摘要
198 7年大兴安岭“5 .6”特大森林火灾 ,过火面积 1.33× 10 6 hm2 。火后 ,不同的火烧强度区和岛状的未火烧区形成了一个异质镶嵌体 ,然而在自然演替和人为干扰作用下 ,十几年过去之后这种异质镶嵌体格局已经发生了明显变化。以大兴安岭北坡 ,图强林业区的育英和奋斗两个林场为研究区 ,总面积 1.2× 10 5hm2 ,在景观尺度上 ,以遥感和地理信息系统为技术手段 ,将地形因子进行梯度划分 ,并运用分布指数和偏相关分析探讨了大兴安岭在 1987年大火后森林景观在地形梯度上的变化情况。研究结果表明 :森林景观在地形梯度上的格局发生了明显的变化 ,但是不同的景观斑块类型变化幅度不同 ,不同的地形因子对不同森林景观斑块类型变化的影响强度也迥异 ;偏相关分析表明地形对主要森林景观斑块类型之间的变化影响最大的是坡度 ,其次是海拔 ,坡向的影响不明显。
On May 6, 1987, a catastrophic forest fire took place on the northern slope of Daxing'anling Mountains, which burned out more than 1.33×10~6 hm^2 of natural forest and produced a strikingly heterogeneous mosaic of burned severities and islands of unburned vegetation across the landscape. Due to natural succession and anthropogenic disturbance, after more than ten years, the spatial pattern of the landscape has changed greatly. In this paper, a typical area was selected for a case study, which belongs to the Tuqiang forest bureau, including two forestry farms, namely, Yuying and Fendou, covering more than 1.2×10~5 hm^2. The main object of this paper is to analyze the influence of topographical factors on the post-fire forest landscape. With the support of the Geographic Information System (GIS) and Remote Sensing (RS) software, 12 land cover types were delineated for the study area: coniferous forest, broad-leaved forest, mixed forest, shrub, nursery, harvested area, burned blanks, agricultural land, swamp, water, built-up, grass. Three terrain factors were analyzed: elevation, slope and aspect. Elevation is classified into eleven types: the area below 450 m is assigned 1, that of 450m~500m is assigned 2, and so on. Slope is classified into six types: the area where the slope is less than 5° is assigned 1, that of 5°~10° is assigned 2, and so on. Aspect is classified into five types according to the amount of sunlight gained: Shaded aspect (North (337.5°~22.5°) and Northeast (22.5°~67.5°)), semi-shaded (Northwest (292.5°~337.5°) and East (67.5°~112.5°)), semi-sunshine aspect (West (247.5°~292.5°) and Southeast (112.5°~157.5°)), sunshine aspect (Southwest (202.5°~247.5°) and South (157.5°~202.5°)), and no-aspect. Slope, altitude and aspect were derived from digital topographical maps using the ARC/INFO and ARCVIEW software. The spatial distribution of forest landscape on the terrain gradient was analyzed using Distribution Index (DI) and the Partial Correlation Analysis, using the SPSS software. The results obtained from the present study are as follows: (1) The spatial pattern and structure of forest vegetation in the study area changed greatly, but the change of each vegetation type was different. Especially, the area of coniferous forest, broad-leaved forest and mixed forest increased remarkably, and the cross-conversion among them was very strong. The burned blanks and harvested area became forest vegetation completely. (2) The change of spatial distribution of different forest landscapes along the terrain gradient was quite different. The Distribution Index (DI) curves of coniferous forest and broad-leaved forest were changed more greatly than that of mixed forest and marsh. The change of spatial distribution of the same forest landscape on different topographic factors (elevation, slope and aspect) was also quite different. The DI curves of coniferous forest and broad-leaved forest on elevation gradient changed more greatly than that on slope and aspect gradient. The curves of mixed forest on slope gradient changed more greatly than that on elevation and aspect gradient. The curves of marsh on aspect and elevation gradient changed more greatly than that on slope gradient. (3) Through Partial Correlation analysis, it is concluded that the effect of different topographic factors on forest succession was quite different. The effect of slope was the strongest, elevation was the second and aspect was the last.
出处
《生态学报》
CAS
CSCD
北大核心
2004年第9期1863-1870,i002,共9页
Acta Ecologica Sinica
基金
国家自然科学基金资助项目 ( 3 0 2 70 2 2 5
40 3 3 10 0 8)
中国科学院知识创新资助项目 ( SCXZY0 10 2)~~
关键词
林火迹地
森林景观
地形梯度
分布指数
偏相关分析
大兴安岭
burned blank
landscape ecology
terrain gradient
distribution index
partial correlation analysis
Mt. Daxing'anling