Abstract
Detecting long-term change in seasonal precipitation using ground observations is dependent on the representativity of the point measurement to the surrounding landscape. In mountainous regions, representativity can be poor and lead to large uncertainties in precipitation estimates at high elevations or in areas where observations are sparse. If the uncertainty in the estimate is large compared to the long-term shifts in precipitation, then the change will likely go undetected. In this analysis, we examine the minimum detectable change across mountainous terrain in western Montana, USA. We ask the question: What is the minimum amount of change that is necessary to be detected using our best estimates of precipitation in complex terrain? We evaluate the spatial uncertainty in the precipitation estimates by conditioning historic regional climate model simulations to ground observations using Bayesian inference. By using this uncertainty as a null hypothesis, we test for detectability across the study region. To provide context for the detectability calculations, we look at a range of future scenarios from the Coupled Model Intercomparison Project 5 (CMIP5) multimodel ensemble downscaled to 4 km resolution using the MACAv2-METDATA data set. When using the ensemble averages we find that approximately 65% of the significant increases in winter precipitation go undetected at midelevations. At high elevation, approximately 75% of significant increases in winter precipitation are undetectable. Areas where change can be detected are largely controlled by topographic features. Elevation and aspect are key characteristics that determine whether or not changes in winter precipitation can be detected. Furthermore, we find that undetected increases in winter precipitation at high elevation will likely remain as snow under climate change scenarios. Therefore, there is potential for these areas to offset snowpack loss at lower elevations and confound the effects of climate change on water resources.
| Original language | English |
|---|---|
| Pages (from-to) | 4301-4320 |
| Number of pages | 20 |
| Journal | Water Resources Research |
| Volume | 52 |
| Issue number | 6 |
| DOIs | |
| State | Published - Jun 1 2016 |
Funding
We plan to make these data available through the Montana Institute on Ecosystems Data Gateway. This repository will be a DataONE member node but is currently under development. In the meantime, we offer these data on request through e-mail with the corresponding author. The MACAv2-METDATA data, Livneh data, GFS-FNL data, and the WRF model are all freely available for download online. The data set MACAv2-METDATA was produced with funding from the Regional Approaches to Climate Change (REACCH) project and the SouthEast Climate Science Center (SECSC). We acknowledge the World Climate Research Programme's Working Group on Coupled Modeling, which is responsible for CMIP, and we thank the climate modeling groups for producing and making available their model output. For CMIP, the U.S. Department of Energy's Program for Climate Model Diagnosis and Intercomparison provides coordinating support and led development of software infrastructure in partnership with the Global Organization for Earth System Science Portals. This research has been supported by the Montana Space Grant Consortium NASA EPSCoR, NSF GSS-1461576, NSF EPS-1101342, and the United States Environmental Protection Agency, Science To Achieve Results (STAR) fellowship. We thank Johnnie Moore, University of Montana Geosciences, and William Kleiber, University of Colorado, Boulder, for their helpful comments. Thanks to Ethan Gutmann, National Center for Atmospheric Research, for his thorough and thoughtful review. Also, we acknowledge Shu-Hua Chen, Department of Land, Air and Water Resources at the University of California, Davis, for her help with the GFS-WRF climate simulations.
| Funder number |
|---|
| 1443108, EPS-1101342, GSS-1461576 |
Keywords
- Bayesian
- SNOTEL
- WRF
- complex terrain
- precipitation
- uncertainty
