| 摘要: |
| 以广东省森林资源连续清查体系为基础，采用系统成数抽样方法分别产出2012年、2016年和2017年全省森林资源面积数据，对比分析2016—2017年、2012—2016年和2012—2017年3个间隔期森林覆盖率变化量、抽样精度、变化趋势判断等，并通过成本比较基于遥感抽样和遥感全面更新的优化方案，提出可行的优化措施。结果表明：全省2012、2016和2017年森林覆盖率分别为51.26%、53%、53.51%，平均抽样精度达到96.93%，2016—2017年、2012—2016年和2012—2017年间全省森林覆盖率净增量分别为0.51%、1.74%、2.25%，抽样精度低于53.03%。基于现行连清体系估计的森林覆盖率年度或周期变化均无精度保证，利用高分辨率遥感影像判读加密样地可实现精度保证、成本可接受的森林覆盖率年度变化监测，较遥感全面更新效率更高、成本更低。 |
| 关键词: 森林资源连续清查 森林覆盖率 森林面积 年度出数 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| ExperimentalAnalysis ofAnnual Output of Provincial ForestArea Based on National Forest Inventory System |
|
Wang Qiulai1, Xue Chunquan2, Lin Shouming2, Chen Chuanguo2, Yang Zhigang2, Wen Xiaorong1, Ye Jinsheng1, Zheng Wensong1
|
|
1.Nanjing Forestry University,Nanjing;2.Forestry Surveying and Designing Institute of Guangdong Province,Guangzhou
|
| Abstract: |
| In this paper experimental analysis of annual output of provincial forest area had been taken based on NFI of Guangdong province. Forest resource area of the province in 2012,2016 and 2017 were output using systematic percentage sampling, and variation, sampling precision, judgment trends of change of forest cover rate during three intervals which were 2016-2017,2012-2016, and 2012-2017 had been compared and analyzed. Sampling methods with remote sensing and fully update with remote sensing mentioned as optimization program had been compared about cost, and some reasonable optimization measures were proposed. The results showed that the province forest cover rate were 51.26% in 2012, 53.0% in 2016, and 53.51% in 2017, the average sampling accuracy was 96.93%, and net increments of the province forest cover rate during 2016-2017,2012-2016,2012-2017 were 0.51%,1.74%,2.25% respectively, while sampling precision was lower than 53.03%. In general, annual or period variation of provincial forest cover rate output by the currently NFI was unreliable and unwarranted due to low sampling precision. The systematic sampling or simple sampling which can estimate change of provincial forest cover rate with high sampling precision and accepted cost by identi?cation of more plot density with remote sensing were more efficient and cheaper rather than the method of fully update with remote sensing. |
| Key words: national forest inventory forest cover rate forest area annual data output |
