全球变化知识资源中心

Key Points

Parameters of a stochastic weather generator are estimated from 11 bias-corrected simulations of two Representative Concentration Pathways (RCPs) 4.5 and 8.5, and weather observations of the recent past.

We generate one hundred realizations of daily weather for RCP 4.5, RCP 8.5, and ten recent past (PAST) scenarios and input them into a model capturing natural hydrology and water management in an intensively managed system.

Model outputs allow us to quantify probability distributions of allocated and unsatisfied irrigation water and their spatial patterns and indicate that warmer scenarios are associated with higher and more variable unsatisfied demand.

Abstract In intensively managed watersheds, water scarcity is a product of interactions between complex biophysical processes and human activities. Understanding how intensively managed watersheds respond to climate change requires modeling these coupled processes. One challenge in assessing the response of these watersheds to climate change lies in adequately capturing the trends and variability of future climates. Here we combine a stochastic weather generator together with future projections of climate change to efficiently create a large ensemble of daily weather for three climate scenarios, reflecting recent past and two future climate scenarios. With a previously developed model that captures rainfall-runoff processes and the redistribution of water according to declared water rights, we use these large ensembles to evaluate how future climate change may impact satisfied and unsatisfied irrigation throughout the study area, the Treasure Valley in Southwest Idaho, USA. The numerical experiments quantify the changing rate of allocated and unsatisfied irrigation amount and reveal that the projected temperature increase more significantly influences allocated and unsatisfied irrigation amounts than precipitation changes. The scenarios identify spatially distinct regions in the study area that are at greater risk of the occurrence of unsatisfied irrigation. This study demonstrates how combining stochastic weather generators and future climate projections can support efforts to assess future risks of negative water resource outcomes. It also allows identification of regions in the study area that may be less suitable for irrigated agriculture in future decades, potentially benefiting planners and managers.