Abstract:

It is necessary to conduct national forest inventory and generating annual estimates of two binding targets,which are forest coverage and forest volume,for the annual examination and evaluation of green development targets and ecological civilization progress targets.Based on the data of 1 669 permanent plots of the 8^th and 9^th national forest inventories in Beijing,and according to the inventory plan “1/5 of plots being completed every year with a 5-year cycle”,five sets of sample plots were selected by systematic sampling and simple random sampling,respectively.The combined estimation method and double regression method were used to generate annual estimates of forest volume and the precision in the first year.The results show that the precisions of forest volume estimates from two methods are more than 90%. The study chose the 9^th inventory result of forest volume in Beijing as the base for comparison and found that the relative differences are between -1.93%~3.73% for the estimates based on combined estimation method,and the those are between -2.61%~4.98% for the estimates based on double regression method.The following conclusions can be achieved:The sample plots selected by systematic sampling are more representative than those selected by simple random sampling;Both combined estimation method and double regression method could be used to generate forest volume estimates for the first year and the subsequent years;Combined estimation method is more applicable and little better than double regression method,which can be used as an approximate simple method.

Key words: forest volume, sampling survey, regression model, combined estimation, double regression