| Pinus yunnanensis is an endemic widely distributed in the Southwest of China, which is an important forestry tree in wood products. It's reported that pinus yunnanensis can provide timber forest wood products which accounts for 95% in Yunnan province. However, such forests are greatly severed by forest fires in China. In addition, investigations for forest fuel cost enormous labor, material and financial resources , which is equivalent to the workload of investigations for forest resources . Therefore, it is a very important work to get the relationship and distribution between pinus yunnanensis forest fuel load and stand factor using the "3S" technologies, in order to forecast the fire, the fire behavior, and ensure the production of timber.In this article, firstly, based on the data of typical sample survey and 2003 TM Image information of Anning region,143 pieces of fuel load samples data of pinus yunnanensis forest were collected.Secondly , we extracted pinus yunnanensis forest vegetation types according to information from remote sensing and GIS data processing, analysis, and considered similarly gray values with the corresponding sample data ,vegetation index, the ratio of gray, and other qualitative and quantitative factors (aspect, slope, elevation, aspect index,etc.) as the variables regression models, and also considered sample measured fuel load of pinus yunnanensis forest as attributive variable, established fuel load linear regression model to use of quantitative methods and principles. By the study and test quantitative relations of stand and fuel load factor, we estimated pinus yunnanensis fuel load in Anning region.There are four regression models to be adopted, when estimating the pinus yunnanensis fuel load of the sampled plots is estimated. The precision of the regression model with remote sensing factors as variables is 73%; the precision of the regression model with remote sensing and topographical factors as variables is 84%; the precision of the regression model with weight factors as variables, which come from factor analysis is 78.6%. This article studied the indirect estimating method using stand factors such as average tree height, canopy density, stand average diameter at breast height as middle variables, but regression equation by these models aren't significant except that model with canopy density as variables. Although the precision of the regression model with canopy density as variables is 88%, but the precision of the fuel load model with canopy density and ages as variables is only 65%. We compared precisions of each fuel load estimating models, and chose this model with remote sensing(TM1,NDVI,TM2,RVI,TM4,PVI,TM3,VI3,TM5,TM7,TM4*TM3/TM7,TM3/∑(2+3+4+5+7),TM(4+5-2)/(4+5+2),TM7/TM3) and topographical factors(slope,aspect-index,aspect,elevation) as variables. The model is as bellows: yyyy == 1112 23.... 999773 -2225 58.... 222668X1 ++ 66 1.... 999556X2– 8.. 855 0X33 ++ 111100009999.... 999661X4 ++ 77772222.... 000887X5 ++ 33330000.... 999775X6– 777.. 100 9X7 ++ 55556666.... 999446X8 - 111122229999.... 000229X9 ++ 66 8.. 65556 69X11 0 ++ 111144448888.... 444113 X11 1 ++ 55558888....2222 99955X1112 2 -11 6.. 86666 44411X1113 3 ++++ 111100005555....1111 66600X1114 4 - 11116666....5555 00066X11 5 + 0000 ... 21111 88833X11 6 ++++ 88882222....8888 22233X11 7 ++++ 3333....4444 99933X11 8We estimated initially the pinus yunnanensis fuel load,and got the pinus yunnanensis fuel load distribution map of the Anning region.The minimum value is 0.135t.hm-2, The maximum value is 40.6961t.hm-2, The mean is 16.1566 t.hm-2... |
PDF Full Text Request