Evaluating The Cultivated Land Quality And Its Spatial Variability Across The Pearl River Delta

Cultivated land plays an important role in the economic development and social harmony and stability.The quality of cultivated land is a hot issue related to food security and social security.Since the reform and opening up,the economic development brought about urban expansion,transportation and water conservancy and other infrastructure increased and industrial structure adjustment led to a sharp decline in the number of cultivated land,especially urban construction land led to a large number of high-quality farmland loss in towns surrounding around,irreversible effects on the agricultural production.In addition,investment in new technologies and new varieties,to actively promote the development of agriculture at the same time also brings hidden safety problems such as land pollution to cultivated land,to bring a serious threat to social and economic sustainable development and ecological environment.As a result,the quality of cultivated land resources,cultivated land occupation by construction land quantity continuously declining trend,cultivated land quality survey in positive,which is adapt to the social economic development and solve the food security and reliable choice.Scientific evaluation of cultivated land quality can accurately grasp the current situation and spatial pattern of cultivated land in the rapid economic development area,which is of great significance to guide the reasonable utilization and protection of limited cultivated land resources and realize the quantity-quality of cultivated land on comprehensive balance and management.The Pearl River Delta is not only one of the fastest economic growing core urban agglomerations in China,but also an important agricultural production base in Guangdong Province.Based on this,this paper takes the Pearl River Delta as the research area and the cultivated land in the Pearl River Delta as the research object,for the cultivated land quality evaluation techniques and approaches to carry out the research.The main contents are as follows:(1)This paper analyzes the present situation of cultivated land in the Pearl River Delta and the quantitative changes in the past 15 years from 2000 to 2014,including the general trend of cultivated land resources,the change of cultivated land use degree,the dynamic change of cultivated land,the location entropy,the cultivated land space and the landscape pattern.The main conclusion is that the number of cultivated land decreased rapidly in the early period,and then gradually slowed down;the land reclamation rate decreased slowly and the multiple cropping index showed a trend of fluctuation.From the loss of cultivated land,Guangzhou,Foshan,Dongguan and Zhongshan were serious.The change of cultivated land landscape is obvious,and the degree of fragmentation of cultivated land is also increasing.(2)According to the pearl river delta region and county cultivated land quality evaluation on the two dimensions,select the adaptation of the evaluation index system,evaluation unit and evaluation methods.For the Pearl River Delta scale,correlation analysis combined with rough set screening index to build evaluation system,design evaluation method based on GA-BP neural network,select the representative and covers a sample of 4000 land use types,through the experiment of GA-BP neural network model training times was less than that of BP neural network,and the mean square error of the maximum and minimum difference is smaller than the BP network model of 0.1051,and the actual results more close to the cultivated land quality grade,used for cultivated land quality evaluation is more stable and better fitness.The results of cultivated land quality evaluation show that the quality of cultivated land in the Pearl River Delta region is generally better,among which the proportion of the second and third places is the largest,and the quality of cultivated land is basically the same as that of the normal distribution.The overall appears as the central area of good quality,around the characteristics of low quality.County scale of cultivated land quality evaluation,based on MCDM combined with GIS technology to build the evaluation index system.The initial weight of the index is determined by using the analytic hierarchy process(AHP),and the quality of cultivated land is evaluated by multi-factor method.The results show that the quality of cultivated land in Zengcheng area is higher,but the quality of cultivated land is less,and the proportion of middle and low yield farmland is larger.On the whole,the overall quality of cultivated land in Zengcheng area is characterized by the low central and southern regions.(3)In the evaluation of cultivated land quality on the county,it is a simple and effective method to use the expert's experience to determine the evaluation index system and the correlation and weight between the indicators.However,there are inevitably human subjective factors in the model establishment process,reduce the credibility and accuracy of the evaluation.The evaluation results are uncertain and may change with the weight changes.Therefore,in order to investigate the reliability of the evaluation results,the sensitivity of the weights is introduced to analyze the stability of the evaluation results from two aspects: the local uncertainty and the comprehensive sensitivity.The quality of cultivated land in Zengcheng area is the overall stability and local sensitivity.(4)The spatial distribution pattern of cultivated land quality is analyzed from four aspects: natural quality,utilization quality,economic quality and ecological quality,the results show that the quality of various types of spatial distribution pattern are different.The spatial autocorrelation is used to study the spatial distribution of cultivated land quality in the Pearl River Delta,the quality of cultivated land in the Pearl River Delta was globally and locally.In the global autocorrelation analysis,the Moran's I value of the economic quality is the largest,indicating that the spatial positive correlation is the relatively strongest,and the spatial positive correlation of the quality is relatively the weakest.In the county global autocorrelation,the Moran's I value in the same county scale,there is a large internal differences.In the local spatial autocorrelation,there are great differences in the spatial aggregation types of cultivated land quality,and most of the townships show spatial non-salient.Based on the GWR model,the factors influencing the spatial pattern of cultivated land quality are analyzed.The results show that the model has a good effect.From the P value of the significance test,the number of agricultural employees has the greatest difference in the spatial pattern of cultivated land quality,followed by the proportion of secondary and tertiary industries.From the regression coefficient of each factor,it can be seen that the influence of different factors on the spatial pattern of cultivated land is not the same.Under the influence of human factors,the pattern of cultivated land quality is characterized by aggregation and differentiation.