Distance of meteorological data from test site and fine fule forecast model accuracy

Abstract

To determine the influence of meteorological data at different distances from a research site on the accuracy of a prediction model for water content of fine fuels, a model was established and tested in the spring and autumn of 2010. Continuous observations were conducted on the moisture content of fine fuels in Pinus sylvestris var. mongolica, Larix gmelinii, and Betula platyphylla stands in the Pangu Forest Farm of Tahe County, Da Xingan Ling Prefecture and compared to meteorological data such as precipitation (R), total precipitation for n d (Ran), average relative humidity (Han), daily maximum temperature (Tmax), and drought days (d). Results indicated that for meteorological factors in spring and autumn, R was positively correlated with Ran and Han (n=1-2), but negatively correlated with Tmax and d. The model using the meteorological data of the Pangu Meteorological Station was more accurate than the Tahe Model which was further away having a significantly lower mean absolute error (EMA) of 2.7% and mean relative error (EMR) of 22.6% (P < 0.01). From the heterogeneity of the fuels and the accuracy of the model, attention should be paid to the selection of time and the spatial scale of the meteorological factors when modeling meteorological data collection so as to ensure model accuracy.