Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Office of Planning, Environment, & Realty (HEP)
PlanningEnvironmentReal Estate

HEP Events Guidance Publications Awards Contacts

The Use of Climate Information in Vulnerability Assessments

Notes

[1] In this memorandum, discussion of "climate" will remain focused on temperature and precipitation. This is consistent with the discussion of climate information that took place during the Breakout Session at the Newark Pilot Peer Exchange workshop, May 4-5, 2011. As such, other aspects of climate (e.g., atmospheric circulation patterns, ocean currents, streamflow, snowpack) will not be discussed at length in this memorandum. Although some of these other aspects of climate may affect transportation infrastructure, the applicability, reliability, and availability of data will be highly variable from location to location; temperature and precipitation, on the other hand, will likely need to be considered in any vulnerability assessment.

[2] Climate models' ability to simulate the daily-scale statistics of climate in a particular region is limited. Quantitative information derived from daily-scale statistics should be scrutinized and used with care (as discussed further in the "Translating to Impacts" section). For example, changes in daily-scale extremes can span a wide range - projections for California exhibit increases in the frequency of intense heat waves ranging from a doubling or tripling to increases by a factor of 20 (Cayan et al., 2009). Although all models and scenarios project a warmer future, including an increase in the frequency and intensity of heat waves, there is significant uncertainty as to how this warming translates into daily-scale changes in weather in particular locations across California.

[3] The relationship between El Niño-Southern Oscillation and precipitation in Southern California is particularly strong. Many locations in the United States, especially away from the West Coast, will not exhibit such a strong connection.

[4] Much of this data is available online. USGS Streamstats (http://water.usgs.gov/osw/streamstats/) has historical streamflow data and statistics for various locations across the country Federal Emergency Management Agency allows users to access maps of floodplains (referred to as Flood Insurance Rate Maps, or FIRMs; http://www.fema.gov/hazard/map/firm.shtm), which have been developed using historical data; NRCS has snowpack and soil moisture data across much of the western United States (http://ftp.wcc.nrcs.usda.gov/).

[5] Technically speaking, the model projections should be compared to the model simulations for the 20th century, and then the projected change should be compared to the observed statistics (e.g., those taken from station data). For any model, especially at the regional-scale, the simulated 20th-century climate is likely to differ slightly from the observed 20th-century climate -this is often referred to as bias. Focusing on the projected change avoids incorporating the model's bias into any interpretation of expected regional-scale climate changes.

[6] Notably, scenarios do not include any assumptions about policies designed to explicitly discourage emissions of greenhouse gases.

[7] For the next round of IPCC model simulations (the Fifth Assessment, or AR5), new scenarios are being developed. These scenarios, or Representative Concentration Pathways (RCPs), willoften involve similar assumptions about future population growth and changing socio-economic conditions as the scenarios used in previous Assessments. However, the RCPs have been developed with greater emphasis on greenhouse gas concentrations, rather than emissions. Concentrations can be used more efficiently as inputs to integrated assessment models that examine impacts. Also, the RCPs will include information about possible mitigation policies, which was also lacking from the previous scenarios. For more information, see the recent issue of Climatic Change, which is devoted to discussing various aspects of the RCPs (http://www.springerlink.com/content/f296645337804p75/).

[8] Sub-grid scale processes are represented using parameters that are based on empirically-derived relationships, rather than direct calculations that are emerge from physical or chemical first principles.

[9] To clarify, this section is specific to understanding the statistics of daily-scale events as derived from climate models. For seasonal and annual timescales, all model and scenario combinations project increased temperatures for the United States and the globe (i.e. this is a robust finding of the models). For seasonal and annual precipitation, some models and some scenarios project increases in precipitation, while others project decreases. The results, and the consensus (or lack of a consensus) among models, are dependent on the geographic region.

[10] Calibration often involves the choice of a vertical datum. As defined by NOAA (2011), "A datum is a base elevation used as a reference from which to reckon heights or depths. A tidal datum is a standard elevation defined by a certain phase of the tide. Tidal datums are used as references to measure local water levels." Example tidal datums include mean higher high water and mean high water.

[11] The sea-level rise projection must be chosen to account for the vertical accuracy of the elevation data for the land (and vice versa). An accurate map requires the root mean square error of the elevation data to be smaller than the projected change in sea-level rise (NOAA 2009). For more in-depth discussion of land elevation data resolution and accuracy, see Chapter 2 of CCSP (2009).

[12] Adjustments to the vertical datum are a necessary part of mapping inundation. The land elevation data are usually referenced to a vertical datum called the North American Vertical Datum of 1988 (NAVD88). This datum is not tidal, meaning that a value of 0 does not equate to any particular local tide value. Correcting this issue requires converting the elevation data from NAVD88 to a tidal datum, such as mean high tide (NOAA, 2011).

[13] SLOSH (Sea, Lake and Overland Surges from Hurricanes) is a computerized model run by the National Hurricane Center (NHC) to estimate storm surge heights and winds resulting from historical, hypothetical, or predicted hurricanes. (http://www.nhc.noaa.gov/HAW2/english/surge/slosh.shtml)

[14] This memorandum was originally written prior to Hurricane Irene (August 20-29, 2011). Some of the impacts of Irene are discussed in the New Jersey team's final report.

Updated: 03/27/2014
HEP Home Planning Environment Real Estate
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000