GC31C-01 INVITED
Climate Change Has Cascading Ecological Effects on Mountain Ecosystems
Evidence that ecosystems of the Northern Rocky Mountains are responding to climate change abounds. Alpine glaciers, as iconic landscape features, are disappearing rapidly with some glaciers losing one half of their area in five years. A model developed in the 1990s to predict future rates of melt has proved too conservative when compared to recent measurements. The largest glaciers in Glacier National Park are almost 10 years ahead of schedule in their retreat. The cascading ecological effects of losing glaciers in high-elevation watersheds includes shifts in distribution and dominance of temperature-sensitive stream macroinvertebrates as stream volume dwindles (or disappears) in later summer months and water temperatures increase. Critical spawning areas for threatened bull trout (Salvelinus confluentus) will be lost without the consistent supply of cold water that melting snow and ice provide and raise management questions regarding the efficacy of recovery efforts. Snowpacks are documented as becoming smaller and melting earlier in the spring, facilitating the invasion of subalpine meadows by trees and reducing habitat for current alpine wildlife. Even vital ecosystem disturbances, such as periodic snow avalanches that clear mountain slope forests, have been shown by tree-ring studies to be responsive to climatic trends and are likely to become less prevalent. Monitoring of high-elevation mountain environments is difficult and has largely been opportunistic despite the fact that these areas have experienced three times the temperature increases over the past century when compared to lowland environments. A system of alpine observatories is sorely needed. Tighter integration of mountains studies, and comparisons among diverse mountain systems of the western U.S. has been initiated by the USGS-sponsored Western Mountain Initiative and the Consortium for Integrated Climate Research in Western Mountains to begin addressing this need.
GC31C-02 INVITED
Complex Patterns in Climate and Atmospheric Nitrogen Deposition Influence Rocky Mountain Ecosystems
Long-term monitoring of physical and biogeochemical characteristics in Loch Vale watershed, Rocky Mountain National Park, has revealed complicated patterns in temperature, precipitation, and atmospheric nitrogen deposition. July mean and maximum temperatures have increased since 1985 by 0.1-0.2 ° C, while March mean and maximum temperatures became 0.1-0.3 ° C colder. There is no long-term trend in annual or monthly precipitation; annual totals range 75-140 cm yr-1. Atmospheric N deposition has increased approximately 2% yr-1 since 1985, and there are strong upward trends in July and September deposition. A combination of observations, ecosystem modeling (DayCent-Chem model), and structural equation modeling (SEM) suggests this alpine/subalpine catchment is responding physically, biologically, and chemically. Observed stream discharge was greater than measured precipitation in several recent years, indicating melt from glacier ice contributes to flow. Model results suggest a strong increase in alpine microbial activity and plant N uptake, and a moderate increase in forest microbial activity driven by increased temperatures and increased N deposition. Alpine lichen activity appears to also have been significantly stimulated. There has been a significant increase in observed stream nitrogen concentrations and flux. Annual mean stream N concentrations in alpine/subalpine Loch Vale watershed of Rocky Mountain National Park have increased from approximately 1.0 to 1.5 mg NO3 L-1 between 1991 and 2005; the annual amplitude has also increased. Mean annual N efflux from the catchment doubled between 1991 and 2005. SEM suggests N loss from Loch Vale appears to result most strongly from the combined influence of temperature and precipitation on stream flow, and secondarily from the influence of terrestrial nitrogen cycling.
GC31C-03
Recent Accelerated Warming in Western United States Mountains
The mountainous western portion of the United States and North America have been warming since the mid 1970s, by a total of about 1 degree C. This warming has been most pronounced in spring, present but less obvious in winter, and nearly absent in autumn. In summer, a slow rise in temperature was under way until the late 1990s. Over the past 8 years a succession of very warm summers has occurred, with multi-year temperatures well above all similar past multi-year averages. Of further interest is that the major Colorado River drought now underway also began about this time. Contributing to that drought were several springs that featured a month with much above normal temperatures that greatly hastened snow melt at a time of year when snow pack is usually still accumulating. These warming effects have been seen in lower elevation and higher elevation onitoring networks. They do not appear to be a product of observational methodologies. There is some evidence that other parts of the global climate system have behaved somewhat differently over those years. Potential relationships or causes for this apparent acceleration in warming will be discussed.
GC31C-04
Deriving high resolution historical and future climate databases for mountainous environments
General circulation model (GCM) and regional climate model (RCM) data are currently available in resolutions sufficient to conduct reasonable climate change studies for hydrologic systems on a scale of thousands of square kilometers and/or for terrains with minimal topographic variability. The Canadian Forest Service has created monthly climate datasets for North America at a 10 km resolution by interpolating climate station data (McKenney et al Agric. For. Meteorol, 138, 69-81, 2006). However, much finer spatial resolutions are needed to investigate changes in watershed and ecosystem processes, particularly in regions of diverse topography. This work is building a series of interpolated high resolution hydro-meteorological surfaces for predicting hydrologic change in mountainous regions. Level A climate stations (with complete historical record 1961-2006) are interpolated using the ANUSPLIN thin plate smoothing spline technique to create historical daily climate field for the study region. Error surfaces from ANUSPLIN define regions within the study area where data availability limits confidence in the analysis. To minimize these errors we are synthesizing climate data using multiple regression infilling techniques for level B stations – with limited data records; and where needed, we use the SIMGRID alpine microclimate model to create level C climate stations with wholly synthetic data. SIMGRID has been used to successfully simulate snow pillow data and to provide high-resolution spatiotemporal climate data for hydrological simulations (Lapp et al., IJOC 25 (4), 521-526, 2005). The goals are two fold: first, to develop a technique for creating high resolution hydro-meteorological surfaces in mountainous regions; and second, a classified series of alpine climate response units for application in hydrologic and ecologic research.
GC31C-05
Climate change and Elevational Dependence at a Mid-Latitude Mountain System, Niwot Ridge, Colorado Rocky Mountains
Mid-latitude mountain systems are critically sensitive to recent and projected climate change under an elevated greenhouse gas world. It is often taken that climatic change at high elevation sites will reflect those at lower sites - regional warming is assumed to be consistently played out in mountains, or even amplified by the snow-albedo feedback. The anticipated outcome is that the alpine will eventually be "pushed off the top of mountains." There are several reasons why this might not be the case, or at least considerably delayed - one is whether high elevation climates reasonably reflect regional lowland trends or if they are decoupled from them as a result of mountain climatic processes. We evaluated standard climatological variables (minimum & maximum temperature, precipitation) and derived variables [diurnal temperature range, growing season length (using both 0° & -3°C thresholds), and growing degree days (0°C base)] from subalpine (C1, 3048m) and high alpine (D1, 3749m) sites from 1953 to 2006 at Niwot Ridge in Colorado, the longest high- elevation climate record in the US. Over the last 54 years, mean maximum temperature (Tmax) increased through much of the year in the subalpine (trend in annual Tmax=+0.4°C/decade), but in the alpine decreased in early winter (-0.4 to -0.6°C/decade). These patterns resulted in altered seasonal cycles for the two sites, but in different ways: a positive offset in the subalpine (C1) and amplification in the alpine. Precipitation increased at the alpine site from October through April (trend in annual ppt=+100mm/decade), but not during any season in the subalpine. At both sites, summer onset is later and termination earlier, so that the "growing season" has shortened - this reflects long-term tendencies in minimum temperatures. An apparent contradiction is that growing degree-days have gone up at the subalpine site; this due to the positive trend in maximum temperatures. The alpine showed no corresponding trend. An integrated view of these trends infer synoptic dynamics and surface energy processes that act differently in the high alpine near the Continental Divide vs. in the subalpine dominated by closed conifer forest. At the same time, these climates are affected by multidecadal hemispheric circulation changes. Nearly all temperature-related timeseries for both sites show a period of cooling until around 1980, followed by warming. Precipitation series show corresponding periods of increasing then decreasing precipitation. On the face of it, this pattern resembles that of the Pacific Decadal Oscillation (PDO). This suggests that alpine and subalpine climate signals are not as decoupled as they appear, but rather that across a relatively short elevational gradient (Δ700m) synoptic and landscape-scale processes react differently to and differentially modify a prevailing hemispheric signal.
GC31C-06
Evidence of Regional Warming during the 20th Century in Alpine and Subalpine Lakes in the Western United States
Subfossil midge analyses have been used to develop high-resolution (sub-decadal) reconstructions of 20th century temperature change in the Sierra Nevada, CA with success. Expansion of this earlier work to additional sites in the western United States suggests that a widespread increase in lake water temperatures has occurred in this region during the late 20th and early 21st centuries. Inference models for summer surface water temperature (SSWT) were developed combining midge abundance data from 56 lakes in the eastern Sierra Nevada, California, with subfossil midge remains from the Uinta Mountains, UT. The newly merged Sierra Nevada–Uinta Mountains calibration set contains a greater diversity of chironomid assemblages and spans a wider SSWT range than the previously published Sierra Nevada calibration set. The lakes in the merged calibration set spanned elevation, depth, and SSWT temperature ranges of 900 m, 12.7 m, and 11.3 oC, respectively. A robust inference model for SSWT (3-component WA-PLS), based on 90 lakes, had a high coefficient of determination (r2jack = 0.66) and a low RMSEP (1.4 oC). The midge-based SSWT inference model was applied to subfossil chironomid remains extracted from well-dated sediment sequences recovered from alpine and subalpine lakes in the Sierra Nevada, CA, Snake Range, NV and Uinta Mountains, UT. A close correspondence exists between the chironomid-inferred temperature profiles for the 20th and 21st centuries and mean July or summer temperatures measured at nearby meteorological stations. Application of this midge-based SSWT inference model to other intact, late Quaternary sedimentary sequences found in subalpine and alpine lakes in the Great Basin will help resolve the impact of late Quaternary and recent climate change in this region, improve our understanding of regional climate and aquatic ecosystem variability, and can be used to monitor the effects of climate change on aquatic ecosystems and establish ‘baseline' conditions against which future biotic changes can be compared.
GC31C-07 INVITED
Critical Hydrologic and Atmospheric Measurements in Complex Alpine Regions
The Alps are often referred to as the « Water Towers of Europe » and as such play an essential role in European water resources. The impact of climatic change is expected to be particularly pronounced in the Alps and the lack of detailed hydrologic field observations is problematic for predictions of hydrologic and hazard assessment. Advances in information technology and communications provide important possibilities to improve the situation with relatively few measurements. We will present sensorscope technology (arrays of wireless weather stations including soil moisture, pressure, and temperature) that has now been deployed at the Le Genepi and Grand St. Bernard pass. In addition, a Distributed Temperature Sensor array on the stream beds has been deployed and stream discharge monitored. The high spatial resolution data collected in these previously "ungaged" regions are used in conjunction with new generation hydrologic models. The framework as to what is possible today with sensor arrays and modeling in extreme mountain environments is discussed. http://eflum.epfl.ch
GC31C-08
Climate Change in Tropical East Africa: Combining High-Altitude Measurements, Proxy Records and Numerical Modeling
Several types of proxy data indicate a rapid climate change in tropical East Africa around 1880, leading from wet to relatively dry conditions. The change manifested itself in the drop of lake levels, alteration of limnological sediment composition, and - at high altitude - glacier recession. To understand these changes, we run an extensive field program on the glaciers in the vicinity of Kilimanjaro summit (5895 m a.s.l.). Application of the field data to a glacier mass balance model, which resolves the physics of the glacier-climate interaction, allows the derivation of former glacier extents and thus local climate conditions of the pre-1880 wet period. From the latter we can deduce the magnitude of changes in precipitation, air temperature, air humidity, and solar radiation in the mid-troposphere. To explore the potential value of upscaling from regional to large-scale tropical climate change, a 200-year paleoclimate simulation with the Community Climate System Model (CCSM 3.0) is analyzed. Results suggest that changes in Indian Ocean dynamics (atmosphere-ocean interactions), and related moisture transport into East Africa, most likely contributed to this rapid regional climate change.