基于CA-Markov模型的石羊河流域生态承载力时空格局预测

生态承载力是区域可持续发展的重要物质基础。为了探讨我国干旱区内陆河流域未来生态承载力的时空格局变化,以石羊河流域为研究区,基于该流域1992、2002年和2012年3期Landsat TM遥感影像,选取10个影响生态承载力变化的主要驱动因素,利用元胞自动机-马尔科夫模型(CA-Markov),以土地利用预测为切入点,对该流域2022年的生态承载力时空格局进行了模拟预测。首先基于1992—2002年数据预测了2012年石羊河流域土地利用状况,并与当年实际土地利用状况进行对比和验证,结果显示:Kappa系数为0.7956,说明本文所选用的预测方法预测结果可靠,可以用于该流域土地利用预测;其次,以2012年为起始年,模拟预测了2022年土地利用空间数据,并计算生态承载力时空格局,结果显示:2022年石羊河流域单位面积生态承载力与1992年、2002年和2012年相比整体空间分布格局变化不大,但区域内部变化却呈明显的空间分异特征;1992—2022年流域生态承载力总量以2002年为拐点呈先减少后增加的趋势,其中建设用地的生态承载力增加最为显著;预测可知,较2012年,2022年流域上游山区的林地、中游绿洲的建筑用地生态承载力均增加较快,且呈现较明显的斑块聚集,而中、下游绿洲区耕地、草地生态承载力增、减变化复杂,斑块分布较为离散和破碎;2012—2022年土地利用类型将发生频繁转换,导致流域生态承载力结构组成变化较大,其中未利用地的转出对2022年流域生态承载力的增加贡献突出。研究结果表明自2002年以后,该流域实施的退耕还林还草、关井压田等生态工程已经并将继续对该流域生态效益的提高起到积极的作用。

Spatio-temporal pattern prediction of the biocapacity in the Shiyang River Basin on the basis of the CA-Markov model

The biocapacity is an important material foundation of regional sustainable development. In order to explore the spatio-temporal pattern of biocapacity in inland rivers, arid China in the future, for a case study of the Shiyang River Basin, using the CA-Markov model as a predication method and choosing 10 main driving factors influencing the biocapacity, we predicted the land use and assessed biocapacity in 2022 based on Landsat^TM images in 1992, 2002, and 2012. First of all, by comparing the predicted land use with the actual land use situation in 2012, we got the Kappa coefficient of 0.7956, which showed that the driving factors and the CA-Markov model were reliable for predicting the land use of the basin. Second, based on the model above, we predicted the land use condition and assessed biocapacity in 2022 with the year 2012 as a starting point. The results showed the following:in the Shiyang River Basin as a whole, the distribution pattern of biocapacity per unit area in 2022 will be similar to the situation in 1992, 2002, and 2012; however, there was remarkable spatial heterogeneity within the basin. The total biocapacity of the basin in 1992-2022 has been increasing since the turning point in 2002, and the increase of biocapacity in the built-up area was the most significant part of them. In the comparison with 2012, the biocapacity of the forestland in the mountainous area of the upstream reach and the built-up area in the oasis area of the midstream reach in 2022 should increase faster than other areas of the river basin, and there was a significant patch cohesion. Nonetheless, in the oasis areas of midstream and downstream reaches, the changes of farmland and grassland biocapacity should be complicated and the distribution of the land use patch should be discrete and fragmented. The frequent transition of land use types in 2012-2022 should lead to a large structural change of the biocapacity in the river basin, including an important contribution to the increase of biocapacity supported by the transfer-out area of unused land in 2022. Above all, the results provided a more extensive understanding of the spatio-temporal dynamic and spatial pattern of biocapacity in the Shiyang River Basin. Ecological engineering such as the Grain for Green Project will continue to produce obvious ecological benefits in the future.