Simulation And Optimization Of Forest Landscape Of The Hilly-gully Region On The Loess Plateau

Forest ecosystems play important roles in mitigating climate change.However,forest ecosystems were always affected by climate change and human activities.The traditional method of studying landscape configuration can hardly take into consideration of climate change and human activities;therefore,it was urgent to propose a new approach to optimizing the forest landscape.In this study,a typical forest ecosystem in Yongshou County on the Loess Plateau was selected as an example,because it is an ecologically sensitive area in China.Based on remote sensing and process models,we analyzed the current status of land cover types and vegetation changes in Yongshou County.Then,we simulated responses of forest landscape and forest ecosystem services to the climate change.Finally,we proposed a novel approach to optimize the forest landscape in the study area.Main results were listed as following:The variation of vegetation in Yongshou County showed an increasing trend during 1998 and 2013,and the area where the vegetation was significant improved occupied more than 73%of the study areas.Vegetation variation presented an obvious seasonal trend,with the maximum value occurred in May and the minimum occurred in February.The area with high and extreme sensitivity occupied about 40%of the whole region,which indicated that Youshou County was still a fragile ecosystem.The two main land use types were forest land and cultivated land,which greatly affected the vegetation change of the whole area.Forest land,orchards and construction land showed systematic expansion after comparing the classification of TM images during 1992-2000 and during 2000-2013.Converting from immature forest land to forest land,from cultivated land to orchards,and from orchards to construction land were the three main land cover type changes during the two periods.We proposed a revised mode to detect the land cover dynamic.Based on the model,we found that the northwestern and northern parts of the study area showed relatively low change,while the eastern,middle and southern parts of the study area showed a high change.Human activities directly affected the land cover dynamic in the middle and southern part of the study area.Agriculture,mining and urbanization were the main drivers for land cover dynamic during 1992-2000.The land cover dynamic increased during 2000-2013 mainly in the northwestern and eastern parts of the study area,which was mainly influenced by human activities and land use policies.The processes and landscape changes of the forest ecosystem were simulated with LANDIS-Ⅱ model.At the early stage,Robinia pseudoacacia occupied nearly a half of the forest land in Youshou County,but the areas decreased gradually during the succession process.Although Pinus tabuliformis occupied the largest area,it will be replaced by Platycladus orientalis after 2260.Among the deciduous and broadleaf forests,the areas of Populus davidiana was larger than those of Robinia pseudoacacia,Quercus wutaishansea and Salix matsudana.The results presented the detailed information of forest succession process under climate change and they also showed the different strategies between exotic and indigenous trees to adapt climate change.Forest ecosystem services were evaluated by InVEST based on the vegetation distribution obtained from LANDIS-II.There was a great variation of the carbon storage,which initially increased and then decreased.R.pseudoacacia and P.tabuliformis were the main carbon sources at the early stage.The carbon storage of P.tabuliformis decreased,while the carbon storage of P.orientalis and P.davidiana increased.The distribution of water supply was consistent with precipitation in 2010.The water showed a slow increased trend,and the high values will occur in the southern region during 2010-2070.In the late period,water supply with high values will be located in the southeast,north and southwest of the region.The patterns of soil retention will change little and the high values will be distributed mainly in the deep-slope region.This study proposed a novel approach to optimize forest landscape based on the progress-based model and to provide an optimized scheme to manage forest landscape in future climate change scenarios.The results showed that the forest management could increase the number and types of patches and the degree of landscape heterogeneity and fragmentation.During the simulated 100 years,2040 would be an inflection point,which divides our aims into two parts.The one is a short-term forest management and the other one is a long-term forest management.The most effective management plan for a short-term aim is to clear-cut 5%of the area and plant mixed forest with Q.wutaishanica and P.labuliforms.Differently,the most effective management plan for a long-term aim is to clear-cut 5%of the area and plant mixed forest with P.orientalis.These results will not only help local managers make the forest management strategies and land use policy,but also provide theoretical and technical supports for ecological scientists.This study presents a novel method to optimize forest landscape on the Loess Plateau.