with over 5,600 employees. Murray Energy and its subsidiary companies currently operate fifteen (15) coal 1 On December 28,2017, the U.S. EPA issued its Advance Notice of Proposed Rulemaking (ANPRM) seeking public comment in its consideration of proposed emission guidelines to limit greenhouse gas (GHG) emissions from existing electric generating units (EGUs) (the proposed "CPP Replacement Rule"). Docket ld. No. EPA-HQ-OAR-20 17 -0545; 82 Fed. Reg. 61507. Murray Energy has filed comments in Docket ld. No. EPA-HQ-OAR-2017-0545 opposing any consideration of a new CPP rule absent a clear Congressional grant of authority and based in any degree, on the illegal and technically flawed 2009 Endangerment Finding. 5 11093257 v2 mines, consisting of eleven (11) underground longwall mining systems and forty-six (46) continuous mining units in Ohio, lllinois, Kentucky, Utah, and West Virginia. In addition, Murray Energy operates two surface mines in Colombia. Murray Energy produces approximately 75 million tons of bituminous coal each year. It supplies coal to many of the largest coal-fired electric utility generating facilities in the United States. Murray Energy is also engaged in related business operations and activities, including owning and operating four mining equipment manufacturing and rebuild facilities along with a number of river, truck and rail terminals, and twenty-five river towboats and over 500 coal barges on the inland waterway system. Many of Murray Energy's mining complexes are strategically located near its customers' electric generating stations, and all have excellent, low cost transportation infrastructures to Murray Energy's markets. The vast majority of the coal produced from Murray Energy's mines in the U.S. is used for the generation of electricity. Murray Energy is dependent on the continuing viability and operation of coal-fired generation in the U.S .. Unless fully repealed, the Obama Clean Power Plan would force many of these coalfired power plants to close, destroying the jobs and livelihoods for thousands of coal mining families and related industries. Given the current threats to coal-fired generation, Murray Energy, along with other coal producers and related industries, and numerous generating companies and electric utilities, is threatened with bankruptcy and significant economic harm if coal capacity is forced out of the market by unreasonable and unsupportable regulations. Prior to the election of President Obama, fifty-two percent (52%) of America's electricity was generated from coal, and this rate was much higher in the Midwest. The percentage of coal-fired generation declined under the Obama Administration to thirty percent (30% ). Under the Obama Administration, over 531 coal-fired 6 11093257 v2 generating plants, or 59,000 megawatts of generating capacity through 2016, were closed prematurely, many as a result of new and potential regulations that were put into place illegally, without proper cost analysis, and without proven environmental benefits. Further, an additional 12,700 megawatts of coal fired-generation will be closed by the end of 2020, bringing coal's share of electricity to as low as twenty-seven percent (27% ). These closures are the functional equivalent of entirely eliminating the combined electricity supplies of Ohio, Pennsylvania, Indiana, and West Virginia. In the PJM Interconnection, LLC ("PJM") footprint alone, which covers all or part of thirteen (13) states and sixty-five (65) million people, 11,000 megawatts of coal-fired electricity generation has been closed over the past four (4) years. In addition, 20,056 megawatts of this baseload capacity in PJM is contemplated for closure. This devastation has had far-reaching consequences. By early 2016, the total value of the American coal industry had declined from $68.8 billion five years before to $4.08 billion, a ninety-four percent (94%) reduction in value. A total of fifty-two (52) coal companies were in bankruptcy proceedings with only four (4) major companies remaining financially solvent. Local rural communities in coal producing regions, and in areas that depend on coal-fired power plants, are losing jobs and millions of dollars in local tax support due to the closure of coal-fired generation plants. This devastates the residents and the employees supporting local businesses, governments, and school districts. Now is the time to end this catastrophic destruction wrought by the Obama Administration's "War On Coal". Fortunately, President Trump's Executive Order of March 31,2017 is already having a positive impact in coal communities. Indeed, this action alone stopped fifty-six (56) more coalfired plant closures, totaling 53,000 megawatts of generation, and the layoff of 25,000 more coal miners on top of the 63,000 families already laid off under the Obama Administration. Far more 7 11093257 v2 must be done, however, for our communities to recover from the devastation of the past eight (8) years. III. CONTINUED OPERATION OF AMERICA'S COAL-FIRED ELECTRICITY GENERATION FLEET IS ABSOLUTELY VITAL TO ENSURING RELIABLE, EFFICIENT AND COST-EFFECTIVE SUPPLIES OF ELECTRICITY TO THE NATION. President Trump's Executive Order 13783 is far reaching, directing all executive departments and agencies to "immediately review existing regulations that potentially burden the development or use of domestically produced energy resources and appropriately suspend, revise, or rescind those that unduly burden the development of domestic energy resources beyond the degree necessary to protect the public interest or otherwise comply with the law." Executive Order 13783, Section l(c). America enjoys an abundant resource of proven coal reserves. Coal is a critical component of America's energy resources and continued operation of America's coal-fired electricity generation fleet is absolutely vital to ensuring reliable, efficient and cost-effective supplies of electricity to the nation. For over a century, coal-fired generation has been the safe, reliable, low-cost, and fuelsecure source of electricity in America, providing the baseload generation, as well as the capacity, reserve, and ancillary services that are absolutely necessary to maintain the integrity and reliability of our Nation's power grids. The historical fleet of coal-fired generating units, particularly in the Midwest, has served the economy well, providing as much as eighty to ninety percent (80 - 90%) of in-state generation in many states over the years. Coal-fired generation has also served the commercial, manufacturing and industrial sectors of this Country, providing 8 11093257 v2 low-cost, reliable, high capacity and peak demand services that are absolutely necessary for American manufacturers to operate and to compete in the global marketplace. Over the years, coal-fired generation has been less susceptible than other sources to both short-term and long-term fuel price variation and supply. Coal-fired generation has been the constant through the years of the Arab oil embargo, the natural gas shortages of the 1970's and 1980's, the ensuing volatility in natural gas prices thereafter, nuclear power regulatory challenges, and extreme weather conditions, most recently the 2014 Polar Vortex and 2018 Bomb Cyclones. There is no better illustration of the need to protect baseload generation than the so-called "Bomb Cyclone," which immersed the eastern United States in extremely cold, windy conditions from December 27,2017 through January 8, 2018. Notwithstanding that this cold snap occurred primarily over the holidays, at least two (2) million Americans lost their power, and, tragically, twenty-two (22) people lost their lives. Without the electricity provided by our coal-fired and nuclear power plants, the devastation of this very short twelve ( 12) day Bomb Cyclone would have been far worse. The United States Department of Energy's National Energy Technology Laboratory recently issued a report ("Government Study") analyzing the reliability and resiliency of different sources of electricity generation during the Bomb Cyclone. The Government Study confirmed what many of us have already known, that coal was the single most reliable and resilient form of electricity production during that critical time. Coal and nuclear power provided eighty-nine percent (89%) of the electricity during this Bomb Cyclone. During this time coal-fired generation averaged an output level of 46,038 megawatts, over fifty percent (50%) greater than the average of 29,849 megawatts. Indeed, if it were not for the electricity generated by our Nation's coal-fired 9 11093257 v2 power plants, with ample capacity and on-site fuel availability, the grids would have experienced a massive nine (9) to eighteen (18) gigawatts of shortfall, leading to system collapse. During this cold snap, coal far outperformed all other fuel sources, particularly natural gas and renewables. At least 37,000 megawatts of supposedly available natural gas-powered electricity was entirely unavailable due to the priority for home heating use and frozen natural gas pipelines. Where natural gas was available, prices peaked at over $95 per million BTU in the PJM, and over $175 per million BTU in New York City, which is sixty (60) times the normal cost. Also, during this time, the cost of electric power from natural gas-fired plants peaked at over $500 per megawatt hour, compared to a predominant rate of about $28 per megawatt hour. The ISO New England regional transmission organization has confirmed that their region is at major risk of fuel insecurity, due to New England's dependence on natural gas and the retirement of coal and nuclear generating capacity. Similarly, windmills and solar panels contributed virtually nothing to our Country's electricity needs at that dire time, as cloud cover and wind speeds caused these resources to be unable to dispatch. The Government Study concluded that wind energy was down twelve-percent (12%) across the eastern United States. When considered together, wind and solar electricity generation declined nineteen percent (19%) in Midcontinent Independent System Operator ("MISO"), twenty-nine percent (29%) in Southwest Power Pool ("SPP") and thirty-two percent (32%) in Electric Reliability Council of Texas ("ERCOT"). Fortunately, coal-fired electricity was able to step up and to fill the void for seventy-four percent (74%) of this incremental lost generation. lO 11093257 v2 The Government Study valued the resilience provided by coal at $3.5 billion in the PJM alone, which equates to $288 million per day. PJM's President and CEO, Mr. Andrew Ott, recently stated that 1,410 megawatts of nuclear capacity and 3,688 megawatts of coal-fired capacity that operated during the recent cold snap in the eastern United States are scheduled to be deactivated within the next five (5) years. These problems from the recent cold snap were not an isolated incident. During the socalled "Polar Vortex" of early 2014, PJM came within 500 megawatts of a major system disruption on a demand of 140,000 megawatts. A total of 9,300 megawatts of supposedly available natural gas-fired generation was not available due to gas supply disruptions to the generators. Further, the cost of producing electricity in the Midwest and Mid-Atlantic area rose above $1,000 per megawatt-hour for the first time in American history. During this time, an Ohio-based electric power company was ordered by the State's Public Utility Commission to be connected to 3,800 megawatts of wind and solar power. Only fifteen (15) megawatts of the 3,800 megawatts were available during the crisis. What the utility relied on during the cold snap was 8,170 megawatts of coal-fired generation. As a118,170 megawatts have been closed, what will happen next time? The recent Bomb Cyclone and 2014 Polar Vortex demonstrate that our electric power grids are not as reliable as the independent power grid operators, some electric utilities, and the Federal Energy Regulatory Commission ("FERC") claim. Indeed, we have a power grid reliability and resiliency crisis in much of America. But, will a system collapse occur before they recognize and do something about it? 11 11093257 v2 During the 2018 Bomb Cyclones, the consequence of lack of fuel diversity was seen in New England ISO (NE ISO) pricing. Comparing the first half of January 2018 to the first half of January 2017, natural gas prices (Algonquin hub) were up from an average of $5.60 per MMBtu in 2017 to $22.78 per MMBtu in 2018, a 307 percent increase. Power prices (Mass Hub) were up from an average of $41.80 per megawatt-hour to $147.74 per megawatt-hour, a 253 percent increase. Also relevant was the over 7000 percent increase in use of oil for power generation as a result of supply constraints on natural gas due to the lack of storage and pipeline capacity. Dual fuel gas and oil plants had to switch to oil to meet load. Pricing was also up in PJM West, which had an average energy price of $119.53 per megawatt-hour in the first half of January 2018. The average energy price and price increases were higher in NE ISO than PJM West because the coal generation in PJM increased by about 10 percent in the first half of January 2018 which significantly reduced the increased generation required from oil. There is no question that had it not been for the coal capacity in PJM, MISO and elsewhere the power prices would have been significantly higher. Renewable energy sources are not a viable or credible alternative to baseload coal-fired generation. Wind and solar generation sources are intermittent and unreliable and therefore cannot be relied upon to meet peak or base load demand. Without the price support provided by the Wind Production Tax Credit, wind generation will be a high cost resource. Natural gas-fired generation is not the answer either, as gas pricing is volatile and gas supply is unreliable given limited gas storage capacity, pipeline limitations and a requirement to meet residential and commercial customer requirements ahead of power generation. The high power prices during the 2018 Bomb Cyclones in certain regions were due to gas to oil switching in many ofthe dual fuel units due to insufficient gas delivery capability. 12 11093257 v2 The Obama CPP drastically distorts the electricity grid by forcing power plants effectively to dispatch based upon carbon emissions rather than cost. As a result, lower cost coal generation will be impaired at the expense of higher cost generation from natural gas and renewables. The cost impacts will be magnified to the extent low cost coal plants are forced to retire, leaving only high cost generation available to meet demand. A recent study performed by the leading global economic consulting firm, lliS-Markit concludes that, on a going forward basis (excluding sunk costs), the costs of continuing to operate many recently-retired coal-fired plants is significantly lower than the long-term marginal cost of building new generation. 2 In some instances, on a properly-calculated apples-to-apples basis, the cost of electricity generated by a newly-constructed power plant may be approximately twice that of a baseload coal or nuclear plant that has recently retired. 3 The fact that utilities would be required to close coal power plants that generate electricity much more cost effectively than alternative new generation to meet the CPP requirements is a fatal flaw of the Obama CPP. Furthermore, baseload coal and nuclear plants typically operate at high capacity factors, have stable operating costs in part because fuel can be purchased under long-term contracts with fixed pricing. As such, coal plants are valuable assets which limit exposure to price spikes, keep electricity costs at reasonable levels and historically have been the backbone of the operation of the grid. From an economic standpoint, it seldom should make sense to shut down these generating units, especially since, once shut down, these generating units are permanently lost. Yet that is precisely what is occurring today. 4 2 IHS Markit, Ensuring Resilient and Efficient Electricity Generation: The Value of the current diverse US power supply portfolio, at p. 8 (Sept. 20 17) (hereinafter, "IHS Study"). 3 IHS Study at 36. 4 Many of the companies that historically have been leaders in electric generation, such as AEP, Duke, NRG and Calpine, have announced that, except for generating units supported by long-term Purchase Power Agreements, they will no longer build new merchant generation and, in several instances, are liquidating their entire merchant generation portfolio. This is reducing the number of experienced players interested in continuing to own and operate generation. 13 11093257 v2 A related problem that will worsen with further retirements of baseload coal and nuclear will be the increased frequency, severity, and duration of price spikes that will arise with increased dependence upon natural gas. In particular, during the past several years the ability of grid operators to shift back and forth between natural gas-fired generation and coal-fired generation has played an increasingly critical role in managing price volatility. When gas prices rise, coal generation increases; when gas prices fall, coal generation declines. With additional coal plant retirements, however, the ability to reduce gas use by increasing use of coal-fired capacity declines, reducing the amount of available fuel switching by a startling 11 BCF/day in the past six years. 5 As a result, natural gas price increases are expected as coal generation is not available to cap gas demand and price. Further, the reduced potential for fuel switching is not the only change that is occurring that could cause adverse volatility and price spikes. LNG exports from the U.S. began in earnest in 2016 with the completion of the Sabine Pass facility which reached 2 billion cubic feet per day (BCFD) by year end. Another six plus BCFD of LNG capacity is under construction and 13.5 BCFD of LNG capacity is in advanced development. 6 As exports of LNG grow, natural gas pricing is expected to increasingly be affected by the global price, thereby increasing volatility and making it even more important to keep existing coal-fired units online in order to maximize the availability of fuel switching. lliS calculates that retirement to the existing coal and nuclear generation capacity would result in an increase of retail power prices by about 25 percent and net consumer costs by about In addition, negative energy prices primarily as a result of wind production tax credits are becoming increasing prevalent, with crushing impacts on every type of base load. 5 ABB. Actual and Projected Coal Capacity Retirements in the United States, 2011-2020, Ventyx Database, October 18,2017. 6 EVA, Quarterly LNG Outlook, December 2017. 14 11093257 v2 $98 billion per year. 7 Therefore, failure to maintain the resource diversity by prematurely retiring nuclear (and coal) baseload units could, extrapolating over the next 20 years, increase electricity costs by as much as $2 trillion. These effects are magnified further as soaring electricity costs ripple through the broad economy, with large adverse impacts over the three year period on U.S. GDP (a loss of 0.8% ), on real disposable income (a drop of about $845 per household in 2016 dollars), and jobs (a loss of 1 million). 8 IV. COMMENTS IN RESPONSE TO PROPOSED ACTION A. The CPP Exceeds The Authority Granted to U.S. EPA Under Section 111 of the CAA. The Obama CPP requires States to adopt standards of performance that cannot be met by fossil fuel-fired generation sources under current state-of-the-art technology. Thus, the purpose of the CPP's standards of performance is to require fossil fuel-fired generation sources to transfer their generation to non-fossil fuel generation facilities. However, that illegally-compelled generation shifting is precluded by the unambiguous language and clear structure of section 111 of the CAA. In this proposed action to repeal the CPP, the U.S. EPA correctly concludes that a proper construction of section 111 (a)( 1) of the CAA is limited to emission restriction measures that can be applied to or at an individual stationary source and does not authorize generation shifting to alternate sources. After reconsidering the statutory text, context, and legislative history, and in consideration of the EPA's historical practice in its other existing CAA section 111 regulations, the Agency proposes to return to a reading of CAA section 111(a)(l) (and its constituent term, "best system of emission reduction") as being limited to emission restrictions that can be applied to or at an individual stationary source. That is, such measures must be based on a physical or 7 8 IHS Study at 5, 37-38. HIS Study at 5, 39. 15 11093257 v2 operational change to a building, structure, facility, or installation at that source, rather than measures that the source's owner or operator can implement on behalf of the source, at another location. 82 Fed. Reg. at 48039. The U.S. EPA properly concludes that this interpretation accords with section 111, aligns with Congressional intent, aligns with prior agency interpretations of section 111, avoids illogical results and avoids conflict with the State's sovereign rights. 82 Fed. Reg. at 48039. The U.S. EPA also properly concludes that this interpretation requires immediate repeal of the Clean Power Plan. ld. at 48038. Under section 111(d) of the CAA, U.S. EPA's role is to establish a procedure for States to submit plans establishing standards of performance for any existing source. CAA section 111(d)(1). State plans, in turn, must apply a standard of performance to any particular source. ld. The CAA defines a "stationary source" as "any building, structure, facility, or installation which emits or may emit any air pollutant." CAA § 111(a)(3). Thus, section 111(d) permits U.S. EPA to require the States to establish performance standards only for the facility whose emissions are being controlled. Requiring an owner or operator of a fossil fuel-fired source to construct, or to subsidize generation at other facilities, as the CPP does, is not a standard/or that existing source at all. Further, U.S. EPA's previous application of the "standards of performance" to multiple, combined sources at the level of an entire "source category" as opposed to the individual sources at "single source" level, also directly contravenes the express requirements of CAA section 111. Section 111 clearly provides for EPA to "list" source categories and then, where section 111(d) applies, requires the States to set "standards of performance for any existing source" within a source category. By applying emission performance standards to require the shifting of generation from existing fossil-fuel fired sources (one source category) to renewable generation 16 11093257 v2 facilities (a different source category), the U.S. EPA went well-beyond even its professed single "source category" application of performance standards. In actuality, U.S. EPA is applying the CAA performance standards across multiple sources in multiple source categories. There is absolutely no basis for this application of the performance standards to multiple source categories under existing law. U.S. EPA may not "embellish" the statutory definition of "stationary source" by "rewrite[ing] the definition of a stationary source." ASARCO Inc. v. EPA, 578 F.2d 319,324, 326 n.24 (D.C. Cir. 1978). The U.S. EPA correctly concludes in this action that section 111(d) requires that standards must be set for individual sources. The EPA's proposed interpretation is also guided by CAA section 111(d)'s direction that standards be established "for any existing source," (emphasis added) and not for other sources or entities. See also 42 U.S.E. 7401(a)(3) (finding that "air pollution control at its source is the primary responsibility of State and local governments") (emphasis added). Further, the "for any existing source" phasing in CAA section 111(d) mirrors the "for new sources" phasing in the first sentence of section 111(b)(1)(B). In other words, as applied to both new source standards and existing source standards promulgated under CAA section 111, if standards must be set for individual sources, it is reasonable to expect that such standards would be predicated as measures that can be applied to or at those same individual sources. 82 Fed. Reg. at 48039. Finally, U.S. EPA also unlawfully redefines a CAA-defined "source" to "include ... the 'owner or operator' of any building ... for which a standard of performance is applicable." 80 Fed. Reg. at 64,762. Again, section 111 performance standards apply to a "source," not to the "owners and operator" of that statutorily-defined source. CAA § 111 (a)(3). A "source" is not defined to include the "owner or operator" of the "building, structure, facility, or installation." Indeed, section 111(a)(5) separately defines the term "owner or operator" to mean "any person who owns, leases, operates, controls, or supervises a stationary source". Had Congress intended 17 11093257 v2 to include a facility's owner or operator within the term "source," it would not have separately defined these diverse and mutually-exclusive terms. Again, the U.S. EPA in the proposed action to repeal the Obama CPP correctly concludes that emissions limits apply to the source, not to the owner or operator of the source . . . . Here, contrary to the conclusions in the CPP, the EPA is proposing to interpret the phase "through the application of the best system of emission reduction" as requiring that the BSER be something that can be applied to or at the source and not something that the source's owner or operator can implement on behalf of the source at another location. Interpreting the statute as carrying this additional limiting principle ensures conformity with the statutory context and congressional intent. 82 Fed. Reg. at 48039. Under section 111(d), U.S. EPA must show that Congress clearly authorized the agency to restructure power markets. The CPP' s attempt to reconfigure the sources of generation for the power grid is precisely the sort of significant and transformative assertion of authority that, under the Supreme Court's decisions, requires "clear congressional authorization." Util. Air Regulatory Grp. V. EPA, 134 S. Ct. 2427, 2444 (2014) ("UARG"). The clear congressional statement rule applies with particular force here where U.S. EPA has "no expertise" in the subject matter so as to justify Chevron deference to its unprecedented assumption of authority under the CPP. "[G]rid reliability is not a subject of the Clean Air Act and is not the province of EPA." Del. Dep't of Nat. Res. & Envtl. Control v. EPA, 785 F.3d 1, 18 (D.C. Cir. 2015). Congress did not delegate to U.S. EPA the authority to reconfigure the entire grid to lower overall emissions while ·maintaining reliable and low-cost generation. The "clear congressional statement" requirement is fatal to the CPP. There is no plausible claim that Congress in section 111 (d) authorized U.S. EPA to set emission performance rates on the basis that the owners of existing fossil fuel-fired sources would be required to meet 18 11093257 v2 the rates by transferring generation to lower-emitting generation to displace their own generation. In sum, the unambiguous CAA section 111 requirement that standards of performance must be set ''for" and be "applicable ... to" individual sources forecloses U.S. EPA's claim to authority to enforce CPP compliance by reordering electric generation from one source to another source within a State's grid. CAA sections 111(d)(1), 111(a)(2) (emphasis added). "Generation shifting" does not entail setting standards that are "for" or "applicable" to operations within individual, regulated sources. Rather, it involves replacing or reducing the generation of individual regulated sources in a particular generation category with the generation of entirely different kinds of facilities in a different generation category. That unilateral redistribution of electric generation is plainly beyond what CAA section 111 permits. Murray Energy therefore fully supports the U.S. EPA's proposal to repeal the CPP. The reasons set forth in this action, however, are not the only grounds on which the CPP violates section 111. The CPP also violates section 111 in that it mandates that a regulated source cease producing electricity, rather than addressing how the regulated source is to reduce emissions while continuing to produce electricity at that existing source. A CAA "standard of performance" is a measure by which to regulate the operation of a source, not an order to cease the operation of a source. A CAA "standard of performance" must reflect reductions from an "emission limitation," which in tum must "limit ... the quantity, rate, or concentration of emissions of air pollutants on a continuous [i.e., operating] basis." CAA section 302(k) (emphasis added). As Congress made clear, the terms "standard of performance" and "emission limitation" are defined to preclude performance rates based on "intermittent controls," such as cutting or shifting production to other facilities. Id. sections 111(a)(l), 302(k). The CPP is thus 19 11093257 v2 directly contrary to the CAA's central premise that a "standard of performance" apply to a generation source and should be repealed for this reason as well. In addition, as U.S. EPA notes, the CPP established standards that were more stringent for existing sources than for new or newly-modified sources, either under BACT or EPA's 111 (b) standards. See 82 Fed. Reg. at 48041; /d. at 48041, n.16. EPA cites these errors in legal interpretation as grounds for reversing the agency's prior position that imposition of requirements beyond the source itself, such as requiring generation shifting, cannot be the basis for an existing source performance standard. See /d. at 48042. This is correct, but not the only necessary conclusion from EPA's analysis. These conclusions are also independent grounds for repeal of the CPP. An existing source performance standard that is more stringent that the standards applicable to new sources is arbitrary and capricious, regardless of whether it includes regulation beyond the source itself or requires generation shifting. B. The CAA Section 112 Exclusion Unambiguously Prohibits the CPP. The CAA section 112 Exclusion prohibits U.S. EPA from employing section 111(d) to regulate a source category that is already regulated under section 112. And because it is undisputed that fossil fuel-fired generating units are already regulated under section 112, see 77 Fed. Reg. 9304 (Feb. 16, 2012), the section 112 Exclusion prohibits any attempt by the U.S. EPA to invoke section 111 (d) to re-regulate those same plants. The section 112 Exclusion's prohibition against employing section 111 to regulate "any air pollutant" emitted from a "source category ... regulated under section [1] 12" has a straightforward and unambiguous meaning. If a source category is "governed by [a] rule" under section 112, U.S. EPA may not require the States to set a standard of performance for sources in that category under section lll(d). As the Supreme Court has said, "EPA may not employ 20 11093257 v2 [section 111 (d)] if existing stationary sources of the pollutant in question are regulated under ... § [1] 12." Am. Elec. Power Co. v. Connecticut, 131 S. Ct. 2527, 2537 n.7 (2011) ("AEP"). This literal reading of the section 112 Exclusion is, as U.S. EPA itself has explained, consistent with the statutory and legislative history of the CAA's 1990 Amendments. Before 1990, section 112 covered an extremely narrow category of life-threatening pollutants. See S. Rep. No. 91-1196, at 20 ( 1970). In 1990, Congress expanded the reach of the section 112 program, significantly broadening the definition of pollutants under section 112 to include those "which present, or may present ... a threat of adverse human health effects ... or adverse environmental effects," CAA § 112(b)(2). As U.S. EPA has said in the past, the House of Representatives (where the current text of the section 112 Exclusion originated) responded to this fundamental expansion in section 112 by "chang[ing] the focus of [the Exclusion and] seeking to preclude regulation of those pollutants that are emitted from a particular source category that is actual! y regulated under section 112." 70 Fed. Reg. at 16,031. That is, the House determined that existing sources, which have significant capital investments and sunk costs, should not be burdened by both the expanded section 112 program and performance standards under section 111 (d). !d. at 16,031-32. Because the U.S. EPA has already regulated fossil fuel-fired EGUs under CAA section 112, the section 112 Exclusion barred promulgation of the CPP and prohibits any replacement rule. As part of its final CPP repeal, U.S. EPA should rest its rule upon this ground in addition to the defensible and common-sense logic already set out in the proposal and the other grounds set forth in these comments. While the proposal clearly sets forth sufficient grounds for repealing the CPP, the U.S. EPA must include this additional (and broader) ground for repeal because the section 112 Exclusion supports that the CPP promulgation fell outside of United States policy 21 11093257 v2 that "environmental regulations comply with the law." Executive Order 13783, at Section 1(e). In order to fully comply with the executive order, the U.S. EPA needs to address the illegal interpretation of the CAA (in violation of the section 112 Exclusion) that the Obama Administration advanced in its blind march to adopt the CPP. C. The Proposed CPP Repeal Rule Does Not Address All Concerns Over the Scope of EPA's Authority to Imposed Existing Source Performance Standards for GHGs at Electric Generating Units. In the present action, the U.S. EPA specifically invited comment as to whether the Proposed CPP Repeal Rule, by substantially diminishing the potential economic and political consequences of any future regulation of COz emissions from existing fossil fuel-fired EGUs, has the advantage of not implicating the "clear statement" rule, in that it would avoid potentially transformative economic policy, and political significance in the absence of a clear congressional statement of intent to confer such authority on the Agency. 82 Fed. Reg. at 48042. While the Proposed CPP Repeal Rule fixes several errors on the jurisdictional justification U.S. EPA used to promulgate the CPP, these fixes do not address all problems with the CPP. For example, the Proposed Rule does not address EPA's erroneous conclusion that it has authority to regulate GHGs under section 111 of the CAA. Further complications arose in the CPP from EPA's erroneous attempt to regulate natural gas and coal fired EGUs under the same standards. Any future rulemaking of EGUs under section 111 must not only avoid generation shifting and the imposition of requirements beyond the source itself to avoid violation of the "clear statement" rule. It must avoid extending EPA's jurisdiction beyond that established by Congress. This includes avoiding undue influence on the nation's generation mix and energy markets, even in forms that come through standards applicable at the source level. 22 11093257 v2 D. The CPP Unlawfully Abrogates Authority Granted to the States by the Clean Air Act and FERC under the Federal Power Act. In the proposed action, the U.S. EPA has specifically invited comment as to whether the CPP exceeded the proper role and authority of the Agency by regulating aspects of the nation's energy sector that are within the proper jurisdiction of the States and FERC. 82 Fed. Reg. at 48042. The CPP exceeded EPA's authority in both of these regards. As addressed above, section 111 (d) grants the authority to establish standards of performance for existing sources to the States-not to U.S. EPA. CAA § lll(d)(1). Under section lll(b), U.S. EPA is empowered to adopt "regulations ... establishing Federal standards of performance for new sources." (emphasis added). In contrast, under section lll(d), U.S. EPA's authority is limited to adopting a "procedure" under which "each State shall submit to [EPA] a plan which ... establishes standards of performance for any existing source," and to "prescrib[ing] a plan for a State in cases where the State fails to submit a satisfactory plan." /d. § 111 (d)( 1), (2) (emphasis added). As U.S. EPA admits, the CPP forbids the States to impose emission standards that are less stringent than EPA has mandated through the national performance rates. 80 Fed. Reg. at 64,870. By establishing a minimum requirement for emission standards that are to imposed by the States and then leaving only the work of implementation for the States, EPA has unlawfully rewritten the statutory text in which Congress expressly gave only to the States the authority to "establish[] standards of performance." CAA § 111(d)(1) (emphasis added). "[T]he regulation of utilities is one of the most important of the functions traditionally associated with the police power of the States," Ark. Elec. Coop. Corp. v. Ark. Pub. Serv. Comm'n, 461 U.S. 375, 377 (1983), which the Supreme Court has specifically recognized should not be "superseded" "unless that was the clear and manifest purpose of Congress." Pac. Gas & 23 11093257 v2 Elec. Co. v. State Energy Res. Conservation & Dev. Comm'n, 461 U.S. 190,206 (1983) ("PG&E"). Under the Federal Power Act, "the States retain their traditional responsibility in the field of regulating electrical utilities for determining questions of need, reliability, cost, and other related state concerns." !d. at 205. Other aspects, including electric utilities engaged in interstate commerce, including wholesale sales, transmission of electric energy in interstate commerce, and reliability that fall outside regulation by the States, lies within the jurisdiction of PERC. 16 U.S.C. §§ 824-824w. To meet the CPP's arbitrary emission reduction standards, States will be compelled to enact legislation and regulations restructuring their power generation and distribution systems, decommissioning coal-fired plants, and granting regulatory and siting approval to new, renewable energy projects. Even if the CPP's demand that States take these actions were constitutional, EPA may not make these "decision[s] of the most fundamental sort" for the States without clear authorization from Congress. Gregory v. Ashcroft, 501 U.S. 452,460 (1991). "Although the Constitution grants broad powers to Congress, our federalism requires that Congress treat the States in a manner consistent with their status as residuary sovereigns and joint participants in the governance of the Nation." Alden v. Maine, 527 U.S. 706,748 (1999). Among the powers that the Constitution denies to the federal government is the power to "use the States as implements of regulation"-in other words, to commandeer them to carry out federal law. New York v. United States, 505 U.S. 144, 161 (1992). The CPP violates this anti-commandeering principle by forcing the States and state officials to exercise their sovereign powers to revamp their utility sectors to comply with EPA's unilateral dictates. Under the CPP, the state actors will be the ones to account for the CPP's impact on electric reliability, 40 CPR§ 60.5745(a)(7), through such means as "[public utility 24 11093257 v2 commission] orders," 80 Fed. Reg. at 64,848, and "state measures" that make unregulated renewable energy generators "responsible for compliance and liable for violations" if they do not fill the gap, 40 CFR § 60.5780(a)(5)(iii). Even under a federal implementation plan, state agencies will have to be involved in decommissioning coal-fired plants, addressing replacement capacity, addressing transmission and integration issues, and undertaking all manner of related regulatory proceedings. See 80 Fed. Reg. at 64,678. In fact, EPA's proposed federal plan expressly relies on state authorities to address reliability issues caused by the CPP. 80 Fed. Reg. at 64,981. Just as the federal government may not commandeer the States to carry out federal policy, it also may not coerce them to the same end by denying them "a legitimate choice whether to accept the federal conditions." Nat'! Fed. of Indep. Bus. v. Sebelius, 132 S. Ct. 2566, 2602 (2012) (Roberts, C.J.) (plurality opinion); see also id. at 2659 (Scalia, Kennedy, Thomas, and Alito, JJ., dissenting). The CPP violates this anti-coercion doctrine by threatening to disrupt the electric systems of the States that do not carry out federal policy. If a State declines to implement the CPP, EPA will impose a federal plan that does so. 40 CFR § 60.5720. But because the CPP's aggressive emission rates cannot be achieved by operating fossil fuel-fired sources under existing technology, the States will have to force fossil fuel-frred sources to transfer generation to other sources; the only alternative will be for fossil fuel-fired sources to shut-down, which will result in electricity shortfalls and the associated consequences for state services and operations, public health and safety, and the economy. The CPP places the States in an untenable position. The entire point of the CPP is to force the States to compel fossil fuel-fired sources to transfer generation to renewable sources. The States would not compel such transfer of 25 11093257 v2 generation absent the CPP' s coercion. EPA has no authority under the CAA to engage in such coercion. Moreover, EPA's attempts to control interstate electricity transmission and regulate to the detriment of grid reliability invade the exclusive jurisdiction of PERC. The U.S. EPA Should Immediately Review and Rescind the E. 2009 Endangerment Finding and Legal Memoranda. 1. The U.S. EPA Continues to Lack Authority to Regulate C02 Under Section 111 Because It Has Not Made (And Cannot Make) the Statutorily-Required Endangerment and Significant Contribution Findings. Regulation of a stationary source category under CAA section 111 must be predicated upon the agency's finding that: (1) emissions of the regulated air pollutant "may reasonably be anticipated to endanger public health or welfare"; and (2) the continued emission of the air pollutant "contributes significantly" to that endangerment. 42 U.S.C. § 7411(b)(l)(A). 9 Without both of these findings, the category of stationary sources cannot be subject to new source regulations for emissions of that pollutant or, consequently, subject to existing source emission regulations for that pollutant. Nevertheless, the Obama Administration attempted to implement the CPP-a regulation of a new pollutant (C02) from a new source category (fossil fuel-fired electricity generating units or EGUs)-without attempting to make either the required endangerment finding or the required significant contribution finding. Fortunately, the stay by the Supreme Court prevented formal implementation. It nonetheless triggered numerous power plant retirements in expectation of its implementation. The CPP was the center piece of the Obama's Administration's "War on Coal" even though it did not and could not put together an endangerment finding (much less a significant 9 While both the endangerment finding and the significant contribution finding language comes from CAA section 111(b), this language is clearly linked to the listing of categories of stationary sources, which is the requisite for regulation under both section 111(b) and section 111(d). Thus, Murray Energy refers to these required findings as the section 111 findings . 26 11093257 v2 contribution finding) that correlated C02 emissions from fossil fuel-fired EGUs with recognizable dangers to domestic public health and welfare. The present action is the appropriate mechanism by which the U.S. EPA should now correct the Obama Administration's error. Proponents of the CPP might point out that the Obama Administration did manage to make an endangerment finding under CAA section 202(a)(1) as respects transportation sources. See 74 Fed. Reg. 66,496 (Dec. 15, 2009). But that endangerment finding is irrelevant to this rulemaking because it does not satisfy the requisite findings for section 111 regulation for EGUs. First, section 111 requires a specific finding that the proposed source category subject to regulation "contributes significantly" to the endangerment. 42 U.S.C. § 7411(b)(l) (emphasis added). The endangerment finding alone is not sufficient to regulate under CAA section 111, and an endangerment finding under CAA section 202(a)(l) does not include a significant contribution finding; it merely requires that the agency finds that air pollution from motor vehicles "cause or contribute to," the endangerment. See 42 U.S.C. § 7521(a)(1). Thus, even if the section 202(a)(1) endangerment otherwise supported section 111 regulation-which, as provided below, it does not-the U.S. EPA would need to make an additional significant contribution finding in order to promulgate section 111 regulations. The U.S. EPA did not make, still has not made, and cannot factually support a significant contribution finding for C02 emissions from fossil fuel-fired EGUs. Thus, the statutory prerequisite to promulgate the CPP was never met. A second reason that the section 202(a)(l) endangerment finding did not satisfy the statutory prerequisite for promulgating the CPP is because section 111 requires a finding that the regulated pollutant from the specific source category sought to be regulated "may reasonably be 27 11093257 v2 anticipated to endanger public health." 42 U.S.C. § 74ll(b)(l)(A). The section 202(a)(l) endangerment finding did not address COz from fossil fuel-fired EGUs. By the U.S. EPA's own wording, during section 202(a)(l) finding, the agency merely set out "to determine if emissions of the well-mixed greenhouse gases from CAA section 202(a) source categories contribute to the air pollution that endangers public health and welfare." 74 Fed. Reg. at 66,499 (emphasis added). The U.S. EPA did not determine in that finding whether any emissions from fossil fuelfired EGUs contributes significantly to air pollution reasonably anticipated to endanger public health or welfare. This is an important distinction that leaves the section 202(a)(l) endangerment finding irrelevant to the agency's authority to promulgate section Ill regulations. The agency simply did not possess authority for section 111 regulation based on the section 202(a)(l) endangerment finding. In finalizing the CPP repeal in Docket ID No. EPA-HQ-OAR-2017-0355 and in the present action, the U.S. EPA now has before it a simple task of determining that the lack of a section 111 endangerment finding renders the CPP contrary to "the policy of the United States that necessary and appropriate environmental regulations comply with the law." See Executive Order 13783, at§ l(e). 2. EPA's Legal Memoranda Should Also Be Rescinded In Their Entirety. The Legal Memoranda comprise over 200 pages of erroneous legal analysis used by EPA to support the CPP' s improper jurisdictional overreach and minimize the legal and technical hurdles posed by the Clean Air Act, years of case law, and the numerous petitions and comments or parties who sought to limit the agency to its jurisdictionally proper role. EPA has correctly identified several key flaws in the Legal Memoranda related to their defense of an interpretation of the Clean Air Act that is inconsistent with its plain text, the structure and purpose of the Act, 28 11093257 v2 Congressional intent, and the Agency's prior regulatory actions. These are not the only flaws in the Legal Memoranda, however. These same memoranda support the violation of the section 112 exclusion supported by the CPP, the interference with State and FERC jurisdiction over the power sector, as achieved by the CPP, and the regulation ofGHGs and fossil-fuel fired EGU's in the absence of a proper endangerment finding. The Legal Memoranda are not based on sound legal analysis and do not reflect the current views of the Administration. They should, therefore, be rescinded in their entirety. To the extent the Agency believes any portion of the Legal Memoranda accurately reflect the Agency's current position, the Agency should restate those positions in a separate legal memorandum. This is the only way to make clear what positions have been rescinded and what positions are supported by the Agency. To do otherwise will cause confusion as to which conclusions remain in effect, which have been specifically rescinded, and which can no longer be considered viable because they depended in whole or in part on logic, policy, or reasoning reflected in those portions that have been rescinded. 3. Even If The 2009 Endangerment Finding Could Serve As The Requisite For Section 111 Regulation of EGUs, The Finding Is Severely Biased, Not Supported By Sound Scientific Evidence And Highly Speculative. Even if the section 202(a)(1) endangerment finding could serve as the requisite for section 111 regulation, repeal of the CPP should still be based on the factual and procedural shortcomings of that finding. A clear procedural flaw of the finding is that the proposed finding was never submitted to the Science Advisory Board for peer review as statutorily required. This flaw has been fully drawn out by the Texas Public Policy Foundation in its reconsideration petition (Appendix A herein) filed on behalf of seven petitioners. Murray Energy, as part of these Comments, fully adopts the reasoning therein by reference and inclusion in Appendix A. 29 11093257 v2 The agency's failure to make its own judgment is another procedural flaw of the section 202(a)(1) endangerment finding. The CAA requires that U.S. EPA exercise its own judgment in promulgating the endangerment finding. 10 This necessarily entails that the agency create a record and then, in its own judgment, articulate whether danger to public health and welfare is anticipated. As part of the section 202( a)( 1) endangerment finding, the agency did not exercise its own judgment. Instead, it simply adopted as its own determination the findings by a select number of other organizations, most notably the finding of the U.N. Intergovernmental Panel on Climate Change (IPCC). See, e.g., Technical Support Document, Endangerment and Cause or Contribute Findings for Greenhouse Gases under Section 202(a) of the Clean Air Act, Doc. No. EPA-HQ-OAR-2009-0171-0137, at 6. In promulgating the endangerment finding, the U.S. EPA all but admitted that it had not exercised its own judgment. See 74 Fed. Reg. at 66,511 ("[T]he Administrator is placing primary and significant weight on these assessment reports in making her decision on endangerment."). The problems with the U.S. EPA's adoption of the findings of others as opposed to exercising its own judgment has been well-documented, see generally the State of Texas Petition for Reconsideration 11 , and the U.S. EPA Administrator himself has acknowledged 12 this procedural shortcoming of the finding. This is true for endangerment findings required under section 111 or section 202(a)(l). See 42 U.S.C. § 7411(b)(1) ("The Administrator ... shall include a category of sources in such a list [of categories of stationary sources] if in his judgment it causes, or contributes significantly to, air pollution which may reasonably be anticipated to endanger public health or welfare." (emphasis added)); id. § 7521(a)(1) ("The Administrator shall by regulation prescribe ... standards applicable to the emission of any air pollutant from any class or classes of new motor vehicles or new motor vehicle engines, which in his judgment cause, or contribute to, air pollution which may reasonably be anticipated to endanger public health or welfare." (emphasis added)). 11 This petition can be viewed at: https://www.epa.gov/sites/production/files/20 16-08/documents/petition for reconsideration state of texas.pdf. 10 In an interview with Time Magazine, Administrator Pruitt stated that in promulgating the 202(a)(1) endangerment finding, the U.S. EPA "took work product of the U.N. Intergovernmental Panel on Climate Change and adopted it, and transferred it to this agency." The full interview can be found at: http:/ltime.com/4998279/company-man-inwashington/. 12 30 11093257 v2 To be sure, the IPCC report could have been part of the record that the agency had decided to consider in exercising its own judgment. But the statute requires the agency itself, with its unique expertise, to exercise its own judgment and come to its own conclusions. In this case, borrowing conclusions from the IPCC report is particularly problematic, because the U.S. EPA should have made a determination as to whether domestic sources endangered domestic public health or welfare. This is not the task that the U.N. Intergovernmental Panel carried out, and the intent of Congress in enacting_the CAA was circumvented when the U.S. EPA substituted conclusions from a non-U.S. entity for its own. Finally, the section 202(a)(l) endangerment finding rested on significant factual errors. The assumptions the agency relied upon in finalizing that finding are increasingly being proven to be factually inaccurate. For example, we now have very credible data demonstrating that the models utilized by the U.S. EPA simply are not accurately predicting how C02 emissions affect the atmosphere. This undercuts the entire factual basis for the endangerment finding. And it is becoming progressively clearer that regulating C02 in the United States will have no discemable effect on the atmosphere or our climate, much less that such regulation would address endangerment of public health or welfare. These factual bases are described with legal and scientific detail in the documents attached to these Comments as Appendix B (U.S. House Committee on Science, Space & Technology Testimony of Professor of Atmospheric Science John R. Christy of the University of Alabama, March 29, 2017) and Appendix C (the Competitive Law Institute's Petition to the U.S. EPA, including additional testimony by Prof. Christy). These Comments hereby incorporate, by reference and by inclusion in the appendices, the legal and factual bases set out in those documents. The tenuous factual support for the 202(a)(l) endangerment finding provides even further evidence that the agency should support 31 11093257 v2 its CPP Repeal on the additional ground that no fact-based endangerment finding has been promulgated to provide the authority for the CPP. Attached as Appendix D and incorporated herein by reference is the recent report prepared by Dr. Roy W. Spencer- "Analysis of the Scientific Underpinnings Of The EPA Endangerment Finding and Clean Power Plan" (February 13, 2018). Dr. Spencer is a Principal Research Scientist at the University of Alabama in Huntsville where he directs a variety of climate research projects. Dr. Spencer received his Ph.D. in Meteorology from the University of Wisconsin in 1981, and was formerly a Senior Scientist for Climate Studies at NASA's Marshall Space Flight Center. Dr. Spencer is the U.S. Science Team Leader for the AMSR-E instrument flying on NASA's Aqua satellite, which monitors global sea ice conditions, sea surface temperatures, precipitation, and other climate variables. Together with Dr. John Christy, Dr. Spencer is co-developer of the original satellite method for precise monitoring of global temperatures from Earth-orbiting satellites, for which he was awarded NASA's Medal for Exceptional Scientific Achievement, and the American Meteorological Society's Special Award. Dr. Spencer has testified in both houses of the U.S. Congress several times on global warmingrelated subjects. His climate-related publications have emphasized the measurement of precipitation and temperature from space, as well as methods for using satellites to diagnosis climate feedbacks for the purpose of estimating climate sensitivity, hurricane intensity, and extratropical storm strength. (Appendix D, p. 3). Based on his detailed analysis of the 2009 Endangerment Finding, Dr. Spencer concludes that the scientific "claims" asserted by the U.S. EPA in its Endangerment Finding are severely biased, and not supported by a significant body of peer reviewed and published evidence. Some of the claims verge on pure speculation, others are exaggerated, and overall a large body of 32 11093257 v2 published scientific work was simply ignored. Additionally, newly published information since the 2009 Endangerment Finding also suggest a reassessment is in order. The 2009 Endangerment Finding should be reconsidered in light of new evidence and the procedural and factual shortfalls of the 2009 Endangerment Finding. No new CPP rule should be promulgated absent this review of all available scientific evidence. (/d., pp. 4, 52-53). Initially, Dr. Spencer confirms the concerns addressed above that the U.S. EPA improperly relied on the UN Intergovernmental Panel on Climate Change (IPCC) rather than on independent peer-reviewed evidence. The IPCC is composed of bureaucrats from the world's nations who used like-minded scientists to support the IPCC's goal of reducing C02 emissions. Scientists who did not share that goal were excluded from the process. The IPCC ignored alternative, natural explanations of climate change and the role of natural, internally-driven climate cycles. (/d., pp. 5-6). In his Analysis, Dr. Spencer addresses the fundamental concepts of energy balance in temperature change and the two different classes of energy imbalance which can cause climate change. He addresses each ofthe principal scientific "claims" asserted by the U.S. EPA in its 2009 Endangerment Finding and criticizes a number of the fundamental claims asserted in each of four (4) classes, including observed trends and effects in GHGs and supposedly "modeled" projections of future climate changes. (ld., pp. 7-8). Dr. Spencer first addresses the important concept of "energy balance" which is fundamental to understanding climate change due to any cause. There are two general classes of energy balance in the environment- radiative and non-radiative. An example of radiative energy balance is the balance between absorbed sunlight and thermally-emitted infrared (heat) radiation which is how the climate system naturally emits energy and cools itself to outer space. 33 11093257 v2 In contrast, an example of non-radiative energy balance is the El Nino and La Nina phenomena, where the average rates of energy transport between the atmosphere and ocean are temporarily altered and involve natural changes in the transports of heat between the atmosphere and ocean. Significantly, the IPCC improperly emphasizes radiative energy balance while largely ignoring non-radiative energy balance. The IPCC improperly focuses on "external" radiative forcing, including anthropogenic forcing, while ignoring non-radiative forced energy imbalance, that is the natural changes in ocean vertical circulation. (/d., pp. 9-13). Dr. Spencer concludes: This issue is important because, as we shall see, the energy imbalance associated with climate change is exceeding small (around 1%) and not computable from physical first principles, not observable from even our best surface and satellite measurement systems, and capable of occurring through natural processes alone, thus causing natural climate change. (ld., p. 10; emphasis added). Concerning the U.S . EPA's claims of observed trends in C02 concentrations in the atmosphere, Dr. Spencer concludes, even if true, C02 still represents only 0.04% of the earth's atmosphere. This tiny component, on the other hand, is nevertheless necessary for life to exist on Earth, since photosynthesis on both land and in the ocean is necessary for the food chain. Furthermore, no matter how much C02 humanity produces, an average of 50% of it is removed by nature each year. (ld., pp. 15-16). Dr. Spencer takes particular issue with every one of the U.S. EPA' s claims concerning observed effects associated with global elevated concentrations of GHGs. (/d., pp. 17-36). In summary terms, there is no scientific basis for U.S . EPA to assert that climate change is more due to GHG emissions than to natural cycles in the climate system. First, Dr. Spencer observes that compared to the approximate 240 W1m2 average rates of energy flows, the U.S. EPA asserted 0.6 to 2.4 W/m2 imbalance amounts to 0.25% to 1% of 34 11093257 v2 average flows. But even a 1% global radiative balance cannot be reproduced by climate models from physical first principles alone, so the supposed GHG-imposed imbalance is below what climate models can resolve from physical first principles. Instead, models must be "tuned" to produce an assumed energy balance (i.e. assumed no natural climate change) and their tuning parameters are not well constrained. Nor can an anthropogenic energy imbalance be measured from our best satellite energy budget instruments. And even if one would accurately measure the Earth's radiative energy imbalance, there is no way to determine how much is due to anthropogenic versus natural forcings. (/d., pp. 17-18). Second, any global warming since the mid-201h century, is fully consistent with the emergence from the Little Ice Age occurring from approximately 1400 AD to 1700 AD and not inconsistent with past warming cycles in the Roman Warm Period from approximately 0 AD to 200 AD or the Medieval Warm Period from approximately 850 AD to 1100 AD. (ld., p. 19-20). The chart below indicates global warming is consistent with prior climate change cycles. 35 11093257 v2 1.0 N. Hemisphere Temperature proxies (Ljungqvist, F.C. 2010) 0.8 ._ ~~ 06 • 0.4 r:i.!i 0.2 00 1-: -0.4 .i.f ~!!. ,; .:.. .0.2 Medieval Warm Period Roman Warm Period Modern Warm Period 1\ . "A /\ 1\J\ A <>(:, l"'no"" ~~'""V ·~v~v -:~: z! -0.6 · (10 yr resolution) ~ ~~~ ~~ I 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 YEAR (AD) It is Not Obvious That Recent Warming Is Outside the Range Of Natural Temperature Variations of the Last 2,000 Years 1.0 u 0.8 ....:f 0.6 .~ :i; 0.4 Global Thermometer Data (HadCRUT4) 0.2 < 0.0 ~ -o.2 Trailing 10-ye•r •ver•ges i-0.4 j -0.6 .5! "-o.8 ·1.0 1880 1900 1920 1940 1960 1980 2000 Third, there is no solid basis to attribute recent warming to anthropogenic GHG concentrations. As Dr. Spencer concludes: There is no way to know just how much of recent warming was due to the observed increase at atmospheric C02. The primary IPCC climate model simulations regarding natural forcing alone were with changes in total solar irradiance, stratospheric ozone depletion, and volcanoes. Clearly, this is insufficient; there are many more potential sources of natural climate change. For example, indirect solar effects on global cloudiness; natural, and especially unforced, (non-radiative) fluctuations in the climate system which can also change the global energy balance. So, once again we are presented in the claim with a statement of faith, an argument from ignorance. (ld., pp. 21-22). Fourth, the U.S. EPA failed to distinguish between GHG concentration effects and Urban Heat Island effect. Dr. Spencer concludes: 36 11093257 v2 While the U.S. has likely warmed in recent decades, there is now evidence that as much as half of the warning could be spurious, due to the Urban Heat Island (UHI, e.g. Oke, 1995) effect. When only the most pristine stations in the U.S. are analyzed- that is, those with the least amount of manmade structures and spurious heat sources encroaching upon the thermometer sites - the rate of warming is considerably reduced compared to official NOAA estimates (Watts et al., 2015). This also raises questions about warming trends reported in other land areas of the globe as well. Furthermore, unstated in the claim is that most of the concern for human activities and agriculture would be warming during the summer months (JuneJuly- August), not winter. As can be seen in official NOAA data, warming during the summer in the U.S. has been weaker than in the annual average temperatures, with a warming trend of only +0.11 deg. F/decade (+0.06 de g. C/decade): (!d., p. 22). Fifth, contrary to the U.S. EPA claim, there is no demonstrative correlation between sea level changes and GHG emission. Dr. Spencer concludes: There are a number of points which must be made regarding sea level rise. The first is that, based upon global tide gage data produced by Jevrejava et al. (2014), sea level has been rising since well before human-caused GHG emissions could be blamed (data from http://www.psmsl.org/products/reconstmctions/gslGPChange2014.txt): 37 11093257 v2 10 8 Global Average Sea Level Changes from 1,277 Tide Gages Trailing 5-Vear Averages (Jevrejeva et al., 2014) 6 4 In IV 2 .c u c 0 - -2 -4 -6 * could be only 1.1 or 1.2 mm/yr based upon recent reanalyses of data -8 1855 1875 1895 1915 1935 1955 1975 1995 2015 Thus, the claim deceptively excludes the possibility that recent sea level rise is mostly natural. Importantly, there has been no obvious acceleration of sea level rise during the period of greatest greenhouse gas emissions (generally after the 1940s), as might be expected. In other words, as far as we know, sea level has been rising as we have been coming out of the Little Ice Age. (/d.; p. 26). Finally, the U.S. EPA's claim that there have been widespread changes in extreme temperatures in the last fifty years is a gross exaggeration. Dr. Spencer concludes: At a minimum, the claim is a gross exaggeration. Regarding the U.S., the main concern would be excessive heat (since less excessive cold would be a welcome thing). For 1,114 USHCN stations in the United States, here are the average numbers of days each year that a station exceeded 100 deg. F and 105 deg. F temperatures, from 1895 through 2017, as tabulated from official NOAA data by John Christy (UAH): 38 11093257 v2 Average per station (1114 USHCN Stations) 1895-2017 Number of days daily Maximum temperature above 100•F and 1os•F 20 No significant Trends. 11 of 12 hottest years occurred before 1960 18 16 • 1ooF 14 •10SF 12 10 8 6 1895 1905 1915 1925 1935 1945 1955 1965 1975 1985 1995 2005 2015 John II. Christy Unlwnlty of Aloboml ln Huntnllle 5131fon Iiles from NOAA/HIE occeuetl Z-Z017 As can be seen, there is no obvious trends in very hot days, which would be the main concern. In fact, 11 of the 12 years with the largest number of very hot days occurred before 1960. As previously explained, I have concern that all land-based thermometer data have spurious warming effects from manmade structures replacing natural vegetation, and active heat sources, leading to an Urban Heat Island (UHI) effect. It is not clear how well this has been adjusted for, and there is evidence that in the U.S. warming in recent decades has been exaggerated by as much as a factor of 2 (Watts et al., 2010). (/d., p. 31). Concerning the last two classes of U.S . EPA claims made in the 2009 Endangerment Finding -projections of future climate change and effects - these claims are not based on observational evidence but rather are based purely on projections of future climate change states based on climate modeling. Based on recent information, this modeling has proved biased, exaggerated and faulty. (/d., at pp. 37-52). Dr. Spencer concludes: 39 11093257 v2 There is now a great deal of published evidence that the amount of future warming projected by the models will be too large. The claim, rephrased, is that warming in the 21st Century will accelerate, that is, the rate of warming will be greater than in the 20th Century. But, to date, the models have produced approximately twice the amount of atmospheric warming as has been observed since 1979, which is when we have had our first capability to monitoring the tropospheric temperature over virtually the entire Earth: Tropical Mid-Tropospheric Temperature Variations Models vs. Observations 1.2 5-Year Averages, 1979-2017 Trend line crosses zero at 1979 for all time series • ACCESSl.l IICC.aMUimJ ACC£SS1.0 1.0 ICC-cMS1.1 INU.£SM CCSM4I& runs) CESM1ICAM5) IJ rvnsl CMCC.cM CSI-O.MkJ.i.O ......... EC_EAIITH fi0.£SM GFDL.£SMZG 11 rvnJ GISS.U-1111& rvns) HadGEMZ.£5 IPSI.-CMSA-1.11 -- - 0.4 t FGCJAU.sZ IPSI.~ 0.6 MIIIOC.£SM MPI.£SM.MR Nor£5M1-M JZGtoupMeon I ••• .••••• C.n£SMZ CESM1IIGC) 11 rvn) CNIIMA should conduct a new rulemaking on this issue. Respectfully submitted on behalf of CEI, SEPP, Donna Bethell, Kenneth A. Haapala, Thomas Sheahen, and S. Fred Singer. Dated: February 17, 20 17 (corrected Feb. 23) Is/ Sam Kazman SamKazman Hans Bader Competitive Enterprise Institute 1310 L St., NW, 71h Floor Washington, D.C. 20005 (202) 331-1010 Attorneys for Petitioners 11 Fiscal Year 2017 EPA Budget: Hearing Before the H. Comm. on Energy and Commerce, Subcomm. on Energy and Power, I 14th Cong. at 73-74 (Mar. 22, 20 I 6) (testimony of EPA Administrator Gina McCarthy), available at bup://docs.house.goy/mectio2SIIFIIF03/10160322/10471S/HHRG-t 14-IRJ3-Transgipt-20160322.pdf. 6 ATTACHMENT U.S. House Committee on Science, Space & Technology 2 Feb 2016 Testimony of John R. Christy University of Alabama in Huntsville. I am John R. Christy, Distinguished Professor of Atmospheric Science, Alabama's State Climatologist and Director of the Earth System Science Center at The University of Alabama in Huntsville. I have served as Lead Author, Contributing Author and Reviewer of United Nations IPCC assessments, have been awarded NASA's Medal for Exceptional Scientific Achievement, and in 2002 was elected a Fellow of the American Meteorological Society. It is a privilege for me to offer my analysis of the current situation regarding (1) the temperature datasets used to study climate, (2) our basic understanding of climate change and (3) the effect that regulations, such as theo Parisfagreement, might have on climate. I have also attached an extract from my Senate Testimony last December in which I address (1) the popular notion that extreme climate events are increasing due to humaninduced climate change (they are not), and (2) the unfortunate direction research in this area has taken. My research area might be best described as building datasets from scratch to advance our understanding of what the climate is doing and why - an activity I began as a teenager over 50 years ago. I have used traditional surface observations as well as measurements from balloons and satellites to document the climate story. Many of our UAH datasets are used to test hypotheses of climate variability and change. (1.1) Upper air temperature data from satellites and balloons I shall begin with a discussion that was precipitated by an increasingly active campaign of negative assertions made against the observations, i.e. the data, of upper air temperatures. Figure 1 in particular has drawn considerable attention from those who view the climate system as undergoing a ? rapid, human-caused transformation into a climate to which people would have great difficulty adapting. This simple chart tells the story that the average model projection, on which their fears (or hopes?) are based, does poorly for the fundamental temperature metric that is allegedly the most responsive to extra greenhouse gases - the bulk atmospheric temperature of the layer from the surface to 50,000ft. [The layer shown is known as the mid-troposphere or MT and is used because it overlaps with the region of the tropical atmosphere that has the largest anticipated signature of the greenhouse response by bulk mass -between 20,000 and 50,000 feet.] The chart indicates that the theory of how climate changes occur, and the 1 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology associated impact of extra greenhouse gases, is not understood well enough to even reproduce the past climate [much more in section (2)]. Indeed, the models clearly overcook the atmosphere. The issue for congress here is that such demonstrably deficient model projections are being used to make policy. 1.0 r = = = = = = = = = = = = = = = = = = = = = = ; - - - - -"77""- - , Global Bulk Atmospheric Temperature (Surface-SDK ft) Average of 1021PCC CMIP-5 Climate Model runs 0.4 0.0 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 Fig. 1: Five-year averaged values of annual mean (1979-2015) global bulk (termed "midtropospheric" or "MT") temperature as depicted by the average of 102 IPCC CMIP5 climate models (red), the average of 3 satellite datasets (green- UAH, RSS, NOAA) and 4 balloon datasets (blue, NOAA, UKMet, RICH, RAOBCORE). Because this result challenges the current theory of greenhouse warming in relatively straightforward fashion, there have been several well-funded attacks on those of us who build and use such datasets and on the datasets themselves. As a climate scientist I've found myself, along with fellow like-minded colleagues, tossed into a world more closely associated with character assassination and misdirection, found in Washington politics for example, rather than objective, dispassionate discourse commonly assumed for the scientific endeavor. Investigations of us by congress and the media are spurred by the idea that anyone who disagrees with the climate establishment's view of dangerous climate change must be on the payroll of scurrilous organizations or otherwise mentally 2 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology deficient. Also thrust into this milieu is promotional material, i.e., propaganda, attempting to discredit these data (and researchers) with claims that amount to nothing. Several of these allegations against the data appeared a few weeks ago in the form of a well-made video. I shall address the main assertions with the following material, which in similar form has appeared in the peer-reviewed literature through the years. The video of interest was promoted by a climate change pressure group (Yale Climate Connections, http://www.yaleclimateconnections.org/20 16/01 I over-reliance-on-satellitedata-alone-criticized/) in which well-known scientists make claims that are mostly meaningless or completely wmng relative to the evidence in Fig. I. I wish to make four points regarding the video and demonstrate the misdirection for which such agendized videos, along with a happily mimicking media, are so famous. First, the claim is made the satellites do not measure temperature. In reality, the s~nsors on satellites measure temperature by emitted radiation - the same method that a physician uses to measure your body temperature to high precision using an ear probe. Atmospheric oxygen emits microwaves, the intensity of which is directly proportional to the temperature of the oxygen, and thus the atmosphere. That the satellites measure temperature is evident by the following chart which compares our UAH satellite data with temperatures calculated from balloon thermistors. As an aside, most surface temperature measurements are indirect, using electronic resistance. Comparison of Balloon and Satellite Temperatures (Midtroposphere) at 59 U.S. (VIZ) and Australian balloon stations 1.0 - Balloon ----? satellite (UAH6.0) I correlation = 0 .98 r - -.-- 1979 1984 "!' -- -r 1 --?r-- -, ---T?--r---1?-- -;-?--:?- - r - r - - : r - . - - .- 1989 1994 3 ,---,. 2004 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology Fig. 2: Average temperature variations measured at 59 radiosonde stations in the U.S. (VIZ manufacturer) and Australia. Secondly, the scientists claim that the vertical drop (orbital decay) of the satellites due to atmospheric friction causes spurious cooling through time. This vertical fall has an immeasurable impact on the layer (MT) used here and so is a meaningless claim. In much earlier versions of another layer product (L T or Lower Troposphere), this was a problem, but was easily corrected almost 20 years ago. Thus, bringing up issues that affected a different variable that, in any case, was fixed many years ago is a clear misdirection that, in my view, demonstrates the weakness of their position. Thirdly, the scientists speak of the spurious temperature changes that occur as the satellites drift in the east-west direction, the so-called diurnal drift problem (which was first detected and accounted for by us). They speak of a sign error in the correction ' ' procedure that changed the trend. Again, this error ~as not a factor in the MT layer in Fig. 1, but for the different LT layer. And, again, this issue was dealt with for LT 10 years ago. Finally, though not specifically mentioned in this video, some of these scientists claim Fig. 1 above is somehow manipulated to hide their belief in the prowess and validity of the climate models. To this, on the contrary, I say that we have displayed the data in its most meaningful way. The issue here is the rate of warming of the bulk atmosphere, i.e., the trend. This metric tells us how rapidly heat is accumulating in the atmosphere - the fundamental metric of global warming. To depict this visually, I have adjusted all of the datasets so that they have a common origin. Think of this analogy: I have run over 500 races in the past 25 years, and in each one all of the runners start at the same place at the same time for the simple purpose of determining who is fastest and by how much at the finish line. Obviously, the overall relative speed of the runners is most clearly determined by their placement as they cross the finish line - but they must all start together. In the same way I constructed the chart so that the trend line of all of the temperature time series starts at the same point in magnitude and time (zero value at 1979) so the viewer may see how wide the spread is at the finish line (20 15). One way to look at this is seen in Fig. 3 where I provide what is seen in Fig. 1 except this is only the trend line without the variations that occur from year due to volcanoes and such. This is analogous to plotting the overall average speed of a runner along the course even though they likely ran slower on an uphill, and faster on a downhill. This image indicates the models, on average, warm this global layer about 2.5 times faster than the observations indicate. This is a significant difference that has not been 4 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology explained and indicates the theory of greenhouse impact on atmospheric temperature is not sufficiently known to even reproduce what has already happened. We are not talking about 10 or 15 years here, but 37 years- well over a third of a century. That two very independent types of measuring systems (balloons and satellites) constructed by a variety of institutions (government, university, private) all showing the much slower rate of warming gives high confidence in its result. Thus, the evidence here strongly suggests the theory, as embodied in models, goes much too far in forcing the atmosphere to retain heat when in reality the atmosphere has a means to relinquish that heat and thus warms at a much slower rate. 1.o r;:::==============================::::;------~ Global Bulk Atmospheric Temperature SFC-SOK Ft Trend Lines- Rate of Temperature Change 0.8 0.6 o I Average of 102 IPCC CMIP-5 Climate Model runs ---------- I+0.091 ocfdecade I I+0.079 ocJdecade I Observations Blue - Avg 4 Balloon datasets O.O -0.2 1? ~~I ti~e- The Dnear trend (based on 1979-2015 o:l:, series Intersects at zero at 1979 r-r--t-r-r-t 1975 Green - A: 3 SateiHte datasets 1980 I l I i I 1985 I I T ~l r-1 1990 I I I t-.-r-r-r-r - r 1995 2000 J r-:-r 1 - ....--.-.-,- , -r t - r r.----J -r r r-r~y 2005 2010 2015 2020 T t \ 2025 Fig. 3: The linear trend line over the period 1979-2015. The colors represent the same source data as in fig. 1. I've shown here that for the global bulk atmosphere, the models overwarm the atmosphere by a factor of about 2.5. As a further note, if one focuses on the tropics, the models show an even stronger greenhouse warming in this layer. However, a similar calculation with observations as shown in Fig. 3 indicates the models over-warm the tropical atmosphere by a factor of approximately 3, (Models +0.265, Satellites +0.095, Balloons +0.073 ?C/decade) again indicating the current theory is at odds with the facts. (again, see section 2.) 5 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology It is a bold strategy in my view to actively promote the output of theoretical climate models while attacking the multiple lines of evidence from observations. Note that none of the observational datasets are perfect and continued scrutiny is healthy, but when multiple, independent groups generate the datasets and then when the results for two completely independent systems (balloons and satellites) agree closely with each other and disagree with the model output, one is left scratching one's head at the decision to launch an offensive against the data. This doesn't make scientific sense to me. (1.2) Surface temperature issues There are several issues regarding surface temperature datasets that are too involved to discuss in this material. I shall focus on a few points with which I am familiar and on which I have published. (1.2.a) Surface temperature as a metric for detecting the influence of the increasing concentrations of greenhouse gases One of my many climate interests is the way surface temperatures are measured and how surface temperatures, especially over land, are affected by their surroundings. In several papers (Christy et al. 2006 J. Climate, Christy et al. 2009 J. Climate, Christy 2013 J. Appl. Meteor. Clim., Christy et al. 2016 J. Appl. Meteor. Clim.) I closely examined individual stations in different regions and have come to the conclusion that the magnitude of the relatively small signal we seek in human-induced climate change is easily convoluted by the growth of infrastructure around the thermometer stations and the variety of changes these stations undergo through time, as well as the variability of the natural ups and downs of climate. It is difficult to adjust for these contaminating factors to extract a pure dataset for greenhouse detection because often the non-climatic influence comes along very gradually just as is expected of the response to the enhanced greenhouse effect. In examining ocean temperatures (Christy et al. 2001, Geophys. Res. Lett.) I discovered that the trends of the water temperature (1m depth) do not track well with those of the air temperature just above the water (3m), even if both are measured on the same buoy over 20 years. This is important for the discussion below where NOAA used marine air temperatures to adjust water temperature measurements from ships. There are many other factors that render surface temperature datasets to be of low effectiveness for the detection of enhanced greenhouse warming, (a) lack of systematic geographical coverage in time, (b) unsystematic measuring methods and instrumentation 6 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology in time and space, (c) the point measurement represents at best a tiny, local area and (d) is easily impacted by slight changes in the surroundings, which can occur for example when a station moves. There have been huge efforts to try and adjust the raw surface data to give a time series that ?would represent that of a pristine environment, and I have led or been a part in some of these (e.g. for Central California in Christy et al. 2006 and East Africa in Christy et al. 2009 and Christy 2013). Thus, having experience in building surface, satellite and balloon temperature datasets, and taking into account the signal we are looking for to detect the enhanced greenhouse effect, the evidence suggests to me that utilizing the bulk atmospheric measurements provides the best opportunity to answer questions about the climate's response to this human-induced change in atmospheric composition. The deep atmosphere is much more coherent in space and time in terms of its variations. It is not affected by human development at the surface. It is measured systematically. To be sure, satellite and Balloon temperatures require their own adjustments and 'cannot be considered "perfect", but do offer an independence from one another to allow direct comparison studies. Regarding the detection of the enhanced greenhouse effect, the troposphere, as indicated by models, happens to be the atmospheric region that will respond the most, i.e. warm the fastest, and thus, in my view, is a metric that provides a better way to detect human influence on the climate. (1.2.b) The new NOAA surface temperature dataset A series of papers appeared last year (including Huang et al. 2015 J Climate, Karl et al. 2015 Science) describing a new surface temperature dataset constructed by NOAA which indicated a bit more warming in the past 10 to 25 years than the previous versions. The key change dealt with seawater temperatures in the dataset now known as ERSSTv4. This change introduced an additional warming into the record from about 1990 onward. The main reason for this ne\v warming, as the authors note, was the adjustment applied to buoy data, adding about +0.12 ?C to the buoy readings. In 1980, only about 10 percent of the data reports were from buoys, but by 2000 about 90 percent were buoy data. Thus, because the influence of the buoy data grew significantly through time, the simple addition of a bias to all the buoys from the beginning created a warmer trend as they became the dominate source of information. Some background is necessary. Unlike satellite and balloon datasets which measure a systematic quantity (essentially atmospheric air temperature), surface temperature datasets are a mixture of air (over land) and water (over ocean) temperatures measured over a considerable range of instruments, exposures and methods. Over land, weather stations measure the temperature of the air in varying types of instrument shelters and by 7 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology varying techniques at a level about 5 ft above the ground. Over the ocean, however, the temperature utilized is that of the water itself, not the air above, so traditional global surface datasets do not measure a homogenous physical parameter over land versus ocean. Further, the depth of the water temperature measurement is quite varied from 2 ft to 50 ft or so, by methods that range from buckets drawn up on deck into which a thermometer is inserted to engine-intake temperatures much deeper in the water and to buoys, drifting or moored to the bottom. So the fact temperature varies by depth is an issue to tackle before the possibility of constructing a systematic dataset may be attempted. Then too, the measurements are not spatially or temporally consistent with large regions, such as Africa and the southern oceans, unmeasured. Keep in mind that even though the trend of this NOAA dataset became more positive in the past 10 to 20 years, it is still below climate model projections over the longer term. For longer periods, such as the period since 1979 when satellites began measuring bulk o atmospheric temperatures, the new global dataset is similar to that of the oHadley Centre (1979-2015: NOAA +0.155 ?C/decade, Hadley Centre UKMet, +0.165 ?C/decade). However, there are questions that remain concerning the new NOAA seawater dataset, especially how it indicates more warming in the last 20 years than others. Figure 4 displays the ocean trends for the region 20S to 60N (i.e. tropical and northern hemisphere oceans - there was too little data south of 20S for generating near-surface air temperatures there). There are 4 datasets represented, NOAA (NOAA, red), Hadley Centre (HadCRUT4, orange), a preliminary near-surface air temperature over the oceans by my graduate student Rob Junod (yellow) and the UAH deep layer air temperature from satellites (blue). Both NOAA and HadCRUT4 are temperatures of the seawater near the surface, so should be the same. NOAA used a curious reference variable to calibrate the water temperatures measured from ship intakes -the Night Marine Air Temperature (NMAT). This is curious because there are considerable adjustments required for the NMATs themselves, i.e. corrections for height of ship deck, etc. In any case, from this, the buoy data were then adjusted to match the ship data. It appears, then, that the foundational adjustment process depends on NMATs to adjust the ship data to then adjust the buoy data. The fmal product from NOAA mixes all of these together, and because the geographic representation of the different systems changed dramatically (as noted, from approximately 10% buoys and 90% ships in 1980 to 90% buoys and 10% ships today - Huang et al. 20 15), an adjustment applied to the buoys will automatically influence the trend. I'm aware that the Committee sought information about this curious process and asked NOAA to generate datasets based only on consistent measuring systems, i.e. ships alone, buoys alone and NMATs alone, to see if one system might have impacted the trends 8 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology improperly due to distribution changes. NOAA was unable to accommodate this request. At the same time I asked my graduate student, Rob Junod, to do the work for NMAT. What is presented here is preliminary, but follows much ofthe previous work on NMATs (developed at the National Oceanographic Centre and the Hadley Centre in the UK) with that added advantage ofbeing updated to 2014. The best geographical data coverage was found to be 20?S to 60?N, so this area was also applied to the other datasets for an apples to apples comparison. The results are shown in Fig. 4 in which all trends end in 2014 but cover periods in two-year increments from 20 years to 10 years. Trends in Ocean Temperatures (20S-60N} 10 to 20 Year periods ending in 2014 0.15 . - - - - - - - - - 0 0 0 0 0.10 - - - - - - - -- t:=- NOAA Sea Sfc HadCRUT4 Sea Sfc UAH Air (Near Sea Sfc) UAH LT Deep Layer Air o r- ---- - - - ? - - - ? ? - - - - ~ 0.05 o cu '"0 'V' 0 -,- 0.00 -0.10 --- ,..._ r- ra u -r --, I ?.- D= .., - - - - - - - - - - - - - - -------??-?----?-?-???--?1995-2014 1997-2014 1999-2014 2001-2014 2003-2014 2005-2014 Period of Trend Figure 4. Decadal trends (?C/decade) of four temperature datasets over the oceans from 20?S to 60?N for varying periods ending in 2014. Red and orange are surface seawater temperature datasets from NOAA and the I:Iadley Centre (HadCRUT4). Yellow is a near-surface air temperature dataset (Night Marine Air Temperature) built by UAH (preliminary). Blue is the temperature trend of the deep atmosphere (surface to 35,000 ft or Lower Troposphere) from microwave emissions captured by satellites (also UAHv6.0b5.) A number of observations are evident in Fig. 4. (1) In terms of the temperature trend, the air temperatures are less than those of the water (as indicated in my 2001 study mentioned above.) (2) NOAA warms the fastest in all periods. (3) In the past 10-14 9 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology years, the trends of the HadCRUT4 agree better with the near-surface air temperature dataset (being near zero and supporting the notion of a hiatus) than with the trends from its physically-identical quantity from NOAA. (4) The magnitude of the NMAT trends lies between the trends of the deep atmospheric and sea water. This figure generates a number of data quality questions too. (1) lfNMATs were used to calibrate the ship temperatures and then the ships were used to calibrate the buoy temperatures, why does the NOAA dataset differ so much from its basic reference pointNMATs? (2) What do the time series look like and what are the sub-period trends for seawater under the condition that only ships and/or only buoys are used to build the dataset for the past 20-25 years? (3) What does the time series of NOAA's NMAT (i.e. their reference) dataset show? The real science questions here are those which have significant importance to the o o understanding of how extra greenhouse gases might affec~ the climate as shown in the following section. (2) Ho?w well do '"'e understand climate change? A critical scientific goal in our era is to determine \Vhether emissions from human activities impact the climate and if so by how much. This is made especially difficult because we know the climate system already is subject to significant changes without the influence of humans. Because there is no measuring device that explicitly determines the cause of the climate changes we can measure, such as temperature, our science must take a different approach to seek understanding as to what causes the changes, i.e. how much is nah1ral and how much is human induced. The basic approach today utilizes climate models. (The projections of these models are being utilized for carbon policies as well.) It is important to understand that output from these models, (i.e. projections of the future climate and the specific link that increasing C02 might have on the climate) are properly defined as scientific hypotheses or claims - model output cannot be considered as providing proof of the links between climate variations and greenhouse gases. These models are complex computer programs which attempt to describe through mathematical equations as many factors that affect the climate as is possible and thus estimate how the climate might change in the future. The model, it is hoped, will provide accurate responses of the climate variables, like temperature, when extra greenhouse gases are included in the model. However, the equations for nearly all of the important climate processes are not exact, representing the best approximations modelers can devise and that computers can handle at this point. 10 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology A fundamental aspect of the scientific method is that if we say we understand a system (such as the climate system) then we should be able to predict its behavior. If we are unable to make accurate predictions, then at least some of the factors in the system are not well defined or perhaps even missing. [Note, however, that merely replicating the behavior of the system (i.e. reproducing "what" the climate does) does not guarantee that the fundamental physics are well-known. In other words, it is possible to obtain the right answer for the wrong reasons, i.e. getting the "what" of climate right but missing the "why".] Do we understand how greenhouse gases affect the climate, i.e. the link between emissions and climate effects? As noted above, a very basic metric for climate studies is the temperature of the bulk atmospheric layer known as the troposphere, roughly from the surface to 50,000 ft altitude. This is the layer that, according to models, should warm significantly as C02 increases - even faster than the surface. Unlike the surface I I temperature, this 'bulk temperature informs us about the crux of the global warming question - how much heat is accumulating in the global atmosphere? And, this C02caused warming should be easily detectible by now, according to models. This provides a good test of how well we understand the climate system because since 1979 we have had two independent means of monitoring this layer - satellites from above and balloons with thermometers released from the surface. I was able to access 102 CMIP-5 rcp4.5 (representative concentration pathways) climate model simulations of the atmospheric temperatures for the tropospheric layer and generate bulk temperatures from the models for an apples-to-apples comparison with the observations from satellites and balloons. These models were developed in institutions throughout the world and used in the IPCC AR5 Scientific Assessment (2013). 11 J.R. Christy 2 Feb 2016 House Committee on Science, Space and Technology Global Mid-Tropospheric Temperature Variations Models vs. Observations 1.5 .,---- - ---1 5-Year Averages, 1979-2015 Trend line crosses zero at 1979 for all time series L2 #--=-=~~anmnu~--------==AcAoorumuu~----1-------------~~~~~~~~ - - IICC-tMSU INli-UM CDM4(6nuu ) asMt(CAMS) (3 ,..,., CHI\M Conduction to air =>Infrared radiation => Evaporation of water ENERGvGAIN from stove 9 This is simply a statement of the pt Law of Thermodynamics, also termed Conservation of Energy. Thus, when discussing temperature change, we can and should always be asking, "How has the energy balance changed?" This issue is important because, as we shall see, the energy imbalance associated with climate change is exceedingly small (around 1%} and not computable from physical first principles, not observable from even our best surface and satellite measurement systems, and capable of occurring through natural processes alone, thus causing natural climate change. In regard to the climate system, there are two general classes of energy balance which affects the temperature changes we humans experience: radiative and non-radiative (e.g. Spencer & Braswell, 2014}. The former is the one the /PCC emphasizes, while the latter is one the IPCC largely ignores. An example of radiative energy balance is the balance between absorbed sunlight (the energy source for the climate system) and thermally-emitted infrared (heat) radiation, which is how the climate system loses energy and cools itself to outer space. The increase in atmospheric C02 is, at least theoretically, causing a radiative energy imbalance in the climate system. In contrast, an example of non-radiative energy balance is the El Nino and La Nina phenomena, where the average rates of energy transport between the atmosphere and ocean are temporarily altered, and the lower atmosphere can either warm (EI Nino) or cool (La Nina). Those energy transfers are largely non-radiative, and involve changes in the transports of heat between the atmosphere and ocean. 3.1 Radiative Forcing of Climate Change The Earth, on average, absorbs about 240 Watts per square meter of solar energy, and emits about the same amount of energy to outer space in the form of infrared (heat) radiation (Trenberth eta/., 2009). 10 In the IPCC view, as long as these rates of energy gain and energy loss are equal, the climate system will not change its average temperature on time scales of several decades or longer. Any externally imposed imbalance in these two energy flows is referred to by the IPCC as "radiative forcing", a concept that is useful, but as has been mentioned, incomplete. By "external" radiative forcing, the IPCC means external to the normal operation of the climate system, such as cooling from volcanic aerosols, anthropogenic forcings (e.g. increasing C02, pollution aerosols, land use changes, stratospheric ozone depletion). As we shall see, however, the climate system can undergo "internal" radiative forcing (Spencer and Braswell, 2014), which the IPCC includes in the class of "unforced natural variability". 3.2 Non-Radiative Forcing of Climate Change Much of life on Earth exists in a relatively thin warm zone, sandwiched between the cold upper atmosphere and the cold ocean depths (the global average ocean temperature throughout its entire volume is about 38 deg. F). 11 Ignored in the 11 External Forcing" Paradigm of Climate Change: "Internal" Forcing* Humanity lives in a relatively thin zone of warmth between the cold upper atmosphere and the cold depths of the global ocean (38-39 deg. F average temperature)... *The IPCC calls this "unforced internal variability" Atmosphelie 20 15 Ocean Temp. (deg. C) 0 40"S 20"S 20W EQ 40"N 60"N Because of the large difference in temperature between the ocean surface and the deep ocean, any change in the vertical circulation of the ocean can cause warming or cooling of the global-average lower atmosphere, which is where people live. Due to the extremely long time scales associated with the ocean circulation (centuries to millennia), the changes in climate could also be long, in human terms. The extent to which climate change is affected by such internally-forced variations in the energy balance of the "warm habitable zone" is not well understood, but as we shall see, the changes in energy balance required are very small, around 1%. Since the coupled oceanatmosphere climate system is what's known as a nonlinear dynamical system, capable of its own chaotic variations (Lorenz, 1963), it is entirely plausible (and there is historic evidence to suggest) that much of climate change is simply "unforced internal variability" due to this nonradiative forcing. El Nino and La Nina are examples of this on a short time scale (several years), 12 and they represent a bifurcation in the average climate of the Earth during Northern Hemisphere winter. The Pacific Decadal Oscillation (PDO, e.g. Zhang et al., 1997) is a longer time scale phenomenon, one full cycle lasting about 60 years. Recent warming of the climate system, and especially the Arctic, that began around 1977 might well have been related to the PDQ. If so, this confounds the attribution of recent warming to human GHG emissions. I cannot overemphasize the importance of the three concepts addressed above: (1) energy balance is required to maintain a constant temperature, and climate change can be caused by either (2) radiatively forced energy imbalance (e.g. increasing C02), or (3) nonradiatively forced energy imbalance (e.g. a natural change in the ocean vertical circulation associated with El Nino, La Nina, the Pacific Decadal Oscillation, etc.) These concepts encompass the paradigm within which we talk about climate change and its potential causes. 13 4. The Four Classes of Scientific Claims Supporting the Endangerment Finding I will now address the Executive Summary scientific claims of the Technical Support Document entitled Endangerment and Cause or Contribute Findings for Greenhouse Gases under Section 202{a) of the Clean Air Act. There are a total of 47 claims grouped in 4 classes, many of which are repeated verbatim, below. As mentioned above, not all claims are disputed, and some will have a common analysis response. 4.1"0bserved Trends in Greenhouse Gas Emissions and Concentrations" (7 claims) Claim #1.1: Greenhouse gases, once emitted, can remain in the atmosphere for decades to centuries, meaning that 1) their concentrations become well-mixed throughout the global atmosphere regardless of emission origin, and 2) their effects on climate are long lasting. The primary long-lived GHGs directly emitted by human activities include carbon dioxide (C02), methane (CHt), nitrous oxide (N20), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF6). Greenhouse gases have a warming effect by trapping heat in the atmosphere that would otherwise escape to space. Analysis: I have no reason to disagree with these statements, but will point out they are qualitative in nature, implying no significant negative impacts, and ignore the large benefits of more C02 in the atmosphere, as will be discussed later. Claim #1.2: In 2007, U.S. GHG emissions were 7,150 teragramst of C02 equivalenh (TgC02eq). The dominant gas emitted is C02, mostly from fossil fuel combustion. Methane is the second largest component of U.S. emissions, followed by NzO and the fluorinated gases (HFCs, PFCs, and SF6). Electricity generation is the largest emitting sector (34% of total U.S. GHG emissions), followed by transportation (28%) and industry ( 19%). Analysis: I have no reason to disagree with these points. Claim #1.3: Transportation sources under Section 202 of the Clean Air Act (passenger cars, light duty trucks, other trucks and buses, motorcycles, and cooling) emitted 1,649 TgC02eq in 2007, representing 23% of total U.S. GHG emissions. Analysis: I have no reason to disagree with these points. Claim #1.4: U.S. transportation sources under Section 202 made up 4.3% of total global GHG emissions in 2005, which, in addition to the United States as a whole, ranked only behind total GHG emissions from China, Russia, and India but ahead of Japan, Brazil, Germany, and the rest of the world's countries. In 2005, total U.S. GHG emissions were responsible for 18% of global emissions, ranking only behind China, which was responsible for 19% of global GHG emissions. 14 Analysis: I have no reason to disagree with these points, but will note that even if the U.S. transportation sector were to no longer produce C02 at all, there would then be only a 4.3% reduction in global GHG emissions, a factor to be kept in mind when evaluating the extent to which any EPA rulemaking regarding GHG emissions can impact the climate system. Claim #1.5: U.S. emissions of sulfur oxides (SOx), nitrogen oxides (NOx), direct particulates, and ozone precursors have decreased in recent decades, due to regulatory actions and improvements in technology. Sulfur dioxide (SOz) emissions in 2007 were 5.9 Tg of sulfur, primary fine particulate matter (PMz.5) emissions in 2005 were 5.0 Tg, NOxemissions in 2005 were 18.5 Tg, volatile organic compound (VOC) emissions in 2005 were 16.8 Tg, and ammonia emissions in 2005 were 3.7 Tg. Analysis: I have no reason to disagree with these points. Claim #1.6: The global atmospheric C02 concentration has increased about 38% from preindustrial levels to 2009, and almost all of the increase is due to anthropogenic emissions. The global atmospheric concentration of CH.l has increased by 149% since pre-industrial levels (through 2007); and the N20 concentration has increased by 23% (through 2007). The observed concentration increase in these gases can also be attributed primarily to anthropogenic emissions. The industrial fluorinated gases, HFCs, PFCs, and SF6, have relatively low atmospheric concentrations but the total radiative forcing due to these gases is increasing rapidly; these gases are almost entirely anthropogenic in origin. Analysis: I have no reason to disagree with these points. Claim #1.7: Historic data show that current atmospheric concentrations of the two most important directly emitted, long-lived GHGs (C02 and CH4) are well above the natural range of atmospheric concentrations compared to at least the last 650,000 years. Atmospheric GHG concentrations have been increasing because anthropogenic emissions have been outpacing the rate at which GHGs are removed from the atmosphere by natural processes over timescales of decades to centuries. Analysis: Even if we assume the claim of highest C02 levels in 650,000 years is true, it should be pointed out that C02 still makes up only a tiny fraction of the atmosphere, as seen in the right panel of the following figure (both panels are mutually consistent, C02 data available from ftp:ljaftp.cmdl.noaa.gov/products/trends/co2/co2 annmean mlo.txt ): 15 While the amount of C02 in the atmosphere has experienced a large increase in relative terms, in absolute terms it still occupies only a tiny fraction. 430 410 100 Atmospheric C02 Concentration (ppm) 90 Atmospheric C02 Concentration (percent) 80 390 70 370 60 50 350 40 330 30 310 20 290 10 270 0 88 N N ~ N Only 4 out of every 10,000 molecules in the atmosphere are C02, which is a very small proportion. Despite the seemingly large yearly emissions numbers EPA claim #1.2 (above), the resulting increase in atmospheric C02 concentration amounts to only 1 in 10,000 molecules of air over the last 100 years. This tiny component of the atmosphere (0.04%) is nevertheless necessary for life to exist on Earth, since photosynthesis on both land and in the ocean is necessary for the food chain. Furthermore, no matter how much C02 humanity produces, an average of 50% of it is removed by nature every year, much of which goes into increased photosynthetic activity. As we will see, below, the positive benefits of more C02 in the atmosphere are not insignificant. Note that if these facts were mentioned, the Endangerment Finding would have had less support. By giving a biased presentation of facts, a maximum amount of alarm can be created. We will see that this is a common theme in the EPA's scientific claims, and in the IPCC's and 16 related organizations' statements: convey a maximum amount of alarm with selective wording, and do not list any positive benefits of GHG emissions which could stand in the way of an Endangerment Finding. 4.2 "Observed Effects Associated with Global Elevated Concentrations of GHGs" (12 claims) Claim #2.1: Current ambient air concentrations of C02 and other GHGs remain well below published exposure thresholds for any direct adverse health effects, such as respiratory or toxic effects. Analysis: I agree. Claim #2.2: The global average net effect of the increase in atmospheric GHG concentrations, plus other human activities (e.g., land-use change and aerosol emissions), on the global energy balance since 1750 has been one of warming. This total net heating effect, referred to as forcing, is estimated to be +1.6 (+0.6 to +2.4} watts per square meter (W/m2}, with much of the range surrounding this estimate due to uncertainties about the cooling and warming effects of aerosols ... The combined radiative forcing due to the cumulative (i.e., 1750 to 2005} increase in atmospheric concentrations of C02, CH4, and N20 is estimated to be +2.30 (+2.07 to +2.53} W/m2. The rate of increase in positive radiative forcing due to these three GHGs during the industrial era is very likely to have been unprecedented in more than 10,000 years. Analysis: Note that compared to the approximate 240 W/m2 average rates of energy flows in and out ofthe climate system (Trenberth et al., 2009), the 0.6 to 2.4 W/m2 imbalance amounts to 0.25% to 1% of the average flows. But even a 1% global radiative imbalance allegedly causing recent warming cannot be reproduced by climate models from physical first principles alone. Instead, all models must be "tuned" in order to produce global energy balance, and those tuning parameters are numerous and their values are not well constrained (Mauritsen et al., 2012). Nor can the anthropogenic energy imbalance be measured from our best satellite energy budget instruments (CERES), as they have insufficient absolute accuracy (Loeb et al., 2018). It is a theoretical calculation which ignores the fact that a chaotic climate system can create its own energy imbalances. Even if we could accurately measure the Earth's radiative 17 energy imbalance, there would be no way to tell how much of it is due to anthropogenic versus natural forcings. The claim that "the rate af increase in positive radiative forcing due to these three GHGs during the industrial era is very likely to have been unprecedented in more than 10,000 years" cannot be supported because we have no idea whether the climate system has been stable to less than 1% radiative energy imbalance in the past. This is purely a statement of faith based upon the assumption that the climate system never changes; it is an argument from ignorance. As further evidence, warming of the global oceans since the 1950s, if the Argo floats' measurements are believed, represent an energy imbalance of only 0.4 W/m2, which is only a 0.17% imbalance (about 1 part in 600) in the rates of energy flow in and out of the climate system (Levitus et al., 2012). The absolute accuracy ofthe Earth's energy imbalance from the CERES satellite instruments is ten times worse than this (Loeb et al., 2018). While the deepocean warming more recently (since 2005) has been estimated to be equivalent to an energy input of 10 23 Joules (a very large number), what the Argo floats actually measure are temperature, and the heating of the oceans between 2005 and 2017 is based upon only a 0.04 deg. C average temperature increase in those 12 years, a 1 part in 260 energy imbalance (about 0.9 W/m2). What this means in practical terms is that the theory of anthropogenic climate change remains just a theory, supported by models that are built upon the assumption that C02 is the main driving force of climate change. The warming of the oceans represents a smaller global energy imbalance than can be measured by satellite, modeled by climate models, or concluded to be larger than Mother Nature can generate just through chaotic changes in the ocean circulation. Claim #2.3: Warming of the climate system is unequivocal, as is now evident from observations of increases in global average air and ocean temperatures, widespread melting of snow and ice, and rising global average sea level. Global mean surface temperatures have risen by 1.3 :1: 0.32 ?F (0.74?C :1: 0.18 ?C) over the last 100 years. Eight of the 10 warmest years on record have occurred since 2001. Global mean surface temperature was higher during the last few decades of the 20th century than during any comparable period during the preceding four centuries. 18 Analysis: While warming has very likely occurred since the mid-20th Century, and we might even be warmer than "any comparable period during the preceding four centurie.s", this is fully consistent with our emergence from the Little Ice Age of centuries past. There is temperature proxy evidence (30 proxies from around the Northern Hemisphere) of unusual cold in recent centuries, and that the most recent historical warming trend began in the 1700s, before increasing C02 could have been blamed (Ljungqvist, 2010): 1.0 0.8 . ._ ~~ 06 ? 0.4 -2 E .s ~ 1'0 c:c: 0 1910 1930 19SO 1970 1990 2010 Time [yr] Fig. 1. Time series and rates of GMSL during the period 1902-2012 . (A) Revised GMSL reconstruction based on 322 tide gauges in comparison with previous estimates (CW11 =ref. 1; ROll =ref. 2, J14 =ref. 5; HlS =ref. 6) and modeling attempts based on historical CMIP5 models (12). The gray shading marks the ln errors of the final reconstruction. The dotted black line represents a GMSL reconstruction with all VLM and geoid corrections, but without methodological adjustments such as area weighting and the use of a common mean. (8) The corresponding rates calculated with a singular spectrum analysis using an embedding dimension of 15 y. While the authors curiously claim the revised observational data (black line in the above figure) shows evidence of a recent acceleration of sea level rise, the blue arrow points to pre1950 rise that looks the same as the post-1990 rise. 27 The components of sea level rise (the sea level budget) are still not known well enough to explain the recent rise; for example Rietbroek et al., 2016 compared the satellite-measured rise during 2002-2014 to the components from thermal expansion, glacier changes, and the hydrologic inputs and losses (rivers, precipitation, evaporation), and found that satelliteobserved sea level is rising faster than can be explained. Until the observations and our physical understanding of them improve, it is doubtful that we can trust future model projections of sea level rise. In summary, we do not know just how much of recent sea level rise is natural versus human-caused. This is analogous to the situation with recent global average temperature rise. The EPA's emphasis on only human causation shows clear scientific bias. The implied claim that sea level rise is human-caused is deceptive without acknowledgement that sea level was rising well before humans could be blamed. At least the claim regarding recent acceleration in sea level rise is admitted to be uncertain as to the cause, since similar episodes of acceleration have occurred before. Claim #2.8: Satellite data since 1979 show that annual average Arctic sea ice extent has shrunk by 4.1% per decade. The size and speed of recent Arctic summer sea ice lass is highly anomalous relative to the previous few thousands of years. Analysis: While the Arctic sea ice extent since satellite monitoring began in 1979 has indeed shrunk, once again we have no idea how much of this is due to natural factors versus human-caused. The claim of the shrinking be "anomalous relative to the previous few thousand years" is totally speculative. We have evidence of past climate change, especially warm periods, in the Arctic. Here are excerpts from the 1943 book Arctic Ice by the original expert on Arctic sea ice, N.N. Zubov: ? "... in late years a most interesting phenomenon has been observed- a warming of the Arctic, as evidence by a gradual and universal decrease in ice abundance. The main evidence of this general warming of the Arctic are: "Receding of glaciers and "melting away" of islands ... Ahlman terms the rapid receding of the Spitzbergen glaciers "catastrophic". 28 "Rise of air temperature. (Over the last 20 years) the average temperature of the winter months has steadily increased... "Rise in temperature of Atlantic water which enters the Arctic Basin ... the temperature of surface water and of Gulf Stream water has steadily risen ... "Decrease in ice abundance .... 15% to 20% (over 20 years) ... "Change in cyclone routes. There is no doubt that the increase in air temperatures, increase in Atlantic water temperatures, intensification of ice drift, etc., are closely connected with an intensification of atmospheric circulation, and in particular with a change in cyclonic activity at high latitudes. "Biological signs of warming of the Arctic....fish have ranged further and further to the north ... cod in large quantities have appeared along the shores of Spitzbergen and Novaya Zemlya ... also mackerel, dolphin where formerly were not found ... during recent years fishing has gradually shifted into the Arctic waters, and this unquestionably must be ascribed in considerable degree ta the warming of these waters .... "Ship navigation .... a number of ship voyages (were made) which could hardly have been accomplished in the preceding cold period. "Still more remarkable is the fact that the warming of the Arctic is not confined to any particular region." These comments sound like they could have come from recent news reports of warming in the Arctic, and suggest that natural changes in Arctic climate and sea ice on multi-decadal time scales are indeed possible, and have occurred in the recent past, before we had satellites to monitor this remote area (before 1979). In fact, there is considerable evidence to suggest that satellites began monitoring sea ice at the end of an unusually cold period, possible hundreds of years, during which Arctic sea ice cover slowly grew, and that in the near-40 year period since we might be seeing a return to more "normal" conditions. A number of published studies have presented evidence of this. I will highlight only the most recent one (Moffa-Sanchez & Hall, 2017). A 3,000 year reconstruction of Arctic sea ice from ocean sediment records of past ocean salinity variations (which are a proxy for water temperature) suggest that Arctic sea ice peaked during the 1800s, after a lengthy Little Ice Age lasting at least 300 years: 29 Comment submitted by Orla E. Collier, Benesch Friedlander Coplan Er Aronoff LLP on behalf of Murray Energy Corporation (Appendix part 2 of 2) The is a Comment on the Envlronmental Protection Agency (EPA) Proposed Flule: Repeal of Carbon Follutlon Emleelon Guidelines tor Stationary Sources: Utillty Generatlng Unlts; Fleopenlng of the Comment Ferlod ID: For related information, Open Docket Folder El Tracking Number: ITweet ?Share .Email Document Information Comment Please see attached Part 2 of 2 of Appendix to Murray Energy Corporation Comments. Thank you Date Posted: May 11. 2mg FIIN: Attachments (1) Not ASSIgned Show More Detalls .- Appendix Part 2 of 2 Vlew Attachment: Arctic Sea Ice Extent 4.0 (North of Iceland) 20th 3.5 3.0 2.5 Period of satellite record 2.0 1.5 Warm Period 1.0 to little Ice Age 500 1000 caen - CD 0. 0.5 Medieval Warm Period 0 ~ (Q Roman Warm Period 1500 Years BP 2000 0.0 2500 3000 modified from Moffa?S~nchez and Hall2017 Note that "today" is at the left end of the graph, and progressing to the right goes back in time. The ocean sediment data suggests that the Little Ice Age experienced cooling of the Arctic unmatched in the previous 3,000 years. The implied increase in sea ice began to subside with the warming of the 20th Century. The authors attribute the changes to natural fluctuations in ocean currents. In its claim regarding sea ice, the EPA makes no mention of published research work which would contradict its claim that recent Arctic sea ice decreases are unusual in the context of the last few thousand years, when such evidence indeed exists. While the above evidence from ocean sediment data is too recent to have been included in the EPA's analysis, it shows how scientific knowledge can change as more information becomes available, and points out the need to revisit the science claims underpinning the Endangerment Finding. 30 Claim #2.9: Widespread changes in extreme temperatures have been observed in the last 50 years across o/1 world regions, including the United States. Cold doys, cold nights, ond frost have become Jess frequent, while hot days, hot nights, and heat waves have become more frequent. Analysis : At a minimum, the claim is a gross exaggeration. Regarding the U.S., the main concern would be excessive heat (since less excessive co ld would be a welcome thing). For 1, 114 USHCN stations in the United States, here are the average numbers of days each year that a station exceeded 100 deg. F and 105 deg. F temperatures, from 1895 th rough 2017, as tabulated from official NOAA data by John Ch rist y (UAH): Average per station (1114 USHCN Stations) 1895-2017 Number of days dally Maximum temperature above 100oF and 1osoF 20 No significant Trends. 11 of 12 hottest yeJrs occurred before 1960 18 16 14 12 10 8 6 4 2 D 1195 1905 1915 1915 1935 1945 1955 1965 1975 1915 1995 1005 2015 As can be seen, there is no obvious t rends in very hot days, which would be the main concern . In fact, 11 of the 12 years with the largest number of very hot days occurred before 1960. As previously explained, I have concern that all land-based thermometer data have spurious warming effects from manmade structu res replacing natural vegetation, and active heat sources, leading to an Urban Heat Islan d (UH I) effect. It is not clear how well this has been adjust ed for, and there is evidence that in the U.S. warming in recent decades has been exaggerated by as much as a factor of 2 (Watts et at., 2010). That urban areas are warmer than rural areas is a common everyday experience. To demonstrate the point, I quantified this effect 31 using pairs of temperature reporting stations around the world that were within 150 km of each other and had different population densities. The results from the year 2000 show what others have found: that spurious warming at a thermometer site rapidly increases at the lowest population densities: 150 km max spacing Year= 2000 (adds 724 station pairs where both are In the lowest pop. Density category) 1.6 1.5 1.4 1.3 -1.2 0 til 1.1 ~ 1.0 ~ 0.9 co E 0.8 . : 0.1 c 0.6 0 i 0.5 C/) 0.4 ... 0.3 0.2 0.1 0.0 0 50 100 150 200 Population Denslty(per sq. km) Half of the stations used in this analysis had population densities below 100 persons per sq. km, and spurious warming hit a maximum of 2.2 deg. C for a population density of 7,000 persons per km. Interestingly, even at only 20 persons per sq. km, the average UHI effect was almost 0.8 deg. C of spurious warming. There is no easy way to remove this effect from the thermometer data. Unfortunately, the character of the UHI effect looks just like global warming: a gradual warming with time. Population density is a convenient, but imperfect, proxy for the UHI effect, because even with the same population density, wealth increases over time causing still further warming due to addition of more and larger buildings, roads, parking lots, and an increase in active heat sources such as air conditioning exhaust, jet engine activity at airports, etc. Using the above chart, and knowing that world population has increased by a factor of 5 since the late 1800s when 32 thermometer records began, we can roughly estimate that the global land thermometer network has experienced about 0.5 deg. C of spurious warming over the last century or so. It is doubtful that this effect has been removed from the data used by the IPCC and thus the EPA. Given the difficulty, the best methodology would probably be to simply throw away all data with substantial UHI effects, and limit land data analyses to those stations which have remained relatively rural. Instead, the NOAA methodology (upon which the IPCC and EPA claims for the U.S. are based) uses a convoluted "homogenization" technique which, in my opinion, is correcting rura l data to look like urban data, rather than the other way around as desired. Until a detailed analysis of the UHI effect on global land temperature trends is made, it is not clear that any claims can be seriously defended about record high temperatures or whether heat waves have increased. And, as we have seen, even using NOAA's own data the case can be made that there has been no long-term trend in the number of very hot days in the U.S. Finally, even if warming has occurred, it remains very uncertain just how much of it was due to human versus natural influences. Claim 112.10: Observational evidence from all continents and most oceans shows that many natural systems are being affected by regional climate changes, particularly temperature increases. However, directly attributing specific regional changes in climate to emissions of GHGs from human activities is difficult, especially for precipitation. Analysis: At least for this claim, the EPA has acknowledged little confidence in the claim regarding the effects of GHG emissions on natural systems (e.g. forests, croplands). This is wise, because multiple studies using several decades of satellite data have now established a "global greening" has occurred, most likely the result of the positive benefits of increased C02 levels on plant growth (Donohue et al., 2013): 33 ..... 30% 20% 10% ?- o? -10% -20% Thus, in contrast to the alleged negative effects of temperature increases and precipitation changes made in the EPA's claim, it appears that the direct benefits of C02 enrichment on plant growth, drought tolerance, and water use efficiency have outweighed the negative effects. Literally hundreds of published scientific studies supporting this are tabulated at http://co2science.org . The fact that the EPA ignored this evidence reveals once again a bias toward science supporting a chosen policy outcome. Claim #Z.ll: Ocean C02 uptake has lowered the average ocean pH (increased acidity) level by approximately 0.1 since 1750. Consequences for marine ecosystems can include reduced calcification by shell-forming organisms, and in the longer term, the dissolution of carbonate sediments. Analysis: First of all, the wording "increased acidity" is deceptive since the oceans are solidly basic or alkaline {currently pH=8.1) and there is no forecast that they will ever become acidic. The oceans are believed to have become slightly less basic, from an estimated preindustrial value of pH= 8.2. Secondly, the negative effect of this small decrease in pH on marine life has mixed results in the scientific literature, as the extra C02 dissolved in seawater {like on land) promotes photosynthesis of phytoplankton at the beginning of the marine food chain. Studies of reduced pH on marine life are often done with very abrupt {hours to days) changes in pH, and with much larger magnitudes than will ever be experienced, even if we burn all known and suspected fossil fuel reserves. This does not allow the marine organisms to adapt to the changes, which biases the results in the direction of damage to the organisms. 34 Craig ldso, Ph. D., has surveyed the literature on the subject for decades, and his summary of the published results (http:Uwww.co2science.org/data/acidiflcation/results.php), although now somewhat out of date, suggests that the net effect of more C02 dissolved in seawater will have a net benefit on sea life, including corals. He recently made the following statement to me, "there is no scientific basis to support claims of impending marine life catastrophe due to ocean acidification. Rather, the predicted decline in oceanic pH (if it occurs) will have little to no lasting negative impacts on the bulk of marine life." So, while I do not qualify as an expert on this subject, "ocean acidification" is one more area where the science needs to be revisited in a more balanced way. Claim #2.12: Observations show that climate change is currently affecting U.S. physical and biological systems in significant ways. The consistency of these observed changes in physical and biological systems and the observed significant warming likely cannot be explained entirely due to natural variability or other confounding non-climate factors. Analysis: This claim is similar to Claim #10, and much of that analysis applies here. The claim that "the observed significant warming likely cannot be explained entirely due to natural variability or other confounding non-climate factors" is not much more than a faith-based assertion, arguing from ignorance, since the EPA and the IPCC have no idea just how much of recent warming is human-caused versus natural, since natural sources of climate change (e.g. our emergence from the Little Ice Age) are poorly understood. In any event, the claim does not say whether the changes are positive or negative ...what are we to assume? Also, we have already established under Claim #10 that there has been a net greening of the United States in response to the direct positive effects C02 has on vegetation. Furthermore, it is likely that these benefits are at least partly responsible for increasing corn yields in the United States, which have been on the same upward trend since approximately 1960, with all four ofthe most recent growing seasons (2014, 2015, 2016, 2017) experiencing higher corn yields than in any previous year (corn yield data are availab le from a wide variety of sources): 35 ,-I 225 United States Corn Yield 200 1930-2017 175 B shelslacre 150 - USDA?NASS Official Yield - Yield Trend 1930-1959 125 100 -r 1 r I T l _L! I --r I r I I l --iI 75 - Yield Trend 1960?2014 50 25 0 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 Even if some areas of the United States are experiencing negative changes, it would be difficult establish causation since ecosystems are continually changing in response to changes in regional weather patterns, and (as discussed under Claims #5, 6, and 9) the natural temperature and precipitation changes in the United States can be large. The best example of this is the Dust Bowl of the 1930s, which occurred well before GHGs could be blamed, and is generally considered to be the result of natural changes in the circulation of the Pacific Ocean; the NOAA Drought Task Force Assessment Report has a list of many publications (http://ocp.ldeo .columbia.edu/ res/div/ocp/ drought/publications.shtml ), the majority of which deal with natural factors, rather than human-related, as the cause of North American drought episodes. On a global scale, ldso (2013) estimated the global increase in agricultural productivity due to the direct effects of increasing C02 on plant growth, and arrived at a net benefit of US$3.5 Trillion. Thus, the direct benefits of increasing C02 associated with climate change were deceptively left out of this claim by the EPA. 36 4.3 "Projections of Future Climate Change with Continued Increases in Elevated GHG Concentrations" (9 claims) In this class of claims, we move from observational evidence (or lack thereof) of climate change with assertions of causation of those changes, to projections of future climate states based upon computerized climate models that assume various future atmospheric increases in GHGs, human-ca used aerosol pollution, land use changes, etc. The primary driver of climate change in the models is the assumed increases in atmospheric C02, w hich indeed should have (at least based upon theory) some warming effect. Claim #3.1: Most future scenarios that assume no explicit GHG mitigation actions (beyond those already enacted) project increasing global GHG emissions over the century, with climbing GHG concentrations. Carbon dioxide is expected to remain the dominant anthropogenic GHG over the course of the 2lsc century. The radiative forcing associated with the non-C02 GHGs is still significant and increasing over time. Analysis : The claim is basically that C02 concentrations in the atmosphere are expected to slowly increase, and so will the radiative forcing from the extra C02. I have no objection to this statement. Claim #3.2: Future warming over the course of the 21st century, even under scenarios of lowemission growth, is very likely to be greater than observed warming over the past century. According to climate model simulations summarized by the IPCC, through about 2030, the global warming rate is affected little by the choice of different future emissions scenarios. By the end of the 21 st century, projected average global warming (compared to average temperature around 1990) varies significantly depending on the emission scenario and climate sensitivity assumptions, ranging from 3.2 to 7.2 deg F ( 1.8 to 4.0 deg. C), with an uncertainty range of 2.0 to 11.5 deg. F ( 1.1 to 6.4 de g. C). Analysis: There is now a great deal of published evidence that the amount of future warming projected by the models will be too large. The claim, rephrased, is that warming in the 21st Century will accelerate, that is, the rate of warming will be greater than in the 20th Century. But, to date, the models have produced approximately twice t he amount of atmospheric warming as has been observed since 1979, which is when we have had our first capability to monitoring the tropospheric temperature over virtually the entire Earth: 37 Tropical Mid-Tropospheric Temperature Variations Models vs. Observations 1.2 S?Year Averages, 1919-2017 Trend line crosses zero at 1979 fM all tfme series .uassu Aa:USI.O ICC 0.4 IPSlt k\'=u\ N M.u~- l ' pd.ucd 1 clxu..u\ :H.. ~.u-: - 1.! ntOnth Nn~ 1unu 70 GO 50 40 .30 20 10 II ~~~~~~~--~~~~~~~~~~~~~~~~~~~~~~~~ 70 72 74 76 78 80 82 84 86 88 90 92 94 9G 98 00 02 04 OG 08 10 12 I.J IG 18 Claim #3.8: By the end of the century, global average sea level is projected by IPCC to rise between 7.1 and 23 inches (18 and 59 centimeter [em]), relative to around 1990, in the absence of increased dynamic ice sheet loss. Recent rapid changes at the edges of the Greenland and West Antarctic ice sheets show acceleration of flow and thinning. While an understanding of these ice sheet processes is incomplete, their inclusion in models would likely lead to increased sea level projections for the end of the 21" century. Analysis: As discussed in the analysis of Claim #2.7, it is not known how much of recent sea level rise is due to natural versus human-caused warming. Sea levels have been rising at least since the mid-1800s at about the same rate as today, with some evidence it might be accelerating, but it is unknown whether this trend will continue or is just temporary. The components of sea level rise (the sea level budget) are still not known well enough to explain the recent rise. The claims of a potential acceleration due to rapid loss of ice from the Greenland and West Antarctic ice sheets are extremely speculative. As addressed in the analysis 42 of claim #2.3, glacier retreat in Alaska reveals an ancient forest grew there 1,000 to 2,000 years ago, indicating century-time scale changes in glacier input to sea level rise can occur naturally. Claim #3.9: Sea ice extent is projected to shrink in the Arctic under all IPCC emissions scenarios. Analysis: This claim, once again, entirely depends upon demonstrably flawed climate model predictions. As discussed in the analysis of claim 2.8, there is abundant evidence of warm Arctic conditions 1,000-2,000 years ago, as well as early in the 20th Century. Until such natural climate fluctuations are understood and predicted, there is no way to know how much of Arctic sea ice change is natural, or how much will occur in the future. 4.4 "Projected Risks and Impacts Associated with Future Climate Change" (19 claims) Claim #4.1: Risk to society, ecosystems, and many natural Earth processes increase with increases in both the rate and magnitude of climate change. Climate warming may increase the possibility of large, abrupt regional or global climatic events (e.g., disintegration of the Greenland Ice Sheet or collapse of the West Antarctic Ice Sheet). The partial deglaciation of Greenland (and possibly West Antarctica) could be triggered by a sustained temperature increase of 2 to 7 deg. F (1 to 4 deg. C) above 1990 levels. Such warming would cause a 13 to 20 feet (4 to 6 meter) rise in sea level, which would occur over a time period of centuries to millennia. Analysis: Taken literally, this claim says that a 1 deg. C sustained tem perature above 1990 levels could cause a 4 to 6 meter rise in sea level. The claim of such a catastrophic response to only a 1 deg. C change is, in my opinion, wildly speculative, alarmist, and scientifically irresponsible. Claims #4.2, 4.3, 4.4: CCSP reports that climate change has the potential to accentuate the disparities already evident in the American health care system, as many of the expected health effects are likely to fall disproportionately on the poor, the elderly, the disabled, and the uninsured. IPCC states with very high confidence that climate change impacts on human health in U.S. cities will be compounded by population growth and an aging population. Severe heat waves are projected to intensify in magnitude and duration over the portions of the United States where these events already occur, with potential increases in mortality and morbidity, especially among the elderly, young, and frail. Some reduction in the risk of death related to extreme cold is expected. It is not clear whether reduced mortality from cold will be greater or less than increased heat-related mortality in the United 43 States due to climate change. Analysis: The climate change portion of these claims are inconsistent with more recent research that shows cold weather kills twenty times as many people as does hot weather (Gasparrini et al., 2015). Claim #4.5: Increases in regional ozone pollution relative to ozone levels without climate change are expected due to higher temperatures and weaker circulation in the United States and other world cities relative to air quality levels without climate change. Climate change is expected to increase regional ozone pollution, with associated risks in respiratory illnesses and premature death. In addition to human health effects, tropospheric ozone has significant adverse effects on crop yields, pasture and forest growth, and species composition. The directional effect of climate change on ambient particulate matter levels remains uncertain. Analysis: Since I do not follow ozone chemistry research, I have no expert opinion on this claim. Claim #4.6: Within settlements experiencing climate change, certain parts of the population may be especially vulnerable; these include the poor, the elderly, those already in poor health, the disabled, those living alone, and/or indigenous populations dependent on one or a few resources. Thus, the potential impacts of climate change raise environmental j ustice issues. Analysis: Since this claim has nothing to do with science, so I have no expert opinion . Claim #4. 7, 4.8, 4.9: CCSP concludes that, with increased C02 and temperature, the life cycle of grain and oilseed crops will likely progress more rapidly. But, as temperature rises, these crops will increasingly begin to experience failure, especially if climate variability increases and precipitation lessens or becomes more variable. Furthermore, the marketable yield of many horticultural crops (e.g., tomatoes, onions, fruits) is very likely to be more sensitive to climate change than grain and oilseed crops. Higher temperatures will very likely reduce livestock production during the summer season in some areas, but these losses will very likely be partially offset by warmer temperatures during the winter season. Cold-water fisheries will likely be negatively affected; warm-water fisheries will generally benefit; and the results for cool-water fisheries will be mixed, with gains in the northern and losses in the southern portions of ranges. Analysis: I have no expertise to critique these claims. 44 Claim #4.10: Climate change has very likely increased the size and number of forest fires, insect outbreaks, and tree mortality in the interior West, the Southwest, and Alaska, and wiU continue to do so. Over North America, forest growth and productivity have been observed to increase since the middle of the 20"' century, in part due to observed climate change. Rising C02 will very likely increase photosynthesis for forests, but the increased photosynthesis will likely only increase wood production in young forests on fertile soils. The combined effects of expected increased temperature, C02, nitrogen deposition, ozone, and forest disturbance on soil processes and soil carbon storage remain unclear. Analysis: It is interesting that the positive effects of C02 on photosynthesis are emphasized here, in the case of wildfires, where more wood production would result in more fuel for wildfires. Why were the positive benefits of increased photosynthesis on agriculture not mentioned, say under claim 2.12? This shows a clear bias towards only negative effects, while ignoring positive effects, of GHG emissions. The long-term trend in U.S. wildfi res is greatly affects by land use practices, as can be seen in these data from the National Interagency Fire Center: US Forest Area Burned 1926-2017 so 40 2017 one-fifth of record~ 10 ooo 111111oo.1111.11oo1.11.1111111111111111111 0 45 During the early part of the 20th Century, we started extinguishing wildfires due to the threat they posed to human settlements. Then, in 1972, a " let burn" policy was started as a way for natural wildfires (especially in national park areas) ignited by dry lightning strikes to burn the excess woody fuel which had built up over several decades. This led to the catastrophic Yellowstone fire during the drought of 1988, during which the "let burn" policy was temporarily rescinded by President Reagan . In any event, it can be seen from the above figure that the natural state of forests in the U.S. is to burn more than is currently allowed. As long as we put out wildfires, we allow more fuel to build up, and later fires to become worse. This is a land use policy issue, not a climate change issue. Claim #4.11: Coastal communities and habitats will be increasingly stressed by climate change impacts interacting with development and pollution. Sea level is rising along much of the U.S. coast, and the rate of change will very likely increase in the future, exacerbating the impacts of progressive inundation, storm-surge flooding, and shoreline erosion. Storm impacts are likely to be more severe, especially along the Gulf and Atlantic coasts. Salt marshes, other coastal habitats, and dependent species are threatened by sea level rise, fixed structures blocking landward migration, and changes in vegetation. Population growth and rising value of infrastructure in coastal areas increases vulnerability to climate variability and future climate change. Analysis: Again, the claim implies that sea level rise is human-caused, and thus controllable through EPA rulemaking. But as has been addressed above, global sea level has been rising naturally since before it could have been due to GHG emissions (prior to about 1940). No mention is made of the natural process of land subsidence in some coastal areas (e.g. Norfolk, VA; Miami Beach, Fl) which makes coastal sea level rise even worse. Neglecting these issues makes the claim deceptive from the standpoint of supporting the Endangerment finding. Claim #4.12: Climate change will likely further constrain already overallocated water resources in some regions of the United States, increasing competition among agricultural, municipal, industrial, and ecological uses. Although water management practices in the United States are generally advanced, particularly in the West, the reliance on past conditions as the basis for current and future planning may no longer be appropriate, as climate change increasingly creates conditions well outside of historical observations. Rising temperatures will diminish snowpack and increase evaporation, affecting seasonal availability of water. In the Great Lakes and major river systems, lower water levels are likely to exacerbate challenges relating to water quality, navigation, recreation, hydropower generation, water transfers, and binational relationships. Decreased water supply and lower water levels are Hkely to exacerbate challenges relating to aquatic navigation in the United States. 46 Analysis: It is not known to what extent precipitation changes in the U.S. will occur. As discussed in the analysis of Claim #2.6, the observed trend in U.S. precipitation has been upward, not downward, and expert discussions of drought in North America {e.g. the 1930s Dust Bowl) focus on natural causation, such as the Pacific Decadal Oscillation. Regarding the Great Lakes, data from the Great Lakes Environmental Research Lab (https://www.glerl .noaa.gov/ data/ dashboard/ data/ levels/1918 PRES/) reveals there has been no decline in water levels over the last century: Great Lakes Water Levels Show No Signs of a Decline 178.0 lakes Michigan-Huron Annual Average Water level {meters) 1920 through 2017 177.5 177.0 176.5 176.0 175.5 175.0 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 J Thus most of what is stated in this claim is highly speculative with little evidence any of it has begun to happen. Claim #4.13: Higher water temperaturest increased precipitation intensityt and longer periods of low flows will exacerbate many forms of water pollution, potentially making attainment of water quality goals more difficult. As waters become wanner, the aquatic life they now support will be replaced 47 by other species better adapted to warmer water. In the long term, warmer water and changing flow may result in deterioration of aquatic ecosystems. Analysis: This claim depends upon the predictions of future warming, which has already been covered. Claim #4.14: Ocean acidification is projected to continue, resulting in the reduced biological production of marine caJcifiers, including corals. Analysis: This claim has been covered under claim #2.11. Claim #4.15: Climate change is likely to affect U.S. energy use and energy production and physical and institutional infrastructures. It will also likely interact with and possibly exacerbate ongoing environmental change and environmental pressures in settlements, particularly in Alaska where indigenous communities are facing major environmental and cultural impacts. The U.S. energy sector, which relies heavily on water for hydropower and cooling capacity, may be adversely impacted by changes to water supply and quality in reservoirs and other water bodies. Water infrastructure, including drinking water and wastewater treatment plants, and sewer and stormwater management systems, will be at greater risk of flooding, sea level rise and storm surge, low flows, and other factors that could impair performance. Analysis: The lack of evidence for human-caused changes in precipitation and sea level rise has already been addressed . Claim #4.16: Disturbances such as wildfires and insect outbreaks are increasing in the United States and are likely to intensify in a warmer future with wanner winters, drier soils, and longer growing seasons. Although recent climate trends have increased vegetation growth, continuing increases in disturbances are likely to limit carbon storage, facilitate invasive species, and disrupt ecosystem services. Analysis: This is a partial repeat of what was contained in claim #4.11, and shows a lack of sufficient editing by the document's authors. Claim #4.17: Over the 21st century, changes in climate will cause species to shift north and to higher elevations and fundamentally rearrange U.S. ecosystems. Differential capacities for range shifts and constraints from development, habitat fragmentation, invasive species, and broken ecological connections will alter ecosystem structure, function, and services, and 48 Claim #4.18: Climate change impacts will vary in nature and magnitude across different regions of the United States. ? Sustained high summer temperatures, heat waves, and declining air quality are projected in the Northeast, Southeast, Southwest, and Midwest. Projected climate change would continue to cause loss of sea ice, glacier retreat, permafrost thawing, and coastal erosion in Alaska. ? Reduced snowpack, earlier spring snowmelt, and increased likelihood of seasonal summer droughts are projected in the Northeast, Northwest, and Alaska. More severe, sustained droughts and water scarcity are projected in the Southeast, Great Plainss, and Southwest. ? The Southeast, Midwest, and Northwest in particular are expected to be impacted by an increased frequency of heavy downpours and greater flood risk. ? Ecosystems of the Southeast, Midwest, Great Plains, Southwest, Northwest, and Alaska are expected to experience altered distribution of native species (including local extinctions), more frequent and intense wildfires, and an increase in insect pest outbreaks and invasive species. ? Sea level rise is expected to increase storm surge height and strength, flooding, erosion, and wetland loss along the coasts, particularly in the Northeast, Southeast, and islands. ? Warmer water temperatures and ocean acidification are expected to degrade important aquatic resources of islands and coasts such as coral reefs and fi sheries. ? A longer growing season, low levels of warming, and fertilization effects of carbon dioxide may benefit certain crop species and forests, particularly in the Northeast and Alaska. Projected summer rainfall increases in the Pacific islands may augment limited freshwater supplies. Cold-related mortality is projected to decrease, especially in the Southeast. In the Midwest in particular, heating oil demand and snow-related traffic accidents are expected to decrease. Analysis: Again, these claims depend upon climate models projections of warming and precipitation changes, which we have already seen are seriously biased and need to be reevaluated by the EPA. Claim #4.19: Climate change impacts in certain regions of the world may exacerbate problems that raise humanitarian, trade, and national security issues for the United States. The IPCC identifies the most vulnerable world regions as the Arctic, because of the effects of high rates of projected warming on natural systems; Africa, especially the sub-Saharan region, because of current low adaptive capacity as well as climate change; small islands, due to high exposure of population and infrastructure to risk of sea level rise and increased storm surge; and Asian mega-deltas, such as the Ganges-Brahmaputra and the Zhujiang, due to large populations and high exposure to sea level rise, storm surge and river flooding. Climate change has been described as a potential threat multiplier with regard to national security issues. Analysis: Again, the projected geopolitical impacts of climate change are speculative, and depend upon the assumption that climate change will continue as projected and is largely human-caused. These claims and assumptions have already been addressed. 49 5. Clouds, Water Vapor, and Climate Sensitivity Given the importance of climate models to the Endangerment Finding and CPP, it is important to cover a couple issues not discussed in detail, above. The modelers themselves admit that models must be "tuned" to the observations because there are portions of the physics which are not that well understood (Mauritsen et al., 2012). These tunings of adjustable parameters can dominated the amount of climate change the model produces. Model deficiencies in one area sometimes have to be adjusted for in an ad hoc manner by tuning other uncertain parameters in the models. So, while much of the physics contained in climate models is well understood, there is very poor understanding of some of the key processes associated with climate change, for example how clouds or precipitation efficiency (and thus the Earth's greenhouse effect) will change with an imposed warming tendency. 5.1 Cloud feedback How clouds will change in response to the weak direct warming influence of increasing C02 (cloud feedback) is still very uncertain. It is possible that cloud feedbacks are negative, and will reduce future warming. As demonstrated by Spencer & Braswell, 2011, in the presence of natural climate fluctuations which have a radiative forcing component, cloud feedback diagnosed from satellite observations can look like they are weakly positive when in fact they are strongly negative. (Most climate models exhibit positive cloud feedback.) This is a large source of uncertainty in climate model projections. 5.2 Precipitation efficiency and water vapor feedback Another example of such an uncertainty is the greenhouse effect of water vapor, which is the atmosphere's main greenhouse gas. While global warming would likely lead to more atmospheric water vapor in total, this does not mean that water vapor feedback is necessarily positive (Spencer & Braswell, 1997). The greenhouse effect of water vapor is mostly controlled by the very small amounts of water vapor well above the relatively humid near-surface boundary layer where most water vapor resides. That "free tropospheric" vapor in turn depends upon the fraction of cloud water that precipitation systems remove as precipitation . The fraction is called "precipitation efficiency", and we do not even understand the processes so which control it, leading to huge uncertainty about whether climate models can be relied upon at all to forecast future climate states (Renno et al., 1994; Zhao et al., 2016). In some models, the fraction is simply set to a constant, even though we have no idea what the true value should be, or whether it even is a constant. The lack of a tropical"hotspot" in the upper troposphere is, in my opinion, evidence that water vapor feedback is not nearly as strong in nature as in models. In the annual AMS Summary of the Climate in 2016 (Biunden & Arndt, 2017), this figure reveals the current discrepancy between models and observations for temperature trends (1979-2016) as a function of height in the troposphere. Tropical Temperature Trends 50 ITMT Layer Tre~d II 20. S-20oN, 1979-2016 : 0 RICH I I 0 RAOBCORE 0 RATPACA 0 UNSW L--o I I - 'ii 5 MeanRaob -9- Mean Model ~ o en 'ij o _.411> + 250 2! :;) en en + 2! Wannest 2.5% Coolest 2.5% UAHv60 RSSv40 NOAAv4 0 + UW ~ X ERA-I X MERRA-2 X JRA-55 500 700 1000 -0 1 0 0.1 04 02 oc decade-o OS f1G. 52.10. Temperature trends (?C decade-o) for the tropics at various tropospheric levels for 1979-2016. The top box indicates trends for the tropical atmospheric layer known as the midtroposphere (THT) and includes satellite observations. The CH IP-S (rcp