Northern Rockies Climate Change Primer Observed trends in Northern Rockies temperature and precipitation Temperature increased about +1.2°F (0.7°C) for maximum temperature and about +2.1°F (1.2°C) for minimum temperature averaged for Northern Rockies stations, from 1916 to 20061,2. This is comparable to the average warming over the U.S. lower 48 states as a whole: an increase of +1.0°F (0.6°C) over the same time period.3 Of all stations, 77% showed an increase in maximum temperatures, and 94% demonstrated higher minimum temperatures. Annual precipitation, averaged over the region, increased by about +9% since 1916,1,2 slightly more than the U.S. national average increase of about +8%.3 Precipitation increased at 81% of stations, and the increases primarily occurred in spring, summer, and fall.

Figure 1. Trends in minimum temperature (top left), maximum temperature (top right) and precipitation (bottom right) at weather stations in the Northern Rockies, 1916-2006. Data from NOAA NCDC2

Projected 21st century Northern Rockies climate changes To make projections, climate scientists use greenhouse gas scenarios – “what if” scenarios of plausible future emissions – to drive global climate model (GCM) simulations of the earth’s climate.4 Greenhouse gas scenarios are based on assumptions about future changes in factors that influence emissions, including population growth, technological change, geopolitics and global development. All GCMs project warming in the Northern Rockies during all seasons, regardless of differences among models or greenhouse gas scenarios (Table 1).5 By about midcentury, projected temperature increases exceed the warmest years observed in the 20th century (Figure 2). Beyond mid-century, the actual amount of warming that is projected depends increasingly on the amount of greenhouse gases emitted globally over the coming decades. In contrast with temperature, projected changes in precipitation are small compared to year-

Table 1. Projected changes in temperature and precipitation, averaged over a region encompassing the Northern Rockies (117W-105W and 42N-49N). Results show the average and range among 10 global climate model (GCM) projections for the 2020s (2010-2039), 2040s (2030-2059), and 2080s (2070-2099), relative to 1970-1999, for a low (B1) and medium (A1b) greenhouse gas scenario.4 Data: GCM projections described by Meehl et al.5 and evaluated by Littell et al.1

to-year variability.

Figure 2. Average yearly temperatures for the Northern Rockies relative to the average for 19701999 (gray horizontal line). The black line shows the average simulated temperature for 1900–2000, while the grey lines show individual model results for the same time period. Thin colored lines show individual model projections for two emissions scenarios (low: B1, and medium: A1b4), and thick colored lines show the average among ten model projections for each scenario. Data: GCM projections developed for the 2007 IPCC report5 and evaluated by Littell et al.1

Climate projections differ among models because models operate at different spatial scales and use different approaches to simulating the physics of the earth’s climate. Comparing past climate from observations with projected changes across many climate models (Figure 3) provides context for the projections. For instance, the lowest modeled mean annual temperature for the 2080s is warmer than the warmest temperatures observed historically.

Figure 3. Range of observed historical2 (blue bars) and projected future changes5 in annual temperature and precipitation. Global climate model (GCM) projections for the A1b greenhouse gas scenario4 are show for the 2020s (2010-2039; orange bars), 2040s (2030-2059; red bars), and 2080s (2070-2099; brown bars). Each individual bar shows the results for a single GCM. Bars show the mean (white), 5th and 95th percentile values (bars), and minimum and maximum values (dots) among all years for each historical and future time period. The large vertical bars show the average projection for all models. Data: GCM projections described by Meehl et al.5 and evaluated by Littell et al.1

Assuming a medium greenhouse gas scenario (A1B4), climate models project, on average, a temperature increase of around +3.2°F (1.8°C) for winter and spring, +4.3°F (2.4°C) for fall, and about +5.0°F (2.8°C) in summer by the 2040s (2030-2059, relative to 1970-1999; Figure 4). For the same scenario and time period, models project increases in precipitation in winter (+8%), spring (+9%), and fall (+4%), but a decrease in summer (−9%).5

Figure 4. Projected changes in seasonal temperature and precipitation for the Northern Rockies for the 2040s relative to 1970-1999. Results are shown for three greenhouse gas scenarios, ranging from low (B1) to medium (A1B) to high (A2).4 Each dot is a single climate model projection, and boxes show the median (bar inside box), 25th and 75th percentiles (box edges), and 5th and 95th percentile changes (“whiskers”).

Projected 21st century changes in Northern Rockies hydrology Hydrologic models combine projected changes in temperature and precipitation with local information on soils, vegetation, and terrain to simulate the hydrologic consequences of climate change. Warmer temperatures, for instance, result in decreased winter snow accumulation and earlier spring melt (Figure 5). Changes are most pronounced at lower elevations. Warming also leads to declines in summer soil moisture, a consequence of both earlier snowmelt and increased summer evapotranspiration.

Figure 5. Projected changes in April 1st Snow Water Equivalent (SWE; left) and July 1st soil moisture (right) for 2040s (2030-2059) in the Northern Rockies (average of 10 climate models, A1B scenario, relative to 1916-2006). SWE is a measure of the amount of water contained in the snowpack. Data: Littell et al.1

In the Northern Rockies, snowpack is important in governing streamflow, soil moisture, and habitat for snow-dependent wildlife such as lynx and wolverine. Lower streamflow in late summer also affects habitat quality for resident and migratory fish. Declining snowpack and soil moisture therefore are likely to affect the region by affecting water availability, plant growth, forest health, fire risk, and wildlife habitat. In watersheds where winter precipitation is currently dominated by rain instead of snow (“rain dominant”), changes in snowpack will not be as profound as in today’s “mixed rain and snow” watersheds, where more cool season precipitation currently accumulates as snow (Figure 6). In the highest watersheds, where most cool season precipitation currently falls as snow, temperatures are cold enough that watershed hydrology will be less affected by warming. Below are maps of watershed snowpack vulnerability, defined as “rain dominant” (40%). In the Northern Rockies, most mixed rain and snow watersheds

become rain dominant, and most snow dominant watersheds transition to mixed rain and snow by the 2080s under all three scenarios.

Figure 6. Projected changes in the ratio of April 1st snowpack to total cool season (October–March) precipitation for the 2040s (2030-2059) and 2080s (2070-2099) in the Northern Rockies, for the A1B greenhouse gas scenario. Projections are shown for three scenarios: the average of ten climate models (middle), a model that projects less warming than the average (left), and one that projects more warming than the average (right). Some watersheds remain mixed rain and snow, but most historically mixed watersheds become rain dominant and most historically snow dominant watersheds become mixed rain and snow by the end of the 21st century. Data: Littell et al.1

Projected 21st century changes in Greater Yellowstone Area hydrology Focusing in on the Greater Yellowstone Area (GYA), similar changes are projected: warming results in snowpack declines at most locations and decreased summer water availability (declining streamflow and soil moisture, increasing water deficit). These changes in turn have consequences for streamflow, ecosystem function and species’ biology. Only a few (about 1% of the GYA) of the highest places experience increases in spring snowpack – where temperatures are cold enough that rising temperatures do not increase melt, and modest increases in winter precipitation result in a net increase in snow accumulation.

Figure 7. Projected changes in April 1st Snow Water Equivalent (SWE) for the 2040s in the GYA (2030-2059, relative to 1970-1999 for the A1B greenhouse gas scenario). Projections are shown for a climate model that projects less warming (left), the average of ten climate models (middle), and a model that projects more warming (right). Averaged over the GYA, the composite projection is for a decline in April 1 SWE of −34%. Data: Littell et al.1

Similarly, summer soil moisture is projected to decline in response to increasing evapotranspiration and declining snowpack. One of the unintuitive features of soil moisture change is that the largest declines do not occur in historically water stressed areas, a consequence of the fact that these areas have “less water to lose”. Instead, as is true for the Northern Rockies, changes are most pronounced in wetter areas where changes in summer temperatures and winter snow accumulation conspire to produce large changes in water availability. As discussed below, these changes have consequences for terrestrial and aquatic ecosystems.

Figure 8. As in Figure 7, except showing projected changes in July 1st soil moisture. Averaged over the GYA, the composite projection is for a decline in mean July 1st soil moisture of −7%. Data: Littell et al.1

Projected 21st century changes in GYA hydrology As in the Northern Rockies in general, watersheds are projected to transition towards greater rain dominance in the GYA (Figure 9): most mixed rain and snow watersheds are projected to become rain dominant, and all snow dominant watersheds become mixed rain and snow by the 2080s (2070-2099) under all climate change scenarios. Mixed rain and snow watersheds – near the snowline, where snow accumulation is important but does not dominate winter precipitation – are projected to experience the greatest changes in response to warming.

Figure 9. As in Figure 7, except showing projected changes in the ratio of April 1st snowpack to total cool season precipitation (October–March). Some watersheds remain mixed rain and snow, but most historically mixed watersheds become rain dominant and most historically snow dominant watersheds become mixed rain and snow by the end of the 21st century. Data: Littell et al.1

Forest and grassland responses to climate change are partially determined by the water available to plants given the supply of water (soil moisture from precipitation and snowmelt) and the demand for water by the atmosphere (potential evapotranspiration). Actual evapotranspiration is an estimate of how much water plants use given the supply and demand they experience and other factors specific to different vegetation types. The difference between potential and actual evapotranspiration (PET – AET) is water balance deficit. The larger the deficit, the further from water balance the environment is for plants. Increasing deficit is related to increasing area burned by fire, decreasing tree growth, and increasing vulnerability to forest insects like the mountain pine beetle.

Figure 10. As in Figure 7, except showing projected changes in summer (July-August-September) water balance deficit. Warmer temperatures and decreasing precipitation increase deficit, and most of the increased deficit in the GYA is due to projected increases in potential evapotranspiration. For the 2040s, deficit increases by +31% on average across the GYA. Data: Littell et al.1

Climate change impacts on Northern Rockies and GYA ecosystems

Figure 11. Examples of recent studies that quantify climate change impacts on species and ecosystems. Top row: Suitable wolverine habitat (dark areas) for historical (left; 1970-1999) and 2080s (right; 2070-2099) for the A1B scenario and the average of ten global climate models.6 Middle row: Length of suitable stream habitat for four trout species (left) and map of core population habitat for cutthroat trout in the 2080s (right; 2070-2099, A1B scenario).7 Bottom row: Average summer (July-September) water balance deficit in low (left) and high (right) fire years, 1980-2009. Deficit is associated with large fire years, particularly in the Northern Rockies.8,9

The climate and hydrologic projections described in this primer have been used to analyze potential ecological responses in the western United States and Northern Rockies (e.g., Figure 11). For example, the observed relationship between spring snowpack and wolverine denning were applied to future snowpack estimates under climate change scenarios to demonstrate projected declines in future wolverine habitat and its connectivity (average of −63% by the 2080s compared to historical).6 Another study projects that suitable thermal and flow regimes for cutthroat trout habitat will decline by −58% by the 2080s compared to historical.7 Finally, research shows that the historical relationship between water balance deficit and area burned in the northern Rockies is robust.8,9 Future declines in water availability suggest large increases in area burned. In each of these cases, the climate and hydrologic mechanisms that drive changes are different. The ability to develop quantitative projections depends on identifying these mechanisms and evaluating how they are impacted by climate change. 1

2 3 4

5

6

7

8

9

Maximum temperature, minimum temperature, and precipitation trends from raw USHCN and COOP network data were analyzed as part of: Littell, J.S., M.M. Elsner, G. S. Mauger, E. Lutz, A.F. Hamlet, and E. Salathé. 2011. Regional Climate and Hydrologic Change in the Northern US Rockies and Pacific Northwest: Internally Consistent Projections of Future Climate for Resource Management. Project report: April 17, 2011. Latest version online at: http://cses.washington.edu/picea/USFS/pub/Littell_etal_2010/ Menne, M.J. et al., 2009. The US Historical Climatology Network monthly temperature data, version 2. Bulletin of the American Meteorological Society, 90(7), 993-1007. http://www.ncdc.noaa.gov/cag/ Nakićenović, N., and R.J. Swart, eds. 2000. Special Report on Emissions Scenarios: A Special Report of Working Group III of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, U.K., 599 pp. Available online at: http://www.grida.no/climate/ipcc/emission/index.htm. B1, A1B, and A2 SRES greenhouse gas emissions scenarios are representations of plausible future emissions given assumptions about social and economic global trends. It is important to note that while B1 is always the lowest scenario between 2001 and 2100, A1B has higher greenhouse gas emissions early in the 21 st century, leading to greater warming under A1B than A2 until the middle of the 21 st century, after which A2 emissions cause greater warming past about the 2060s. Meehl, G., C. Covey, T. Delworth, M. Latif, B. McAvaney, J. Mitchell, R. Stouffer,and K. Taylor. 2007. The WCRP CMIP3 multi-model dataset: a new era in climate change research. Bulletin of the American Meteorological Society, 88(9):1383-1394. Climate model output archived at the CMIP3 archive (http://wwwpcmdi.llnl.gov/ipcc/about_ipcc.php) was used in this work, which relies on the same models as the 2007 Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment (AR4). See [1] for evaluation of climate models over the upper Columbia, Missouri, and Colorado River basins. McKelvey, K.S., J.P. Copeland, M.K. Schwartz, J.S. Littell, K.B. Aubry, J.R. Squires, S.A. Parks, M.M. Elsner, G.S. Mauger. 2011. Climate change predicted to shift wolverine distributions, connectivity, and dispersal corridors. Ecological Applications, 21(8):2882–2897. Wenger, S.J., D.J. Isaak, C.H. Luce, H.M. Neville, K.D. Fausch, J.B. Dunham, D.C. Dauwalter, M.K. Young, M.M. Elsner, B.E. Rieman, A.F. Hamlet and J.E. Williams. 2011. Flow regime, temperature, and biotic interactions drive differential declines of trout species under climate change. Proceedings of the National Academy of Sciences doi:10.1073/pnas.1103097108. Littell, J.S. and R. Gwozdz. 2011. Climatic water balance and regional fire years in the Pacific Northwest, USA: Linking regional climate and fire at landscape scales. Chapter 5 in McKenzie, D., C.M. Miller, and D.A. Falk, eds. The Landscape Ecology of Fire, Ecological Studies 213, DOI 10.1007/978-94-007-0301-8_5, © Springer Science+Business Media B.V. 2011. McKenzie, D., and J.S. Littell. 2011. Climate change and wilderness fire regimes. International Journal of Wilderness, 17(1):22-27, 31.