山东黄河流域森林资源空间格局及其评价研究

山东黄河流域森林资源空间格局及其评价研究就是突破传统的森林资源监测与评价模式，利用少量地面样地调查数据，结合RS和GIS技术，借助数学和物理理论，建立以象元为单位的林分郁闭度、蓄积量估测模型，分析森林资源的数量、质量、结构和生态并对其进行评价，为实现流域森林生态系统的平衡、最大限度地发挥森林的生态、经济和社会效益提供科学依据。 　　该文采用2000、2006年两期时相相近的TM/ETM+遥感影像作为研究区的基本数据源，利用GPS、全站仪等仪器，采取典型随机抽样方法采集了378个抽样点的外业数据，借助MATLAB、SPSS、SAS、Microsoft Office Execl等统计分析软件、ERDAS、ArcGIS、Fragstats等建模及空间分析工具，运用主成分分析法，多元线性逐步回归法、空间叠加分析法、模糊数学法、层次分析法、线性规划法研究了山东黄河流域森林资源的空间格局和森林景观，并对其进行了综合评价。主要研究结论有： 　　(1)通过监督分类、非监督分类和专家分类等方法得出流域2000、2006年的森林覆盖率分别为21.4％、25.6％。6年间有林地转移速率最大，其减少量中大部分转化为农田和建设用地。宜林荒山荒地减少量中，有21％转化为有林地，11％转化为农田。农田的减少量中，有12％转化为有林地。很明显，流域的宜林荒山荒地得到大力改造，国家的退耕还林政策开始起到明显的成效。通过主成分分析法和多元线性逐步回归法建模估测出流域的林分蓄积量由2000年的1277万m3增加到2006年的1326万m3。2000年，中龄林和中幼龄林的蓄积量分别占82％、7％，到2006年，中幼龄林的蓄积量迅速增加，达到35％，中龄林的蓄积量只占50％。流域大于海拔200m的林分蓄积量仅占14％，平原的林分蓄积量高达86％，6年间蓄积量海拔梯度变化不大。 　　(2)利用Fragstats景观软件计算并分析了森林景观的21个代表性指数。山东黄河流域农田的斑块面积最大，其占流域总面积的48％，其次是有林地，约占26％，第三是建设用地，约占13％。森林景观中，宜林荒山荒地斑块面积最大，灌木林最小。阔叶林和针叶林的斑块数和斑块密度较大。灌木林的平均斑块面积最大，经济林最小。阔叶林、针叶林的景观破碎化程度高，灌木林破碎化程度最低，宜林荒山荒地的连通性最强。6年间，景观指数变化最大的是宜林荒山荒地和阔叶林景观，其中荒山荒地的斑块数减少了近2倍，斑块的连通性急速提高，单位面积急剧增加。阔叶林景观的斑块数增加近50％，斑块平均面积大幅减小。总之，山东黄河流域森林景观虽然得到改善，但是，森林覆盖率不高、荒山荒地比重较大、森林景观破碎化程度高、水土流失严重、森林生态系统非常不稳定是其主要特点。 　　(3)通过层次分析与模糊数学综合评价法对流域森林资源的空间格局进行了评价。研究表明，流域的森林资源分布极不均匀，济南、泰安、莱芜的森林覆盖率超过30％，但三个地市宜林荒山荒地的比例都超过10％，其它七个地市的森林覆盖率只有15％。森林资源空间格局方面，泰安市优，济南市和莱芜市良，淄博市、滨州市、菏泽市、东营市和德州市中等，济宁市和聊城市较差。6年间，泰安市的森林资源空间格局评价由良好到优秀，济南市和莱芜市的评价得分值虽然都较高，但是，济南市朝好的方向发展，相反，莱芜市的得分值不升反降，而且变坏的趋势明显。淄博市、滨州市、菏泽市、东营市和德州市的得分值虽然没有达到良好标准，但其变好的趋势非常明显，济宁市和聊城市的森林资源空间格局较差。 　　(4)利用线性规划法对山东黄河流域的宜林荒山荒地进行林种的优化配置和改造后，流域单位面积林分蓄积量、有林地百分比、针阔叶混交林百分比、经济林百分比和林种结构得到明显改善，森林资源空间格局由中等达到良好的分级标准。 　　论文的创新点： 　　(1)近些年，众多学者对黄河中上游地区的森林资源进行了大量研究，但对黄河下游地区的森林资源研究的较少。山东黄河流域是黄河入海的最后区域，研究该地区的森林资源及其动态对于研究整个黄河流域的森林资源具有重要意义。该文利用广义3S技术将森林资源空间格局、森林景观生态构成一个整体进行定量和动态研究，其研究成果对山东黄河流域的森林资源经营和管理、山东黄河的治理具有重要的意义。 　　(2)该文基于TM/ETM+遥感影像和相关因子定量估测了流域森林资源的蓄积量、郁闭度，采用层次分析模糊综合评价法从森林资源、森林景观生态、环境和社会经济方面对森林资源空间格局进行了综合评价，其研究方法具有创新性。 　　(3)该文基于线性规划的方法对山东黄河流域的宜林荒山荒地和林种结构进行了优化配置，在不改变森林覆盖率的情况下，可以大幅度地改善森林资源的空间格局，提高森林资源的生态、经济和社会效益，该研究方法具有理论意义和实践价值。

This study on the spatial pattern and evaluation of forest resources in the Shandong section of the Yellow River Basin breaks through traditional forest resource monitoring and evaluation models. Using a small amount of field plot survey data, combined with Remote Sensing (RS) and Geographic Information System (GIS) technologies, and leveraging mathematical and physical theories, it establishes stand canopy closure and stock volume estimation models based on pixel units. It analyzes the quantity, quality, structure and ecology of forest resources and conducts evaluations, providing scientific basis for achieving the balance of the basin's forest ecosystem and maximizing the ecological, economic and social benefits of forests.

This study adopted two phases of TM/ETM+ remote sensing images with similar phenological phases in 2000 and 2006 as the basic data sources for the study area. Using instruments such as GPS and total stations, it collected field survey data from 378 sampling points via typical random sampling method. With the help of statistical analysis software including MATLAB, SPSS, SAS and Microsoft Office Excel, as well as modeling and spatial analysis tools such as ERDAS, ArcGIS and Fragstats, it applied Principal Component Analysis (PCA), Multiple Linear Stepwise Regression, Spatial Overlay Analysis, Fuzzy Mathematics Method, Analytic Hierarchy Process (AHP) and Linear Programming to study the spatial pattern and forest landscape of forest resources in the Shandong section of the Yellow River Basin, and conducted a comprehensive evaluation.

Main research conclusions are as follows:
(1) Through methods such as supervised classification, unsupervised classification and expert classification, it is concluded that the forest coverage rates of the basin in 2000 and 2006 were 21.4% and 25.6% respectively. Over the six years, wooded land had the highest transfer rate, with most of its reduced area converted into farmland and construction land. Among the reduced area of afforestation-worthy barren hills and wastelands, 21% was converted into wooded land and 11% into farmland. Among the reduced area of farmland, 12% was converted into wooded land. Obviously, the afforestation-worthy barren hills and wastelands in the basin have been vigorously transformed, and the national Grain for Green Project has begun to achieve obvious results. Through modeling with PCA and Multiple Linear Stepwise Regression, it is estimated that the stand stock volume of the basin increased from 12.77 million m³ in 2000 to 13.26 million m³ in 2006. In 2000, the stock volumes of middle-aged stands and young and middle-aged stands accounted for 82% and 7% respectively. By 2006, the stock volume of young and middle-aged stands increased rapidly to 35%, while that of middle-aged stands only accounted for 50%. The stand stock volume of areas with elevation above 200m in the basin only accounted for 14%, while that of plain areas reached as high as 86%. The change in stock volume along the elevation gradient was small over the six years.
(2) 21 representative indices of the forest landscape were calculated and analyzed using Fragstats landscape software. In the Shandong section of the Yellow River Basin, farmland had the largest patch area, accounting for 48% of the total basin area, followed by wooded land (about 26%) and construction land (about 13%). Among the forest landscapes, afforestation-worthy barren hills and wastelands had the largest patch area, while shrub forests had the smallest. Broad-leaved forests and coniferous forests had relatively large numbers of patches and patch densities. Shrub forests had the largest mean patch area, while economic forests had the smallest. Broad-leaved forests and coniferous forests had a high degree of landscape fragmentation, while shrub forests had the lowest degree of fragmentation, and afforestation-worthy barren hills and wastelands had the strongest connectivity. Over the six years, the landscape indices with the largest changes were those of afforestation-worthy barren hills and wastelands and broad-leaved forest landscapes. For the barren hills and wastelands, the number of patches decreased by nearly twice, the patch connectivity increased rapidly, and the mean patch area increased sharply. For broad-leaved forest landscapes, the number of patches increased by nearly 50%, while the mean patch area decreased significantly. In general, although the forest landscape of the Shandong section of the Yellow River Basin has been improved, its main characteristics are low forest coverage rate, large proportion of barren hills and wastelands, high degree of forest landscape fragmentation, severe soil erosion and extremely unstable forest ecosystem.
(3) The spatial pattern of forest resources in the basin was evaluated using the comprehensive evaluation method combining AHP and fuzzy mathematics. The study shows that the distribution of forest resources in the basin is extremely uneven. The forest coverage rates of Jinan, Tai'an and Laiwu exceed 30%, but the proportion of afforestation-worthy barren hills and wastelands in these three prefecture-level cities exceeds 10%, while the forest coverage rates of the other seven prefecture-level cities are only 15%. In terms of the spatial pattern of forest resources, Tai'an City has excellent performance, Jinan City and Laiwu City have good performance, Zibo City, Binzhou City, Heze City, Dongying City and Dezhou City have medium performance, while Jining City and Liaocheng City have poor performance. Over the six years, the evaluation score of the spatial pattern of forest resources in Tai'an City improved from good to excellent. Although the evaluation scores of Jinan City and Laiwu City were both relatively high, Jinan City developed in a good direction. On the contrary, the score of Laiwu City decreased instead of increasing, and the trend of deterioration was obvious. Although the scores of Zibo City, Binzhou City, Heze City, Dongying City and Dezhou City did not meet the good standard, their trend of improvement was very obvious. The spatial pattern of forest resources in Jining City and Liaocheng City is relatively poor.
(4) After optimizing the allocation and transformation of forest species for the afforestation-worthy barren hills and wastelands in the Shandong section of the Yellow River Basin using linear programming method, the stand stock volume per unit area, the percentage of wooded land, the percentage of mixed broad-leaved and coniferous forests, the percentage of economic forests and the forest species structure of the basin have been significantly improved, and the spatial pattern of forest resources has reached the good grading standard from medium.

Innovations of this paper:
(1) In recent years, many scholars have conducted a large number of studies on forest resources in the middle and upper reaches of the Yellow River, but there are relatively few studies on forest resources in the lower reaches of the Yellow River. The Shandong section of the Yellow River Basin is the final area where the Yellow River flows into the sea. Studying the forest resources and their dynamics in this area is of great significance for the study of forest resources in the entire Yellow River Basin. This paper uses generalized 3S technology to conduct quantitative and dynamic research on the spatial pattern of forest resources and forest landscape ecology as a whole. Its research results are of great significance for the forest resource management and the governance of the Shandong section of the Yellow River.
(2) This paper quantitatively estimates the stock volume and canopy closure of forest resources in the basin based on TM/ETM+ remote sensing images and related factors. It uses the AHP-based fuzzy comprehensive evaluation method to conduct a comprehensive evaluation of the spatial pattern of forest resources from the aspects of forest resources, forest landscape ecology, environment and social economy, and its research method is innovative.
(3) This paper optimizes the allocation of afforestation-worthy barren hills and wastelands and forest species structure in the Shandong section of the Yellow River Basin based on linear programming method. Without changing the forest coverage rate, it can greatly improve the spatial pattern of forest resources and enhance the ecological, economic and social benefits of forest resources. This research method has theoretical significance and practical value.

搜集汇总

背景与挑战

背景概述

该数据集聚焦于山东黄河流域森林资源，利用2000年和2006年遥感影像及地面调查数据，通过RS、GIS技术和数学模型，分析了森林覆盖率、蓄积量变化及景观格局。研究发现森林覆盖率从21.4%增至25.6%，但景观破碎化严重，空间分布不均，并通过层次分析和线性规划方法提出了优化建议，旨在改善森林生态效益。