山东省森林资源动态变化的非等间距灰色预测

应用灰色系统理论,以1974-2007年山东省森林覆盖率、有林地面积和活立木蓄积量为时间数据序列,建立非等距灰色预测模型。用模型对山东省森林资源主要指标进行拟合,拟合误差均在10%以下,拟合精度较高。使用后验差检验方法对模型进行精度等级检验,求得森林覆盖率、有林地面积和森林蓄积量的后验差检验值分别为0.15,0.23和0.14,均小于0.35,小误差概率均为1,预测精度等级为一级好,预测的结果可以作为制订山东省未来林业发展计划的依据,为建立“数字林业”等提供了一种比较理想的预测方法。预测结果表明：2020年森林覆盖率为26.78%,还达不到2020年森林覆盖率达到30%以上的生态省建设阶段性目标。因此,山东省应加快森林资源发展,稳步提高森林覆盖率,明显增加森林蓄积量,提高林分质量,避免造成森林资源的逆转。

Non-equidistance gray mode forecast of forest resource dynamic changes in Shandong Province

Non-equidistance grey models of time-series data including forest cover rate, forested area and living trees volume in Shandong Province between 1974 - 2007 were established based on the Grey Systems Theory. The model fitted well with the main outcomes of forest resource measurement in Shandong Province. The fitting errors were lower than 10 percent. The models had ideal fitting precision. Posterior variance test method was used to test the precision grade of the models. The test results showed that the posterior variance test c values of forest cover rate, forested area and living trees volume were 0.15, 0.23 and 0.14, which were all lower than 0.35. Small error possibility was 1. The forecast precision of the models reached the ＂first grade＂. The paper provides a more ideal forecast method for constructing ＂Digital Forestry＂. It is a basis for formulating the future development of the forestry program in Shandong Province. The forecast results showed that forest cover rate would be 26.78% in 2020, which was lower than 30%, the phased target of construction of ecology province. Therefore, it is very important to promote the development of forest resource, steadily improve forest cover rate, significantly increase forest volume, improve forest quality, so as to avoid the degeneration of forest resource. [Ch, 4 tab. 15 ref.]