巴西里约热内卢州立大学的Newton de Magalhães Neto及同事利用2000年至2016年间收集的火灾事件、烟羽流动、降水和冰川融化数据建模，评估亚马孙流域的生物体燃烧对玻利维亚Zongo冰川可能造成的影响。他们发现生物体燃烧产生的气溶胶（比如炭黑）可以随风扩散到安第斯山脉热带冰川，沉降在雪中；由于被炭黑或粉尘覆盖的积雪反射的光减少（返照率下降），冰川融化可能因此加剧。
摘要：The melting of tropical glaciers provides water resources to millions of people, involving social, ecological and economic demands. At present, these water reservoirs are threatened by the accelerating rates of mass loss associated with modern climate changes related to greenhouse gas emissions and ultimately land use/cover change. Until now, the effects of land use/cover change on the tropical Andean glaciers of South America through biomass burning activities have not been investigated. In this study, we quantitatively examine the hypothesis that regional land use/cover change is a contributor to the observed glacier mass loss, taking into account the role of Amazonian biomass burning. We demonstrated here, for the first time, that for tropical Andean glaciers, a massive contribution of black carbon emitted from biomass burning in the Amazon Basin does exist. This is favorable due to its positioning with respect to Amazon Basin fire hot spots and the predominant wind direction during the transition from the dry to wet seasons (Aug-Sep-Oct), when most fire events occur. We investigated changes in Bolivian Zongo Glacier albedo due to impurities on snow, including black carbon surface deposition and its potential for increasing annual glacier melting. We showed that the magnitude of the impact of Amazonian biomass burning depends on the dust content in snow. When high concentration of dust is present (e.g. 100 ppm of dust), the dust absorbs most of the radiation that otherwise would be absorbed by the BC. Our estimations point to a melting factor of 3.3 ± 0.8% for black carbon, and 5.0 ± 1.0% for black carbon in the presence of low dust content (e.g. 10 ppm of dust). For the 2010 hydrological year, we reported an increase in runoff corresponding to 4.5% of the annual discharge during the seasonal peak fire season, which is consistent with our predictions.
