Study On The Early Warning And Regulation Of Cultivated Land System Security In Heilongjiang Province

The early warning of cultivated land system security(CLSS)is an effective way to scientifically grasp the status of the CLSS.It can not only grasp advantages of the operation of the CLSS,providing supports for ensuring the security and sustainability of the cultivated land system,but also find out the possible problems and causes of the future operation of the CLSS in advance,providing the basis for implementing measures of prevention risks and mitigating against future threats of the CLSS.It is an important foundation for ensuring food security,promoting socio-economic development and ecological protection.Heilongjiang Province has abundant cultivated land,accounting for 11.7%of cultivated land in the whole country.The region is one of China’s most important for food-production and food-commodity supply,playing a vital role in safeguarding national food security.However,despite the abundance of cultivated land and total grain,the quality of cultivated land is worrying.Cultivated land soil pollution,ecological environment destruction,soil erosion and pesticide and fertilizer application and other issues highlighted in some areas.Besides,the cultivated land area of medium-level and low-level production reached 15.8049 million hectares in Heilongjiang province in 2015,accounting for 99.69%of cultivated land in the whole province.Therefore,accurately identifying the status of the CLSS and timely making scientific early warning.can not only maintain the advantages of the region,guarantee the sustainable development of the secure area of the cultivated land system,but also accurately identify the potential risks in problem-areas of the cultivated land system,and targeted put forward regulation ways and measures of the CLSS,which is of utmost importance to ensure the CLSS of the whole province and realize the common sustainability of food security,economic development,and ecological protection.Current researches on the early warning and regulation of the CLSS have mostly focused on a single spatial scale,one aspect of the cultivated land resources quantity,quality and ecology,or simple combination of several aspects,which lack systematic and multi-scale early warning research.The accuracy of assessment and prediction results of the early warning of current researches are not high,which is impossible to accurately grasp the status of the regional CLSS.Meanwhile,the research on the fundamental driving factors of the early warning of the CLSS is scarce.Most of studies on the CLSS regulation are macro countermeasures and proposals,which are lack of quantitative,targeted and systematic regulation solutions.In view of this,this paper takes Heilongjiang province,the main grain producing area in Northeast China,as the research area.Based on the natural-ecological data,socioeconomic data and cultivated land use data,introducing the biological immunity mechanism,an early warning index system of the CLSS at different scales in provinces,cities,and counties from the system perspective was established.Using the subjective and objective weighting method and numerical transformation method to construct improved catastrophe model,and introducing the Elman neural network model,to assess and predict the early warning results of the CLSS.Based on the results of the early-warning status,the characteristics of spatiotemporal evolution of the early warning on different scales were analyzed,and key driving factors were determined by weighting analysis,constructing reasonable panel data models and geographically weighted regression models,then the spatio-temporal evolution mechanism of the early warning eventually was clarified.Finally,by regulating the direct influencing factors and fundamental driving factors of the early warning,quantitative regional simulation and cultivated land pattern optimization were respectively applied to regulate the CLSS at the scale of cities and counties(combined with patch scales),and the qualitative regulations of the CLSS at provincial scale are combined,all of which were used to regulate the CLSS in study area.The research results for ensuring the national food security and realizing the coordinated development of economic development and ecological protection in Heilongjiang province.Research indicates:(1)Introducing the biological immune mechanism,the concepts of cultivated land system and cultivated land system security(CLSS)were tentatively proposed.The feasibility of applying the immune mechanism to the early warning status of CLSS was analyzed,then the framework of the early warning of CLSS was constructed.In view of the different scales of provinces,cities,and counties,the regulation framework was constructed from the way that promoting regulation by macroscopic,policy-based measures,simulation of sub-divisional scenarios,and the optimization of cultivated land patterns.(2)The improved catastrophe model and the Elman neural network model can provide new method for the assessment and the prediction of the early warning of the CLSS.Assessment results based on the improved catastrophe model can better highlight the "good and bad" characteristics of the assessment object,which is more consistent with the objective reality,and the method was proposed for the defects of the traditional catastrophe model and has the universality.Prediction results with small errors based on the Elman neural network are more accurate,and they are reliable.(3)The results of the early warning of the CLSS at different scales in the province,city,and county are studied,which shows that the status of the CLSS with high the early-warning values is not optimistic and needs to be regulated effectively.①At the provincial scale,the early-warning status of the CLSS in Heilongjiang province from the"secure" and "relatively secure" to the "low-level warning",then to "medium-level warning" during 1995-2014.Under the current development of the early-warning status,the early-warning status will remain between "low-level warning" and "medium-level warning" during 2015～2034.②At the municipal scale,only two cities are relatively secure(critical secure),three are severe-level warning,and the other seven cities are low-level warning and medium-level warning during 1995～2014.Under the current development of the early-warning status,there is a period when the early-warning values will be reduced in 12 cities,but eventually it will be an increasing trend of the early-warning value,that is,the early-warning level will increase in various cities.③At the county scale,of the 72 counties,69 counties were in the early-warning status,about 94%of county total,and 28,32 and 9 counties were low-level warning,medium-level warning and severe-level warning,respectively.(4)The spatio-temporal characteristics of the early-warning status exhibited significant differences at different scales.①At the provincial scale,the historical changes of the early-warning state showed intense fluctuation during 1995～2014,and the early-warning status of the province will develop in the future along the same lines as it did so historically.②At the municipal scale,the early-warning values presented certain fluctuation characteristics during1995～2014.and with the exception of Yichun and Suihua,the early-warning status of the province will develop in the future along the same lines as it did so historically.That is.the early-warning status will initially rise and then decrease,and finally slightly rise during 2015～2034.The severe level warning areas are mainly found in Shuangyashan city and Jixi city located in the southeast,and Daqing city and Qiqihar city located in the west of the region,while the secure and relatively secure areas are mainly concentrated in Heihe city.Yichun city,and Mudanjiang city located in the middle-north and southern areas.③At the county scale,the severe-level warning areas were mainly found in the southeast and west of the region,while the secure and relatively secure areas were mainly concentrated in the north-south axis belt areas in 2014.The counties with positive spatial autocorrelation(high-high type and low-low type)emerged with the "clusters" and had a powerful agglomeration.The counties with negative spatial autocorrelation(high-low type and low-high type)were few and showed sporadic distribution.(5)Constructing the panel data model scientifically and effectively realized the identification of the key driving factors for the spatiotemporal changes of the early-warning status of the CLSS at the municipal scale.①The driver factors including natural population growth rate,natural disaster index,agricultural waste per unit area,waste water load per unit area of cultivated land,and forest coverage rate are the key driver factors of the spatiotemporal changes of the early-warning status in the municipal scale during 1995～2014.One factor(forest cover)is significantly negatively correlated with the early-warning status;the other four factors are significantly positively correlated.②Rational panel data model construction should be verified by unit root test,co-integration test,model selection,instrumental variable test,and endogenous test,so as not to produce pseudo-regression.(6)Constructing the geographically weighted regression model achieved the spatialization of the key driving factors effects on the early-warning status of the CLSS at county scale.①9 driving factors,including 3 natural-ecological factors(slope,elevation and annual average temperature)and 6 socioeconomic factors(per capita GDP(gross domestic product),urbanization level,investment in agricultural fixed assets,water-soil coordination degree,agricultural wastes index,and road network density)are the key driving factors for the spatial differentiation of early-warning status,and the spatial effects differences of these key driving factors are significant.The effects on the early warning of elevation and slope shows a decreasing rule from northwest to southeast of the study area;the average annual temperature has a high effect on the west and the south,and the eastern and northern parts have a low effect;the effect of water-soil coordination degree presents a gradual decrease from the east to the west;the effects of the per capita GDP,urbanization level,investment in agricultural fixed assets,agricultural wastes index,and road network density have gradually decreased from northwest to southeast.②The spatially visualized of effects on the early-warning status of key driving factors based on the geographically weighted regression model,considering the spatial heterogeneity of each county,which can accurately characterize the spatial effects on the early-warning status of various factors,and can provide new ideas and perspectives for the cultivated land protection.(7)The agricultural waste index per cultivated land is a common and key driving factor for the spatiotemporal changes of the early warning at the provincial,municipal,and county scales.It needs to be changed to increase the amount of agricultural wastes to increase the vicious growth pattern of grain production.The driving mechanism of early warning of the CLSS was also clarified that changes of natural circumstances and human economic development lead to the management way cultivation of cultivated land change,then lead to cultivated land benefits change,ultimately affect the CLSS,which provides more abundant and effective information for the CLSS regulation and scientific land use planning.(8)At the provincial level,four macro-policy countermeasures have been proposed:reducing the "hidden risks" of the cultivated land system,strengthening the "immunity"of the cultivated land system,alleviating "pressure" of the cultivated land system,and improving "responsiveness" of the cultivated land system.At the municipal level,the optimal analysis scenarios for the three areas(Songnen Plain,Sanjiang Plain,and hilly areas)are set separately.The status of the CLSS will be improved in next 20 years(2015～2035)by regulation of direct influencing factors of the CLSS.At the county level(combined with patch scale),through regulating fundamental factors of the early warning status of CLSS and introducing the FLUS model,it has been determined that the cultivated land pattern under the comprehensive regulation of food security,ecological protection and economic development is most conducive to mitigate against future threats of cultivated land system from the perspective of optimizing the cultivated land pattern.