Maize Productivity Regionalization In Jilin Province Based On Mutiple Source Data

Along with the promotion and development of spatial information acquisition and data processing, computation technology and statistical theory, how to transform these space-based observation data, from various fields such as global change, natural resources, ecology, environment and humanity, etc. into knowledge or rule for regionalization, combine the classical theory and methodology of regionalization with the computation science to develop the integrated methodology of regionalization, have already become a kind of inexorable trend and current demand.Selected Jilin Province maize as a case study, this paper conducted the study on crop productivity regionalization, by integrating with the GIS technique, multiple-source based data acquisition, geostatistics theory, mathematical statistics model, crop potential productivity model and traditional regionalization theory and method. Compared with the related literature both at home and abroad, this paper mainly innovated in two respects:⑴The paper integrated the spatial information technology, geostatistical theory, statistical method, crop potential productivity model and traditional regionalization theory and method based on quantitative analysis method innovatively;⑵The paper explored and advanced the spatialization technology of the elements in regionalization based on multiple source data, the precision of the elements in regionalization have been improved in aspects of data acquisition and spatialization. The main conclusions of this paper are briefly described as below:⑴Indicator system of regionalization of maize productivityAs the first step toward our final goal, we first considered to develop an indicator system of maize regionalization. According to basic theories, purpose and principle of crop productivity regionalization, and the background of out quantitative knowledge of the environmental factors of maize productivity, the paper established an indicator system of regionalization of maize productivity, which included geographical environmental factors, crop potential productivity, current crop productivity, natural hazards index, utilization and the management index and crop type index, etc.⑵Data acquisition and spatialization of the regionalization element①meteorological conditionsThrough the correlation analysis among the environmental variables, such as meteorological variables, longitude, latitude, elevation, slope, aspect and hillshade, etc, the paper try to characterize the spatial distribution of the meteorological element. By contrast of the precision and validation of the result from three interpolation methods, including kriging, cokriging and multivariate regression with residues kriging, the paper chose the spatial interpolation method with highest precision to map spatial distribution of the meteorological element with a spatial resolution of 1km.②soil nutrientsUsing GIS and geostatistics technique, the paper quantified the spatial variability characteristics of soil pH,soil organic matter, available K, available P and available NH4 in the soil in study area,at the same time , the correlation among these main soil nutrients and other soil properties were analyzed, such as available boron(B) ,available calcium(Ca), available magnesium(Mg), available copper(Cu), available zinc (Zn), etc, and the correlation with the geographical elements was also calculated, such as longitude, latitude, elevation, slope, aspect, plane curvature and profile curvature, etc. Then from the contrast of the results derived from the method of kriging and the cokriging applied to the soil type and microelements, the paper determined the cokriging as spatial interpolation method for the spatial distribution of five types of soil nutrient elements with each grid being 1km by 1km.③maize productivityAccording to the previous agricultural regionalization of Jilin Province, the paper first construct a framework of regionalization with the districtⅠ(Jilin province), districtⅡ(middle plain farming region and western plain farming and pastoral region), districtⅢ(agricultural forest zone of Mid-levels in the east and forest agricultural zone of Changbai Mountain). Then on a GIS platform, the multivariate regression model was developed to simulate the relationship between the statistics-based maize yield and the geographical environmental elements, while the correlation analysis were conducted to validate the result of modeling. Finally, the spatial distribution of maize yield was mapped with a spatial resolution of 1km .④Maize potential productivityUsing spatial analysis technique embedded in GIS platform, combined the variables of meteorological conditions and soil nutrient status, we developed a crop potential productivity model by an combination of the traditional crop productivity attenuating model and the GRID model in GIS, and to quantify the spatial distribution of the maize potential productivity.⑶Maize productivity regionalizationOn the understanding basis of the purpose, principle, the above indicator system and method of crop productivity regionalization, fully considering the mechanism of crop growth and ecological theory on material movement and energy flow, spatial heterogeneity of physical geographic processes, condition and status of social economic development, the advancement of current natural, topographical, agricultural and climatic regionalization, the paper developed the methodology for maize productivity regionalization using GIS technique, experience criterion, in conjunction with the general rules of aggregation (from bottom to top with the partition and from top to bottom). According to the steps mentioned above, six different regions were divided in Jilin Province: DistrictⅠ(highest productivity of the middle and eastern region of Jilin Province), districtⅡ(Second highest productivity of middle and east region of Jilin Province), districtⅢ(moderate productivity of west and middle region of Jilin Province), districtⅣ(low-moderate productivity of west and middle region of Jilin Province), districtⅤ(lower productivity of agricultural forest zone of Mid-levels in the east and forest agricultural zone of Changbai Mountain in Jilin Province), districtⅥ(lowest productivity of forest agricultural zone of Changbai Mountain in Jilin Province). The higher relative potential productivity distributed mainly in the regions of districtⅠand districtⅡ,and the lowest relative potential productivity distributed in the other areas extensively; The spatial or geographic property of the coefficient of potentiality was scattered over the Jilin province. As a general conclusion, the corresponding appropriate countermeasure of agricultural development should be adopted according to the characteristic of different region, and some scientific-based support should be provided for maintaining and promotion the sustainable development of regional economic and ecological system.