Data-model comparisons of the tropical hydroclimate response to the 8.2 ka Event with an isotope-enabled climate model

Abstract

Abstract. The 8.2 ka Event was a prominent climate anomaly that occurred approximately 8200 years before present (8.2 ka) with implications for understanding the mechanisms and characteristics of abrupt climate change. We characterize the tropical hydroclimate response to the 8.2 ka Event based on a multiproxy compilation of 61 tropical hydroclimate records and assess the consistency between the proxy synthesis and simulated hydroclimate anomalies in a new meltwater simulation with the isotope-enabled Community Earth System Model (iCESM1.2). We calculate the timing and duration of the hydroclimate anomalies in our proxy reconstruction using two event detection methods, including a new changepoint detection algorithm that explicitly accounts for age uncertainty. Using these methods, we find significant hydroclimate anomalies associated with the 8.2 ka Event in 30 % of our proxy compilation, with a mean onset age of 8.28 ± 0.12 ka (1σ), mean termination age of 8.11 ± 0.09 ka (1σ), and mean duration of 152 ± 70 years (1σ), comparing well with previous estimates. Notably, these anomalies are not hemispherically uniform, but display a rich regional structure with pronounced drying and/or isotopic enrichment across South and East Asia, the Arabian Peninsula, and in parts of Central America, alongside wetter conditions and/or isotopic depletion in eastern Brazil. In contrast, we find no signature of the 8.2 ka Event over the Maritime Continent. The simulated hydroclimate response to the meltwater event generally agrees with the proxy reconstructions. In iCESM, the North Atlantic meltwater forcing causes a southward shift of the tropical rain bands, resulting in a generally drier Northern Hemisphere and wetter Southern Hemisphere, but with large regional variations in precipitation response, including the isotopic composition of precipitation. Over the oceans, the tropical rainbands shift south and precipitation δ18O (δ18Op) anomalies are generally consistent with the “amount effect,” wherein the change in δ18Op is inversely correlated with the change in precipitation amount. However, the δ18Op anomalies are more decoupled from changes in precipitation amount over land. iCESM captures many of the regional hydroclimate responses observed in the reconstructions, including the large-scale isotopic enrichment pattern in δ18Op in South and East Asia and the Arabian Peninsula, mixed hydroclimate patterns in southern Central America, isotopic depletion in parts of eastern Brazil, and a muted hydroclimate response over the Maritime Continent. We decompose the simulated precipitation δ18O response to identify the cause of these isotopic anomalies, finding that changes in amount-weighted δ18Op arise primarily from seasonal changes in δ18Op rather than seasonal changes in precipitation amount. However, the mechanisms of the seasonal changes in δ18Op vary regionally, with the local amount effect dominant in northeastern South America and the northeastern tropical Pacific; while changes in the isotopic composition of the water vapor (via changes in moisture source, circulation, and/or upstream rainout) seem to control the response in East Asia. In the Caribbean, the addition of isotopically depleted meltwater to the North Atlantic contributes to reduced, but isotopically depleted, wet season precipitation. Overall, this study provides new insights into the tropical hydroclimate response to the 8.2 ka Event, emphasizing the importance of accounting for age uncertainty in proxy-based hydroclimate reconstructions and the value of using isotope-enabled model simulations for data-model intercomparison.