Developing a Data-Driven Global Fire Model to Explore the Impacts of Increasing CO2e on the Global Fire Regime and Vegetation Cover

该博士学位的目的是结合当前的文献，现有的全球火灾模型和新兴的数据集（MODIS植被和火灾数据集，全球火灾地图集等），建立一个经验的全球火灾模型。通过摆脱基于过程的模型，并依靠数据中的突发趋势来确定参数的相对作用，该博士适合由Iain Colin Prentice教授领导的REALM项目的当前工作。根据目前向更简单的地球系统模型发展的趋势，一个想法是将新开发的全球GPP方程纳入该火灾模型（Wang et al., 2017; Stocker et al., 2020）。这将大大简化GPP输入，并可以大大减少围绕不同植物功能类型对火灾的反应（例如可燃性和火灾后再生特性）的不确定性。如果能成功预测燃烧面积和火灾趋势，这种简化将在减少全球火灾控制的不确定性方面向前迈进一步。此外，将探索不同的响应变量（火的辐射功率和火的大小），以摆脱对燃烧面积的依赖。一旦模型具有鲁棒性和功能性，将探讨大气CO2e增加对植被-火灾动态关系的影响。

The aim of this PhD is to combine the current body of literature, existing global fire models and new emerging data sets (MODIS Vegetation and Fire datasets, Global Fire Atlas etc.) to build an empirical global fire model. By moving away from process-based models and relying on emergent trends from data to determine parameters relative role, this PhD fits into the current work of the REALM project led by Professor Iain Colin Prentice. In line with the current move towards simpler Earth System Models, one idea would be to incorporate into this fire model a newly developed global equation for GPP (Wang et al., 2017; Stocker et al., 2020). This would allow for a major simplification in terms of GPP inputs and could greatly reduce uncertainties surrounding how different plant-functional types respond to fire (flammability and post-fire regeneration properties for example). If successful in predicting burnt area and fire trends, this simplification could a step forward in reducing uncertainties concerning of global fire controls. In addition, different response variables will be explored (fire radiative power and fire size) to move away from the reliance on Burnt Area. Once the model is robust and functional, the impact of increasing atmospheric CO2e on the dynamical relationship of vegetation-fire will be explored.