Study On Effects Of Tempreture And Humidity On Equilibrium Moisture Content Of Fuel

Forest fuel moisture content (FFMC) prediction is the main task of forest fire danger rating system. Nowadays, Equilibrium moisture content (EMC) method is used widely to make the prediction of FFMC. This paper summarized the study and applyment of the method and analyzed its effect factors—tempreture and relative humidity, in order to improve the accuracy of FFMC prediction and the development of the the following research.To study the effects of tempreture and humidity on EMC, we selected several classes of withered twigs with 0.5cm interval(0.5cm≤D≤3cm), barks, withered leaves, humus and litters of Pinus koraiensis, Larix gmelini, Fraxinus mandshurica and Betula platyphylla. Measured the EMC under the conditions of 15℃-35℃and 35%-75% relative humidity', several regression models were established, which used SPSS11.5 and Excel. The results show that: 1) Under the same humidity, different kinds of fuels' EMC go down with the rising tempreture, and the decreasing degree is various. Generally speaking, fine fuels'EMC (D≤0.5cm) are not effected largely by the tempreture, while the bigger ones are on the contrary in a certain tempreture. Escaped that tempreture influence, the EMC-Tempreture curves of the bigger ones change smoothly. 2) Coniferous as Pinus koraiensis and Larix gmelini EMC-Tempreture curves are not different significantly under the various humidity level, which is much the converse of Fraxinus mandshurica and Betula platyphylla. 3) The influences of the tempreture on humus and barks are not severcely, which is opposite to litters and withered leaves for their small mass and the complex composition. 4) Most of the EMC-Tempreture curves apply to the quadratic regression. EMC_max and EMC_min appear at 75% and 35% humidity, 15℃and 35℃respectively. 5) Under the same tempreture, different kinds of fuels' EMC go up with the rising relative humidity, and the rising curves canbe expressed by different kinds of regression models, which is different to the result of Simard,Van Wagner,Anderson and Nelson, who believe that the curves have sigmoid shape characteristics. 6) Simard model, which is accorded with the test data exactly, is the best one among these four EMC prediction models, with the excellent degree up to95%. The model is EMC=b₁+b₂x₁+b₃x₂+b₄x₁²+b₅x₂²+b₆x₁x₂, even though its highly accurate, it still makes lower predicted value than the observed ones.