Elevation-dependent temperature trends in the Rocky Mountain Front Range: changes over a 56- and 20-year record.

Determining the magnitude of climate change patterns across elevational gradients is essential for an improved understanding of broader climate change patterns and for predicting hydrologic and ecosystem changes. We present temperature trends from five long-term weather stations along a 2077-meter elevational transect in the Rocky Mountain Front Range of Colorado, USA. These trends were measured over two time periods: a full 56-year record (1953-2008) and a shorter 20-year (1989-2008) record representing a period of widely reported accelerating change. The rate of change of biological indicators, season length and accumulated growing-degree days, were also measured over the 56 and 20-year records. Finally, we compared how well interpolated Parameter-elevation Regression on Independent Slopes Model (PRISM) datasets match the quality controlled and weather data from each station. Our results show that warming signals were strongest at mid-elevations over both temporal scales. Over the 56-year record, most sites show warming occurring largely through increases in maximum temperatures, while the 20-year record documents warming associated with increases in maximum temperatures at lower elevations and increases in minimum temperatures at higher elevations. Recent decades have also shown a shift from warming during springtime to warming in July and November. Warming along the gradient has contributed to increases in growing-degree days, although to differing degrees, over both temporal scales. However, the length of the growing season has remained unchanged. Finally, the actual and the PRISM interpolated yearly rates rarely showed strong correlations and suggest different warming and cooling trends at most sites. Interpretation of climate trends and their seasonal biases in the Rocky Mountain Front Range are dependent on both elevation and the temporal scale of analysis. Given mismatches between interpolated data and the directly measured station data, we caution against an over-reliance on interpolation methods for documenting local patterns of climatic change.