Upcoming wildfiresmodeling plant moisture dynamics under climate change

Laburpena

Spatial and temporal variation in the moisture content of leaves, twigs, and small diameter branches (live fuel moisture content; LFMC) is a critical driver of the patterns of wildfire activity. LFMC conditions ecosystem flammability by affecting the availability of plant biomass (fuels) to fire and also the degree of spatial connectivity of dry patches. However, despite the crucial role of understanding LFMC dynamics for anticipating wildfire danger, there are still significant knowledge gaps regarding the estimation of temporal and spatial variations, particularly under ongoing global warming. The main purpose of this PhD Thesis is to improve current and future assessments of LFMC by adding more biological realism in LFMC estimations. To this end, I first simulated how seasonal variation in LFMC affects fire behavior. Secondly, I developed a semi-mechanistic model that considers key physiological traits to estimate species-specific LFMC dynamics. Then, I applied this model to forecast LFMC trends under climate change conditions assessing potential wildfire danger increases, including also estimates of dead fuel moisture content (DFMC). Finally, using a key driver of both LFMC and DFMC, I estimated how projected increases in fire danger under global warming may compromise the potential of reforestation strategies to mitigate climate change through carbon sequestration. Simulation results indicated that LFMC declines increase the likelihood of extreme fire behavior and, therefore, disregarding LFMC from fire modeling would lead to wildfire risk underpredictions. The development of a new semi-mechanistic model to infer LFMC, considering plant physiological traits, significantly increased predictive capabilities. The application of this approach to forecast LFMC dynamics allows to project significant increases in the fire season over large parts of peninsular Spain, in a manner that was inversely proportional to the gradient in productivity. Finally, projected increases in future wildfire risks reduce the limited potential of reforestation strategies to offset anthropogenic CO2 emissions, highlighting that, previous estimates on the potential carbon removal by plantations have been greatly overestimated. Overall, I conclude that we may be at the brink of a dramatical increase in the incidence of large wildfires, as fuel dry-down processes induced by climate change become more frequent and intense. The process-based understanding of LFMC dynamics proposed here may serve to anticipate critical transitions in forest flammability and consequently improve our preparedness for increased wildfire impacts in the coming decades. Keywords: Fuel Moisture, Fire Danger, Fire Modeling, Plant Hydraulics, Climate Change, Pyrophysiology.