Did the Roman Empire affect European climate? A new look at the effects of land use and anthropogenic aerosol emissions

Abstract

Abstract. As one of the first transcontinental polities that led to widespread anthropogenic modification of the environment, the influence of the Roman Empire on European climate has been studied for more than 20 years. Recent advances in our understanding of past land use and aerosol-climate interactions make it valuable to revisit the way humans may have affected the climate of the Roman Era. Here we drive the global aerosol-enabled climate model ECHAM-HAM-SALSA with land use maps and novel estimates of anthropogenic aerosol emissions from the Roman Empire at its apogee to quantify the effect of humans on regional climate. In a factorial study, we used the HYDE and KK11 anthropogenic land cover change scenarios with three estimates (low, medium, high) of aerosol emissions from fuel combustion and burning of agricultural land. Land use effects on climate varied from no influence using the HYDE scenario to a significant warming over land of 0.15 K with KK11 relative to a no-land use control. This warming is primarily caused by regional decreases in turbulent fluxes, in contrast to previous studies that emphasised changes in albedo and evapotranspiration. Aerosol emissions from agricultural burning were greater than those from fuel consumption, but on the same order of magnitude. All emissions scenarios result in an enhanced cooling effect of clouds over a no-emissions control scenario. As a consequence, the land surface temperature averaged over our entire study domain decreased significantly by 0.17 K, 0.23 K, and 0.46 K for the low, the intermediate, and the high emissions scenarios, respectively. Cooling caused by aerosol emissions is largest over Central and Eastern Europe, while warming caused by land use occurs in parts of North Africa and the Middle East. Our results suggest that the influence of Roman Era anthropogenic aerosol emissions on European climate may have been as important as that of deforestation and other forms of land use. Our model may overestimate aerosol-effective radiative forcing, however, and our results are very sensitive to the inferred seasonal timing of agricultural burning practices and natural aerosol emissions over land (wildfire emissions and biogenic emissions). Nevertheless, it is likely that human influence on land and the atmosphere affected continental-scale climate during Classical Antiquity.