Climate Resilience and Benefit–Cost Analysis: A Handbook for Airports (2019)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

147 1. The Port Authority of New York and New Jersey is aware of the potential for disruptions at LGA due to flood risk and has already undertaken multiple initiatives to address them. 2. The water heights shown are measured relative to a baseline level known as the mean higher high water vertical datum; this is discussed further in Appendix D. 3. It is important to note that all ACROS projections are based on a future climate scenario known as RCP8.5; see the discussion in Appendix D for more information about climate scenarios. Also, each of the climate vector elements is tied to specific quantitative definitions (e.g., “hot days” refers to days when the maximum temperature reaches at least 90°F). The definitions are presented in more detail in Exhibit 3-3. 4. An obvious limitation here is that ACROS only provides projections for two future years, so if doing a standard analysis based on annual data, the analyst would have to interpolate between the three available years to develop annual projections for the incidence of very hot days. Another limitation is that the climate stressors are predefined assuming specific thresholds that may not be relevant for a particular airport. 5. This is potentially complicated because the analyst might realistically want to consider future growth at the airport as well as changes in the future fleet and whether that fleet would be more or less affected by very hot days than the current fleet. 6. Ideally the analyst would check with more recent climate forecasts described in Chapter 3 to ensure that the ACROS maximum forecasts are still representative of a worst-case scenario. 7. This assumes that airlines pay the full crew cost associated with each cancelled flight, plus incur an aircraft depreciation cost; the latter is a rough estimate of the opportunity cost of the aircraft being out of service due to the cancellation. 8. This result may not be surprising given that most of the benefits occur well into the future as the number of very hot days grows. As is discussed further in Appendix E, the results for long-lived projects are often dependent on the choice of discount rate. 9. A more detailed discussion of Monte Carlo simulation is provided in Appendix C. 10. The results in Exhibit 2-9 are purely for demonstration; they are not based on any actual climate projections for PNS. 11. The software is available for download at https://toolkit.climate.gov/tool/vulnerability-assessment-scoring- tool-vast. 12. In this discussion, the term “consequences” is used deliberately to distinguish them from the benefits in benefit-cost studies or the returns earned in a financial feasibility study. 13. The actual value of R could be estimated by accessing the high-temperature climate data described in Appendix D. 14. However, in some circumstances it will not be possible to do so, for example, when an existing facility is being compared with a replacement. The existing asset will likely have some remaining useful economic life, but that will not line up with the full economic life of a new replacement. This may also occur if various alternatives being compared have different economic lives. In this situation, FAA guidance sug- gests that the BCA time frame should be set equal to the useful life of the longest-lived alternative; then the shorter-lived alternative(s) would be assumed to be replaced as necessary, and a residual value would be assigned to the last one. 15. RCPs form a set of greenhouse gas concentration and emissions pathways designed to support research on impacts and potential policy responses to climate change. RCP8.5 combines assumptions about high population and relatively slow income growth with modest rates of technological change and energy inten- sity improvements, leading in the long term to high energy demand and GHG emissions in the absence of climate change policies. Impacts of climate change are higher in this scenario than in others. Endnotes

148 Climate Resilience and Benefit–Cost Analysis: A Handbook for Airports 16. This abstracts from the possibility that more than one extreme water event could actually occur in a single year. 17. There are nine FAA Regional Airports Offices located across the country: Alaskan, Central, Eastern, Great Lakes, New England, Northwest Mountain, Southern, Southwest, and Western-Pacific. 18. In principle, the number of simulations needed for any given analysis will depend on a number of differ- ent factors, including the standard error of the mean of the actual input distribution, the desired statistical confidence level, and the sampling method used. In the present context, it will usually suffice for the analyst to run a few thousand simulations and then compare the results to those obtained when running, say, twice that number. If the mean and standard deviation of the discounted benefits and costs are very similar, then the analyst can be confident that a large enough number of simulations have been run. 19. The LOCA website at http://loca.ucsd.edu has links to download sites containing the latest LOCA data. The downscaled LOCA technique will be used in the future to provide high-resolution projections for other variables, including snow cover, soil moisture, runoff, and humidity. 20. While it is recognized that weight restrictions may begin to impinge even at temperatures below 100° at some locations, for screening purposes, it is believed that 100° is a reasonable cutoff to assess the future likelihood and impact of such restrictions. 21. In the climate science world, a vertical datum is simply a reference level. Any water level measurement must be referenced to a datum in order to be meaningful. There are many different datums used for different pur- poses. Some are based on tidal levels—MHHW and mean sea level (MSL) are two examples. Some are based on the overall shape of the earth (so-called “geodetic” datums) such as the North American Vertical Datum for 1988, known as NAVD88, which is applicable to large continental areas. See https://noaanhc.wordpress. com/2016/01/29/the-alphabet-soup-of-vertical-datums-why-mhhw-is-mmm-mmm-good/ for more infor- mation. Most analyses studying sea level rise and coastal storms use MHHW because measurements relative to MHHW are a good approximation of the threshold where water inundation can occur. It is straightforward to convert from one datum to another simply by adding or subtracting their relative difference. A comprehen- sive list of vertical datums for different locations can be found at https://tidesandcurrents.noaa.gov/stations. html?type=Datums. 22. The formula is water level above MHHW = m + σ/ξ  [ypξ – 1], where m is the location parameter, ξ is the scale parameter, ξ is the shape parameter of the extreme value distribution, p is the annualized probability of occurrence, and yp = –1/ln(1 – p)] 23. Projections are provided for sites covered by the Permanent Service for Mean Sea Level (PSMSL) based in Liverpool, UK. Available at https://tidesandcurrents.noaa.gov/publications/techrpt083.csv. 24. There are also low and high sub-scenarios presented for each case; focus is on the baseline medium sub- scenarios for purposes of the Excel template. In addition, the raw RSL tables reflect expected changes relative to the year 2000. But given that the EWL curves described here are based on historical observations through 2010, the RSL tables have been adjusted so that they reflect a 2010 baseline instead of 2000. Finally, it is important to note that the RSL projections also attempt to account for changes in vertical land movement, as applicable. 25. It is important to note that the climate science supporting the probabilities shown in Exhibit D-4 is changing rapidly. As noted in the CO-OPS 083 report, recent evidence regarding the Antarctic ice sheet, for example, may lead to significantly increased probabilities associated with the intermediate-high, high, and extreme scenarios, particularly for RCP8.5. 26. Though the raw inputs for historical extreme water levels and future sea level rise are in meters, it is impor- tant to note that all of the results in the Excel templates are presented in feet. 27. Both Excel templates are hard-wired to compute 5,000 simulations. After testing with other higher counts, the project team is confident that, regardless of specific input choices made by the user, the variation in results will be sufficiently represented by running 5,000 simulations. 28. Note that both curves exhibit several discontinuities looking like negative spikes. This is just an artifact of the sorting of results based on the difference between the two curves; it occurs when extreme water events with very high costs (as indicated in Exhibit E-6) randomly occur in the early years of some of the 5,000 simula- tions. This leads to high negative NPV costs under both the baseline and the scenario cases. 29. Again, it is important to emphasize that the results shown in the template are all relative to the MHHW datum. If the airport is using NAVD88 or some other datum, it is essential to first translate the critical eleva- tions to be relative to MHHW. 30. Even though our case study analysis of Phoenix involved the assumption of cancelling flights entirely due to extreme temperatures, the high-temperature template focuses instead on weight restrictions, which are likely to begin to occur at much lower temperatures. 31. See Appendix D for more information on where these data can be obtained. Further instructions for accessing and downloading these data are provided in the template itself on the Weather Data sheet. 32. https://www.faa.gov/documentLibrary/media/Advisory_Circular/draft_150_5325_4c_industry_ commentenabled.pdf.

Endnotes 149 33. Coffel et al. graciously agreed to provide lookup tables from their analysis that provide statistical estimates of weight restrictions for the Airbus A320 and Boeing 737-800, 777-300, and 787-8 aircraft. 34. See the discussion in Appendix D. 35. If desired, the user may elect to ignore the weights and treat all models equally. 36. The weight restrictions are computed by comparing the maximum takeoff weight allowed for each aircraft type (given the airport elevation, runway length, and temperature) with the required takeoff weight implied by the passenger count per flight and required fuel for the route. If the maximum takeoff weight restriction is binding, then following Coffel et al. 2017, it is assumed that each pound of required weight reduction translates into 0.83 pounds of payload (passengers) and 0.17 pounds of fuel. 37. Each model also was tested against historical PHX temperatures for the period 1981 through 2000 to look for any systematic bias. The testing strategy described in Appendix D was used, and it was found that the upper end of the projected temperature ranges were quite accurate, involving differences in the top 5-percentile bracket of about 2°F or less across all models. 38. Additional guidance on this subject is provided in an FAA memo entitled “Planning Information Needed for FAA Headquarters Review of Benefit Cost Analysis (BCA),” which is available at https://www.faa.gov/ airports/aip/bc_analysis/media/planning-information-bca.pdf. 39. See for example, R. Kocherlakota, “The Equity Premium: It’s Still a Puzzle,” Journal of Economic Literature, 1996: 42-71. 40. It is important to note, in this case, that the nearest localized points in the NOAA data sets—both for histori- cal extreme water levels and projected sea level rise—are actually quite distant from the airport itself (they are 40–50 miles south along the actual Gulf of Mexico coast), so their accuracy for MSY is questionable. The airport would likely have to adjust the data by evaluating the relationship between the historical NOAA data sets and the actual experience at the airport. For coastal airports subject to sea level rise, the location (and elevation) of the nearest EWL and RSL stations must be an important consideration when undertaking analyses based on these climate estimates. 41. In this case, the NOAA coastal water level stations are within 2 miles of BOS. 42. There are other impacts that could be addressed as well. Rather than cancel or reschedule flights, airlines may accept payload restrictions on their current flights during periods of high temperatures; such restrictions may begin to occur at much lower levels than the unusually high temperatures faced by Phoenix in 2017. The Excel template directly addresses the issue of aircraft weight restrictions due to high temperatures. A more general concern for an airport might be that airlines faced with increasing costs due to high tempera- tures might choose to concentrate hubs or focus activity elsewhere, which would have important local eco- nomic impacts. From a more general perspective, some or all of these impacts could be addressed through improved engine or other aircraft technologies. 43. A more detailed and systematic analysis of payload restrictions is discussed in Appendix E as part of the Excel model high-temperature template developed for this project. 44. The 6-hour delay was assumed based on the relatively low frequency of daily flights at Little Rock.