Search

Multiyear droughts are a common occurrence in southwestern North America (SWNA), but it is unclear what causes these persistent dry periods. The ocean‐atmosphere conditions coinciding with droughts have traditionally been studied using correlation and composite methods, which suggest that cool conditions in the tropical Pacific are associated with SWNA droughts and warm conditions are associated with wet periods in SWNA. Nevertheless, the extent to which multiyear droughts are truly consistent with this paradigm remains unknown. This is, in part, because the temporal trajectory of ocean‐atmosphere conditions during these dry periods have not been sufficiently characterized. Here we examine the continuum of ocean‐atmosphere trajectories before, during, and after multiyear droughts in SWNA using observation‐based data and an ensemble of climate model simulations from the Community Earth System Model. An examination of sea surface temperature patterns at the beginning, middle, and end of SWNA droughts shows that an El Niño event tends to precede SWNA droughts, a cool tropical Pacific occurs during droughts, and central Pacific El Niño events end droughts. However, moderate El Niño events can occur in the middle of persistent droughts, so a warm tropical Pacific does not always end these dry periods. These findings are important for drought predictability and emphasize the need to improve simulations of the magnitude, life cycle, and frequency of occurrence of El Niño events. Plain Language Summary: Droughts lasting multiple years in southwestern North America (SWNA) have severe consequences for both humans and natural systems. Unusually cold tropical Pacific conditions (e.g., La Niña events) typically coincide with dry years in SWNA, and it is often assumed that persistent cold conditions in the tropical Pacific coincide with persistent droughts. Here we test this idea using climate model simulations and observation‐based data. Although a cool tropical Pacific tends to initiate SWNA droughts and a warm tropical Pacific (e.g., El Niño events) tends to terminate droughts, we show that El Niño events can occur in the middle of SWNA droughts without ending these dry periods. We also examine patterns of sea surface temperatures before, during, and after SWNA droughts. We find that El Niño events tend to precede SWNA droughts and central Pacific El Niño events often terminate droughts in both observations and model data; these El Niño events are perhaps the most consistent feature of multiyear droughts in SWNA. These findings have important implications for public perception of droughts and their causes, in addition to our scientific understanding and modeling of El Niño and its impacts. Key Points: We examine climate patterns associated with the life cycle of droughts in southwestern North America (SWNA) in observations and model dataEastern Pacific ENSO events often precede droughts, cool Pacific anomalies maintain droughts, and central Pacific ENSO events end droughtsThe state of the North Pacific and the type of El Niño event may modulate the likelihood of an El Niño event ending a drought [ABSTRACT FROM AUTHOR]

Drought has severe consequences for humans and their environment, yet we have a limited understanding of the drivers of drought across the full range of time scales on which it occurs. Here, the atmosphere and ocean conditions that drive this continuum of drought variability in southwestern North America (SWNA) are studied using the latest observationally based products, paleoclimate reconstructions, and state-of-the-art Earth system model simulations of the last millennium. A novel application of the self-organizing maps (SOM) methodology allows for a visualization of the continuum of climate states coinciding with thousands of droughts of varying lengths in last millennium simulations from the Community Earth System Model (CESM), the Goddard Institute for Space Studies Model E2-R (GISS E2-R), and eight other members from phase 5 of the Coupled Model Intercomparison Project (CMIP5). It is found that most droughts are associated with a cool Pacific decadal oscillation (PDO) pattern, but persistent droughts can coincide with a variety of ocean–atmosphere states, including time periods showing a warm PDO or weak ocean–atmosphere anomalies. Many CMIP5 models simulate similar SWNA teleconnection patterns, but the SOM analysis demonstrates that models simulate different continuums of ocean–atmosphere states coinciding with droughts of different lengths, suggesting fundamental differences in their drought dynamics. These findings have important implications for our understanding and simulation of the drivers of persistent drought, and for their potential predictability. [ABSTRACT FROM AUTHOR]

Multidecadal hydroclimate variability has been expressed as "megadroughts" (dry periods more severe and prolonged than observed over the twentieth century) and corresponding "megapluvial" wet periods in many regions around the world. The risk of such events is strongly affected by modes of coupled atmosphere-ocean variability and by external impacts on climate. Accurately assessing the mechanisms for these interactions is difficult, since it requires large ensembles of millennial simulations as well as long proxy time series. Here, the Community Earth System Model (CESM) Last Millennium Ensemble is used to examine statistical associations among megaevents, coupled climate modes, and forcing from major volcanic eruptions. El Niño-Southern Oscillation (ENSO) strongly affects hydroclimate extremes: larger ENSO amplitude reduces megadrought risk and persistence in the southwestern United States, the Sahel, monsoon Asia, and Australia, with corresponding increases in Mexico and the Amazon. The Atlantic multidecadal oscillation (AMO) also alters megadrought risk, primarily in the Caribbean and the Amazon. Volcanic influences are felt primarily through enhancing AMO amplitude, as well as alterations in the structure of both ENSO and AMO teleconnections, which lead to differing manifestations of megadrought. These results indicate that characterizing hydroclimate variability requires an improved understanding of both volcanic climate impacts and variations in ENSO/AMO teleconnections. [ABSTRACT FROM AUTHOR]