| The karst rocky desertification(KRD)areas in southern China have prominent human-land conflicts,severe vegetation degradation,fragile ecosystems,and slow socio-economic development.At the same time,rocky desertification and poverty are intertwined.There is an urgent need for catch-up with latecomers to integrate economic development into rocky desertification ecological management.It is imperative for the development of rocky desertification areas to cultivate and develop the eco-industry of rocky desertification control and realize the coordination and unity of regional rocky desertification control and rural revitalization.Eco-industry construction is an important part of the comprehensive control project of KRD.The selection and model construction of eco-industry,the big data mining and analysis of eco-industry,the construction of intelligent monitoring model of eco-industry and the implementation of its decision support system(DSS)are of great significance for testing and consolidating the achievements of rocky desertification control,improving the regional ecological environment,promoting the sustainable economic and social development of rocky desertification areas and promoting the application of management experience.In view of this,according to the comprehensive and regional characteristics of geography,this study combines ecological theory,ecological engineering principles,big data mining and DSS and other theoretical technologies.From 2015to 2021,in the mountainous Guizhou Plateau,which represents the overall framework of karst environment in southern China,select Guanling-Zhenfeng Huajiang Karst Plateau Canyon moderate-intensity rock desertification comprehensive control study area,Bijie Salaxi Karst Plateau mountain potential-light rock desertification comprehensive control study area,Shibing Karst mountain Canyon non-potential rock desertification comprehensive control study area as the study area.Through twelve phases of satellite remote sensing of the three study area data,six aerial photography data,twelve ground field surveys,and two socioeconomic surveys involving village groups in the study area(2015 and2020),characteristic forestry,grassland animal husbandry,agroforestry and tourism of World heritage sites were studied.The paper excavates the valuable information of space-air-ground integration of eco-industry in rocky desertification control area,expounds the appropriate development mechanism of different environmental and ecological industries of rocky desertification control,and constructs the intelligent monitoring model of characteristic forestry,grassland animal husbandry,agroforestry and tourism of world heritage sites for rocky desertification control.On this basis,the economic,ecological,social and comprehensive benefits of characteristic forestry,grassland animal husbandry,agroforestry and tourism of world heritage sites in the control of rocky desertification were evaluated.A spatial decision support system for eco-industry in the control of rocky desertification was developed through spatial database,attribute database,knowledge base and model base.Moreover,the decision popularization and demonstration of the suitability of nine governance models in karst areas of southern China have been realized.(1)Based on TF-IDF,random forest and other deep learning algorithms,this paper deeply digs and analyzes the space-air-ground integration data of the natural ecological environment of the study area and the cultivation and development of the eco-industry of rocky desertification control since the rocky desertification control,obtains valuable data and modeling information of the eco-industry of rocky desertification control,and extracts the evaluation index information of the eco-industry of rocky desertification control.It provides data support for eco-industry benefit comprehensive evaluation model construction and decision support analysis.Based on the deep learning algorithm,the landuse,stony desertification,vegetation coverage,soil erosion modulus,net primary productivity of vegetation and biodiversity of the study area in 2015 and 2021 were excavated.And 15 kinds of evaluation factor information,such as industrial structure change,eco-industry output value,infrastructure coverage,minimum living security standard of rural residents,cultural structure and level of labor force,population density,per capita GDP,Engel coefficient and poverty rate.(2)The comprehensive benefit index system and evaluation model of different eco-industries in the rocky desertification control are constructed,and the evaluation comprehensive index of"two-mountain benefits","evitalization benefits","people-center benefits"and"comprehensive benefits"of ecological benefits,economic benefits,social benefits and their combination are put forward to evaluate the development effect of ecological industries in the study area.The overall development benefit of the characteristic forestry in the control of rocky desertification in study areas was obviously improved,but the development level of the characteristic forestry was different in different degrees of rocky desertification.According to the evaluation grade classification criteria:the evaluation index 0-0.2 is poor,0.2-0.4 is quite poor,0.4-0.6 is medium,0.6-0.8 is quite good,and 0.8-1.0 is good.From 2015 to 2021,the characteristic forestry two-mountain benefits of the study area with medium-intensity,potential-mild and non-potential rocky desertification increased from 0.4894,0.2628 and 0.2054 to 0.5641,0.4087 and 0.3202,respectively.The revitalization benefits increased from 0.1574,0.1245,0.1705 to 0.4313,0.3699,0.4619,and the people-center benefits increased from 0.3784,0.2065,0.1917 to 0.5922,0.5018,0.4685,among which the revitalization benefits increased most obviously.The people-center benefit is followed by the two-mountain benefit.The evaluation results show that:with the support of poverty alleviation and rural revitalization policies,the characteristic forest industry has been cultivated and supported and developed.Remarkable achievements have been made in the construction of the characteristic forest industry in rocky desertification areas,and the governance effects of different rocky desertification levels in the study areas have been improved qualitatively in ecological,economic and social aspects.The overall development benefit of grassland animal husbandry in study areas has been improved.From 2015 to 2021,the comprehensive benefit of the medium-intensity and potential-light rocky desertification control area has increased by about 40%,while that of the non-potential rocky desertification area has decreased by 1%.The revitalization benefit of the potential mild rocky desertification area has increased the most,accounting for 64.78%.The people-center in the potential-mild rocky desertification area increased the least,which was 24.26%.The ecological and social benefits of the area with non-potential rocky desertification in different periods are better than those of the area with medium-intensity and potential-light rocky desertification.The overall benefits of the agroforestry in study areas were relatively obvious,but there were differences benefits in the development level of different rocky desertification.From 2015 to 2021,the comprehensive benefits,the two-mountain benefits,the revitalization benefits and the people-center benefits of the moderate-intensity rocky desertification research zone were 0.3095 and 0.5067,0.2704 and0.3769,0.1644 and 0.3336,0.1842 and 0.3029.The comprehensive benefits,the two-mountain benefits,the revitalization benefits and people-center benefits were0.2639 and 0.5623,0.1955 and 0.4599,0.2045,0.2016 and 0.4602 respectively in potential-mild rocky desertification research zone,the non--potential rocky desertification research area were 0.4146 and 0.6222,0.2964 and 0.3695,0.2793and 0.4353,0.2535 and 0.4396.Among them,the comprehensive benefits of the potential-mild rocky desertification areas,the two-mountain benefits,and the people-center benefits were the largest,with 113.07%,135.24%,and 128.27%respectively.The comprehensive benefits of the non-potential desertification research zone,the two-mountain benefits,and the revitalization benefits were the least,with 50.07%,24.66%,and 55.85%respectively.The results of benefits evaluation show that while the agroforestry played the water preservation and soil in desertification,it playes a good role in promoting the ecological,economic,and social benefits of the rocky desertification area.For potential-mild rocky desertification district governance construction is more obvious.Shi Bing Yuntai Mountain was selected into the Karst World Natural Heritage Land in 2014.From 2015 to 2021,it evaluated its tourism industry in three aspects:ecological,economic and social benefits.the comprehensive benefits of the World Heritage Site tourism industry increased from 0.1742 to 0.3111,an increase of78.59%.The two-mountain benefits have increased from 0.1268 to 0.2148,the revitalization benefits have increased from 0.099 to 0.2287,and the people-center benefits have increased from 0.1217 to 0.1787.The growth rate of revitalization benefits was 128.93%,and the smallest is people-center benefits,with an increase of 46.84%.The low comprehensive benefits and low benefits of each combination indicate that the base number of tourism in the world’s heritage land is poor,but the space for development and improvement is large.(3)Vector databases such as the basic geographic information of the study area,Guizhou Province and southern China,the rocky desertification control project and the spatial layout of ecological industry in the study area,raster databases such as satellite remote sensing image,UAV positive image and digital elevation model were established,and the socio-economic attribute database of the study area was built on the comprehensive evaluation model.The ecological industry knowledge base for rocky desertification control was established with the support of knowledge graph theory,and the database capacity reached 10TB.In the China Geodetic Coordinate System 2000 and the Yellow Sea Elevation System 1985,a basic geographic vector database including the study area,Guizhou Province and the south China karst region was established based on Arc SDE,and the status of land use,rocky desertification,vegetation coverage,soil erosion modulus,vegetation net primary productivity,biodiversity,ecological restoration projects,water conservation projects,ecological industry layout and other spatial databases;The socio-economic database includes the name of administrative village,group name,total number of households,total population,labor force,annual household income,plantation income,forestry income,animal husbandry income,labor service income,income from returning farmland to forest,total GDP,crop planting,land area,etc.The raster database includes 1:1 million topographic maps corrected in the study area,raster data such as lithology,slope,DEM and elevation,satellite remote sensing images and UAV orthophoto images.Based on the evaluation monitoring model,the ecological industry model base based on the weight score is constructed.Based on the scientific research results of experts in various fields,the production experience of scientific research and technical personnel,and professional works/books/papers and other knowledge,the ecological industry knowledge base is generated and stored in the distributed knowledge base system after review,processing,classification and organization.(4)The eco-industry spatial DSS for KRD control was developed,and nine models,including"Huajiang model","Bijie model"and"Shibing model",were found to be the most suitable,more suitable,basical suitable,barely suitable and unsuitable regions in karst areas of South China.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of the"Huajiang model"of characteristic forestry in karst areas of South China are 9.85×10~4 km~2,30.11×10~4 km~2,48.55×10~4 km~2,57.19×10~4 km~2 and49.67×10~4 km~2respectively.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of Huajiang Model of grassland animal husbandry in karst areas of South China are 26.61×10~4 km~2,44.5×10~4 km~2,55.38×10~4 km~2,37.60×10~4 km~2 and 31.28×10~4 km~2 respectively.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of the"Huajiang model"for agroforestry promotion in karst areas of South China are38.91×10~4 km~2,14.8×10~4 km~2,11.49×10~4 km~2,10.88×10~4 km~2 and 97.96×10~4 km~2respectively.Among them,the most suitable area accounts for 5.04%,13.62%and19.92%of the 8 southern provinces(municipalities and districts)respectively.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of the"Bijie model"of characteristic forestry in karst areas of South China are 20.01×10~4 km~2、38.06×10~4 km~2、41.60×10~4 km~2、50.75×10~4 km~2and 44.95×10~4 km~2respectively.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of"Bijie Model"for grassland animal husbandry in karst areas of South China are 20.45×10~4 km~2、44.25×10~4km~2、49.55×10~4 km~2、47.06×10~4 km~2 and 34.06×10~4 km~2respectively.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of the"Bijie model"for agroforestry promotion in karst areas of South China are73.34×10~4 km~2、16.5×10~4 km~2、32.92×10~4 km~2、13.51×10~4km~2 and 59.10×10~4 km~2respectively.Among them,the most suitable area accounts for 10.24%、10.47%and 37.54%of the 8 southern provinces(municipalities and districts)respectively.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of the"Shibing model"of characteristic forestry in karst areas of south China are 14.67×10~4 km~2、36.76×10~4 km~2、46.62×10~4 km~2、57.03×10~4 km~2and 40.29×10~4 km~2respectively.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of"Shibing Model"for grassland animal husbandry in karst areas of south China are 19.58×10~4 km~2、46.83×10~4 km~2、50.50×10~4 km~2、44.99×10~4 km~2 and 33.47×10~4 km~2respectively.The most suitable,more suitable,basically suitable,barely suitable and unsuitable areas of the"Shibing model"for agroforestry promotion in karst areas of south China are73.75×10~4 km~2、25.28×10~4 km~2、38.20×10~4 km~2、49.15×10~4km~2and 8.99×10~4 km~2respectively.Among them,the most suitable area accounts for 7.51%、10.02%and37.75%of the 8 southern provinces(municipalities and districts)respectively. |
