Dust‐Drought Nexus in the Southwestern United States: A Proxy‐Model Comparison Approach. Issue 12 (15th December 2020)

• Record Type: Journal Article
• Authors: Arcusa, S. H.; McKay, N. P.; Carrillo, C. M.; Ault, T. R.
• Is Part Of: Paleoceanography and paleoclimatology. Volume 35:Issue 12(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-12-15
• Subjects: dust -- drought -- U.S. Southwest -- paleo model/data comparison -- age uncertainty -- CESM‐LME; Paleoceanography -- Periodicals; Paleoclimatology -- Periodicals; 551.46
• DOI: 10.1029/2020PA004046
• Languages: English
• ISSNs: 2572-4517

Abstract

Abstract: The U.S. Southwest is naturally prone to dust and drought. With a high risk of unprecedented drought in the future due to climate change, changes in the dust cycle are expected. Whereas paleo records of dust deposition from the region suggest that past megadroughts did not coincide with elevated dust levels, modern studies indicate higher dust during drier conditions. To examine this disagreement, we compare regional paleo records to output from the Community Earth System Model Last Millennium Ensemble experiment. We investigate the role of hydroclimate on dust emissions in the model and analyze how modeled dust during megadroughts compares to all other years. We find that bare ground, as directly modulated by the density of vegetation cover and indirectly by hydroclimate, is the primary control on simulated dust emissions in the Southwest. Furthermore, we find slightly (but significantly) higher dust emission during megadroughts than during years without megadroughts in the model. Using a simple forward modeling approach to assess the impact of age uncertainty in the regional dust records, we find that age uncertainty can explain most of the paleo data/model disagreement. The remaining difference could be related to processes that are not included in the model that inhibit dust emissions during drought, including brown vegetation and soil crusts. Our results suggest the dust‐drought nexus is a continuum and that the impact of drought on dust cycling is subtle. Plain Language Summary: Drylands of the U.S. Southwest naturally experience dust and drought. With a high danger of drought in the future because of climate change, changes in the dust cycle are expected. Whereas lake sediment records of dust deposition from the area suggest that past megadroughts did not happen concurrently with periods with higher dust levels, studies based on instrumental observations of dust point to higher dust during drier conditions. To explain this disagreement, we compare lake sediment records to model simulation. We show that uncovered soil, as directly controlled by the number of plants and indirectly by hydroclimate, is the first control on dust emissions in the Southwest. In addition, we discover only slightly higher dust during megadroughts in the model than in the paleo record. Most of the difference between paleo data and model results could be explained by uncertainties in age measurement. The remaining difference between paleo data and model results could be related to processes that stop dust emissions, including brown plants and soil crusts, that are missing in the model. Our study suggests the relation between dust and drought ranges from non‐existent to not strong and further shows only a slight effect of drought on dust. Key Points: Paleo data/model comparison suggests a subtle dust‐drought nexus in the southwest United States, affected more strongly by land than hydroclimate The Community Earth System Model Last Millennium Ensemble shows higher dust during megadroughts, unlike paleo dust records Most of the disagreement in the paleo data/model comparison could be due to age uncertainty in the paleo dust records