The page you requested has moved and you've automatically been taken to its new location.
Please update your link or bookmark after closing this notice.
Analysis of historical climate allows decision makers to better understand and communicate the projected changes in climate and their associated impacts. The following sections discuss three specific types of benefits that can emerge from the analysis of historical climate:
In many locations, the temperature and precipitation patterns of the 21st century are likely to be considerably different than those experienced in the 20th century. However, the ability to anticipate the consequences of climate change on transportation services and assets will draw heavily on agencies' experiences regarding their performance in the past climate, especially during extreme conditions (e.g., heat waves, heavy rainfall events). Understanding how transportation systems have performed during periods of high temperature, heavy precipitation and flooding, or prolonged lack of precipitation can provide qualitative, and in some cases, quantitative information that can be extrapolated to future climate conditions.
The following questions are examples of the way in which past climate information can be employed to identify this type of sensitivity information:
Performing a sensitivity analysis can provide a foundation from which to identify future vulnerabilities. In addition, it helps to identify the particular climate variables that should be extracted from future climate projections (see discussion below) as part of the vulnerability analysis. For example, if heat waves pose problems for a transportation service, then the model projections for temperature during the spring, summer, and fall months should be investigated closely (increases in winter temperatures are unlikely to result in "heat waves" for most locations). Also, if heat waves are an issue, some attention should be paid to the daily-scale temperature variability associated with model projections (see the Projection of daily-scale extremes), in addition to the monthly and seasonal statistics.
Historical variability and the context for future changes
Examination of past climate allows future projections to be put into context. To illustrate the importance of understanding historical variability, records of annual precipitation and average annual temperature for a weather station in Pasadena, California are shown in Figures 1 and 2. These graphs and the accompanying discussion show that a comparison between historical variability and projected changes can be an important part of the assessment of local-scale impacts.
Total annual precipitation (December-November) is shown for the period 1950-2009. The black line shows the average rainfall (~20 inches per year). The gray lines show the averages for El Niño years (top gray line; wettest of the three averages) and La Niña years (bottom gray line; driest of the three averages). The red dashed lines correspond to hypothetical future increases and decreases in annual precipitation by 20%. The graph demonstrates that the year-to-year variability of precipitation is very large relative to the average precipitation. Future changes in precipitation (+/- 4 inches) are relatively small compared to the historical range, and the difference between the average of El Niño years and La Niña years. Data taken from the NCDC/NOAA U.S. Historical Climate Network.
Note: total precipitation is calculated from December through November to prevent splitting winter months that experience the same signal from El Niño-Southern Oscillation, as would be done by using the calendar years.
The precipitation graph demonstrates that the year-to-year variability of precipitation is very large relative to the average precipitation (middle black line on the graph). Although such variability is not representative of most stations across the country, many locations in Southern California exhibit this level of variability. For Southern California (and much of the U.S. West Coast), the El Niño-Southern Oscillation has a significant impact on precipitation falling during the winter, which in turn has a significant impact on annual precipitation. For Pasadena, the average precipitation falling during El Niño years is 10 inches greater than the average for the La Niña years.
Understanding this variability can help to put future changes in context. Although climate model projections vary regarding the direction of future precipitation changes in Southern California, many models show increases or decreases that are around 20% or less of average precipitation by the 21st century. These changes (both increases and decreases) have been shown in the graph above as red, dashed lines. Although these changes are not negligible, they clearly fall within the range of historical variability. The projected changes are much smaller (by more than a factor of 2) than the difference between the averages for El Niño and La Niña years.
Figure 2: Historical temperature in Pasadena, California
Average annual temperature is shown for the period 1950-2009. The black line shows the average for the period 1961-1990. The red dashed line corresponds to a hypothetical increase in average annual temperature by 3°F. In comparison to the precipitation graph, the hypothetical change in temperature, which is at the low end of projections for the end of the 21st century, is large compared to the historical variability. Data taken from the NCDC/NOAA U.S. Historical Climate Network.
Note: The black line only represents an average of a 30-year portion of the record. This was chosen because 1) the time series as a whole has a linear trend toward warming, and 2) the late 20th century is often used as a baseline from which warming is estimated from the output of climate models.
By comparison, the projection for future average temperature for Southern California (as shown by the red, dashed line in Figure 2) is outside the range of historical conditions. The hypothetical warming shown in Figure 2 is 3°F, which is at the low end of projections for the end of the 21st century. Some models and scenarios project warming of 8°F.
The purpose of the Pasadena example is to demonstrate that consideration of historical variability is an important component of an impacts assessment. Clearly, the increases anticipated for temperature in Southern California represent a fundamental change in the regional climate by the end of the 21st century. The projections for precipitation are both more uncertain (i.e., increases or decreases could occur) and fall within the range of historical conditions. This is not to say that the changes will be insignificant or negligible, but simply that the future average conditions are likely to resemble a significant portion of the historical record. In addition, the analysis shows that some of the year-to-year changes in precipitation may be worth considering as part of the impacts assessment, or in the development of adaptation options (i.e., a system that is resilient to some of the relatively large swings in historical year-to-year rainfall is also likely to be resilient to longer term changes in precipitation).
It is important to emphasize that the variability represented in the Pasadena record is not necessarily representative of other locations. In other places, the variability of precipitation is likely to be smaller, such that a 20% change in precipitation could represent a more significant difference from historical conditions. Similarly, other locations could exhibit greater temperature variability, in which a warming of 3°F may appear less drastic. The diversity in climate conditions across the country underscores the need to assess future climate projections in light of historical regional climate variability.
Past weather events (e.g., strong storms, heat waves) or climate events (e.g,, droughts, warmer-than-normal summers, wetter-than-normal years) that have negatively affected the transportation system can be used as examples to facilitate discussion with staff within a transportation agency or with other groups external to the agency (e.g., political representatives, commuters, transit riders). These examples can be important ways to raise awareness among these groups. Benefits include:
Drawing on shared experience: Past weather and climate events represent a shared experience. Within transportation agencies, this shared experience can be a valuable way to acquire information on sensitivity. More broadly, these events can be a way to invite participation from desired audiences, making the issue of climate change and climate impacts more personal and less technical.
Identification of potential adaptive responses: Especially for internal staff audiences, the experience of dealing with past weather and climate events can provide valuable information about what does and does not work when attempting to minimize damage or maximize opportunities associated with impacts. Past experience provides "lessons learned" that can help spark and guide discussion about adaptation options.
Justification for action: Past examples can provide the answer to "why are you performing a vulnerability assessment?" Discussion of the examples can help make the case that the vulnerability assessment can help to reduce the risks associated with similar, repeat occurrences of the examples.
Education regarding weather and climate: Aspects of the examples can help to educate staff or the broader community about climate variability, climate change, and local-scale impacts. Making the connection between weather and climate, in and of itself, is an important step toward educating people about climate.
Alternative framing of climate-related initiatives: Among some audiences, the terms "climate" and "climate change" are laden with significant political connotations. These connotations are likely unrelated to, or at best are tangential to, the task of increasing the resilience of the transportation system to extreme weather and climate impacts. Framing the vulnerability assessment in terms of past weather and climate events can help to avoid some of these political issues.
Limits of making extrapolations based on historical data
As outlined above, examination of historical climate data and observed impacts from weather and climate events can be useful for both analysis of vulnerability and communication. However, it is important to recognize a fundamental issue in considering climate change: the prevailing or typical historical climate conditions are unlikely to be representative of the future climate conditions. Although analysis of the past can yield useful "analogs" for certain types of weather events, provide insights into the types of impacts that might occur (or might occur more frequently), or serve as efficient communication tools, the climate is changing, and some future climate impacts may go beyond the range of impacts that have occurred historically.
On a related note, care should also be taken when examining and making inferences from trends in historical climate data. Although these trends can be informative when considering the types of changes that may occur (e.g., warming over the last 30 years has accompanied a greater frequency of heat waves), or in identifying sensitive infrastructure (e.g., the lifetimes of equipment or infrastructure have been shrinking as a result of more frequent recent high temperatures), it is unlikely that the trends of past decades will persist unchanged into the future. In most cases, extending past trend lines into the future would represent a poor model of future conditions, especially on longer timescales (greater than 30 to 40 years). For example, for all parts of the U.S., the rates of warming for the 21st century are expected to be greater than the rate of warming between 1900 and 2000. Moreover, for many scenarios of warming, the latter half of the 21st century is likely to exhibit more rapid warming than the first half of the 21st century.
Figure 3: Locations of active COOP weather and climate stations
Historical climate information can be found in a variety of data products that are maintained by the National Oceanic and Atmospheric Administration (NOAA). These products fall into two general categories:
Station Data - these records correspond to a specific location. NOAA's National Weather Service, in conjunction with significant volunteer contributions, maintains the Cooperative Station Network (COOP), which represents a large set of stations across the country (Figure 3), some of which have relatively long periods of record (greater than 100 years). These stations have observations for daily temperature, precipitation, and some provide measures of snowfall. A subset of these stations, the U.S. Historical Climate Network has been quality controlled for use in climatological research, and has over 1200 stations across the country (http://cdiac.ornl.gov/epubs/ndp/ushcn/ushcn.html).
Figure 4: Locations of NCDC climate divisions
Station data portrays the daily and monthly variability that has been witnessed at a particular location, including any extreme events that may have occurred there. Such a record can be particularly useful if it is nearby transportation assets that are included in the vulnerability assessment. However, such a station's record may not be representative of larger regions. Topography, differences in land cover, the presence of water bodies, and features of the average larger-scale circulation can affect the extent to which a station might be considered representative of a larger area.
Climate Division Data
The National Climate Data Center (also a part of NOAA) maintains a climate division data set. In these data, individual stations that fall in a region with a similar climate are averaged together. A map of the 344 climate divisions is shown in Figure 4.
These data have better quality from 1931 to the present than earlier periods (data from 1895-1930 were sometimes based on state averages, rather than actual station observations). These data can be useful for examining trends and variability over larger geographic areas, and avoid any issues of data quality that might arise from looking at an individual station. However, only monthly and annual data for temperature and precipitation are available.
There are many other sources of historical climate information for variables such as streamflow (data from U.S. Geological Survey (USGS)), snowpack measurements (the National Resource Conservation Service (NRCS), which is part of the U.S. Department of Agriculture), and gridded "reanalysis" products for more complex meteorological variables (National Centers for Environmental Prediction, which is part of NOAA). These data sets can be useful for assessing some types of impacts, especially hydrological impacts such as droughts and floods. However, we have restricted our discussion to those products that are most closely related to temperature and precipitation at the surface, and to those products that would be easy to download and use.