COMPARISON OF THREE VAPOUR-EXCHAN GEMETHODS FOR DESCRIBING HOURLY LITTER MOSITURE CONTENT OF TYPICAL FOREST STANDS IN DAXINGANLING PREFECTURE, NORTHEASTERN CHINA

摘要

Advances in fire behavior prediction in Dax- inganling Prefecture, where locates at northeast part of China are driving the need for fuel moisture pre- dictions that go beyond what is possible with fuel moisture models currently in use. To establish accu- rate fuel moisture models, hourly measurements were taken of moisture content in litter beds of dif- ferent forest stands like Larix gmelinii, Betula platyplzylla and Pinus sylvestris var. mongolica un- der open, moderate, and shade conditions from Sep- tember to October in 2010. Models were established using three vapour-exchange methods such as Nel- son and Simard methods, which employed the func- tion of equilibrium moisture content (EMC), and di- rect regression methods. The former two methods also used a timelag concept, and latter method esti- mating model parameters directly from fuel moisture and weather observation data in the field. The results indicated that the Nelson method correctly predicted hourly litter moisture content in all forest stands, fol- lowed by Simard method, and direct regression method. Different models generated from forest un- der shading conditions differently, but the coeffi- cient of determinations of these models predicted by three methods were not obvious. Moreover, the ef- fect of Simard method by extrapolation was the most satisfied for its minor deviations both on itself and calculations of extrapolation matrix, while direct re- gression method is the least satisfied predication model. Consider of tlie Simard and Nelson had sim- ilar accuracy, but Simard was more robust and sim- ple estimation, therefore is recommended for pre- dicting litter moisture in this region.