全球变化知识资源中心

Decision-makers using climate projection information are often faced with the problem of data breadth, complexity, and uncertainty, which complicates the translation of climate science products in addressing management challenges. Recently, the concept of climate scenario planning attempts to simplify climate information by developing a series of plausible future storylines. In some cases, however, these storylines lack quantitative detail on extremes that may be useful to decision-makers. Here, we analyse a large suite of statistically downscaled climate projections from two methods to develop quantitative projections for hydrologic extremes (heavy precipitation and drought) across Oklahoma and Texas in the United States. Downscaled projections are grouped into four specific temperature/precipitation scenarios, including Warm/Wet, Hot/Dry, Central Tendency, and the full multi-model ensemble average. The region is split into three sub-domains spanning the region's west-east precipitation gradient, and projections are examined throughout the mid- and late-21st century, using two emissions scenarios (mid-range and high). Most scenarios project increased frequency and duration of moderate or greater drought across the whole domain, with the high-emissions Hot/Dry projections showing the most severe examples. The Warm/Wet scenario also increases the frequency of dry months, particularly in the Southern High Plains, but does not discernably alter duration, and retains a similar frequency of pluvial (wet) periods. The mid-range projections generally retain similar evolutions among scenarios, but they reduce drought intensity and project no change in drought/pluvial frequency with the Warm/Wet scenario. Notably, the occurrence of intense precipitation increases across all scenarios and emissions categories and does not significantly differ between any of the scenarios, including Hot/Dry versus Warm/Wet. Some observed differences in extreme precipitation magnitudes between the two downscaled data sets are briefly discussed.