Study On Techniques And System For Models And3S Based Digital Rice Farming

Agricultural production system is a complex dynamic system influenced by multi-factors including environments, crop varieties characteristics, farming systems, management technologies and social demand, etc. With the characteristics of strong spatial and temporal variation, regional dispersion and management empirical, those factors are difficult to be quantized and normalized. On the basis of crop-environment-technology, crop growth simulation models could describe quantitatively and simulate dynamically the crop growth and development and yield formation, and crop management knowledge models could generate the precise crop management plan or quantitative cultivation mode. Based on RS, crop growth monitoring models could quickly invert the crop growth status and nutrient conditions. GIS could analyze and manage spatial data. Therefore, constructing a model-and3S-based digital rice farming system using information technologies mentioned above is practically important to promote the rapid development of information agricultural.In the study the investigation was made to evaluate methods in interpolating and rasterizing meteorological data based on the platform of GIS. Spatial and temporal distribution characteristics of agricultural climatic resources had been analyzed based on the rasterized daily climatic data during rice-growing periods in main rice growing regions of China. Furthermore, spatial and temporal distribution characteristics of the rice potential photosynthetic, photo-thermal and climatic productivities, yield increment potential were researched based on the gradually descending model, which is one module of the rice management knowledge model. A model-and3S-based digital rice farming system was built based on the models (the rice growth simulation model, management knowledge model and growth monitoring model, which were built by our laboratory) and integration of the3S technology. Model-and3S-based digital rice farming system realized the digital and precise monitoring, prediction and management in the rice farming system.A comparative study was made to evaluate the methods of inverse distance weighting (IDW), co-Kriging (CK) and thin plate spline (TPS) in interpolating the average meteorological data (including the maximum air temperature, minimum air temperature, sunshine hours and precipitation) of the15th day per month from the1951to2005comprehensive observation data of559meteorological stations in China. The results showed that the RMSEs for the maximum and minimum air temperature in a year interpolated by TPS were the smaller (1.02℃and1.12℃, respectively), and the R2values between the observed and predicted values the highest (0.9916and0.9913, respectively), compared with those interpolated by IDW and CK. In four seasons, the smallest RMSEs for the maximum air temperature and minimum air temperature, interpolated by TPS were observed in autumn(0.83℃) and summer (0.86℃), respectively, and the R2between the observed and predicted values interpolated by TPS were higher than in autumn than in other seasons. The RMSEs for the sunshine hours and precipitation in a year interpolated by TPS were the smallest (0.59h,1.01mm, respectively), and the R2between the observed and predicted values were the highest (0.9118,0.8135, respectively) interpolated by TPS compared with those using CK and IDW. In four seasons, the RMSE for the sunshine hours interpolated by TPS was the smallest (0.49h) in autumn, and the R2between the observed and predicted sunshine hours was the smallest (0.9293) in winter. The RMSE for the precipitation in winter interpolated by TPS was the smallest (0.33mm), while the RMSE for the precipitation in summer interpolated by IDW was the smallest (2.01mm). The R2between the observed and predicted precipitation in winter interpolated by CK was the highest (0.8781). It was suggested that TPS could be the optimal spatial interpolation method in interpolating and raterizing the daily meteorological elements in China.The climate change trend during rice-growing periods in main rice growing regions of China is highly uncertain because of its vast area and marked variation. The daily meteorological data during the years from1961to1970(1960s) and1996to2005(2000s) of333weather stations in main rice growing regions of China were processed to generate the daily meteorological surface data using ANUSPLIN and ARCGIS. The spatial and temporal distribution characteristics of the climate resources for the rice production, including total sunshine hours, total growing degree days above10℃, average difference of diurnal air temperature, total precipitation, total rainfall days and daily precipitation intensity during rice-growing periods between two decades (1960s and2000s) were calculated and analyzed. As compared with1960s, the total sunshine hours in rice-growing periods of2000s decreased by11.93%, and the decreasing rates of total sunshine hours in the Northeast and Southwest China were lower than those in the Middle and South China. From1960s to2000s, the total growing degree days above10℃increased by9.40%, and the increasing rates of total growing degree days above10℃in the Northeast and Southwest China were faster than those in the Middle and South China. However, in part areas of the Middle and South China, the total growing degree days above10℃during rice-growing periods decreased. The average differences of diurnal air temperature during rice-growing periods of2000s decreased by4.86%as compared with1960s, while increased in part areas of the Middle and South China. The decreasing rates of average difference of diurnal air temperature in the Northeast and Southwest China were higher than those in the Middle and South China. Compared with1960s, total precipitation during rice-growing periods in2000s increased by1.59%and daily precipitation intensity increased by3.22%. The spatial distribution of the change rate of daily precipitation intensity was similar to that of total precipitation, the total precipitation and daily precipitation intensity decreased in Ningxia Province and the Northeast China, while increased in the Middle and South China. Compared with1960s, the total rainfall days during rice-growing periods of2000s decreased by1.60%, and the decreasing rates of total rainfall days in the Northeast and Middle China were lower than that in the South China.Further, the annual potential photosynthetic, photo-thermal and climatic productivities were calculated with the gradually descending model, which was one module of rice management knowledge model, and the spatial and temporal distribution characteristics of above three potential productivities between two decades (1960s and2000s) were analyzed. Further, the spatial distribution characteristics of yield increasing potential and percentage from high yield target to potential photo-thermal productivity, yield increasing potential from actual yield of2000s to high yield target were investigated. As compared with1960s, the potential photosynthetic productivity of2000s decreases by5.40%, and the decreasing rates in the Northeast and Southwest China were lower than those in the Middle and South China. From1960s to2000s, the potential photo-thermal productivity increased in the Northeast and Southwest China, while decreased in most part areas of the Middle and South China, with average decrement of2.56%in main rice growing regions of China. Meanwhile, the potential climatic productivity of2000s China decreased in most areas of China as compared with1960s, while increased in part of the Middle and South China, with average decrement of7.44%in main rice growing regions of China. The yield increasing potential from the actual yield (2000s) to the high yield target ranged from2×103kg/ha to 6×103kg/ha in the single-cropping rice growing region, from6×103kg/hato12×103kg/ha in the double-cropping rice growing region. The yield increasing potential from the high yield target to the potential photo-thermal productivity (2000s) was less than10×103kg/ha in most areas of the single-cropping rice growing region, ranged from10×103kg/ha to30×103kg/ha in most areas of the double-cropping rice growing region, and greater than20x103kg/ha in most areas of South China, such as Guangxi, Guangdong and Hainan Province. In addition, differences from the potential climatic to photo-thermal productivities in2000s illustrated that the increasing potential contributed by irrigation were between5×103kg/ha and20×10kg/ha in the single-cropping rice growing region, between20×103kg/ha and40×10kg/ha in the double-cropping rice growing region. Differences from potential climatic to photo-thermal productivities in2000s indicated that full use of rainfall and reasonable irrigation was favorable to explore the rice potential productivity and ensure the rice increasing and high yields. These results were helpful for further exploration of technical approaches to enhancing rice yields, and for quantitative decision support for guiding national rice production and food security in China.Quick and real-time monitoring of crop growth status based on remote sensing can support the decision-making on precision crop management. Based on growth estimating models in wheat and rice established by the authors’group, a RS image-based monitoring system was developed based on the Microsoft.NET framework using GDAL and GDI+as information processing methods and EM algorithm for classifying crop growth indices. This system realized the multiple functions as accessing the RS images with common formats, extracting crop information, inverting growth indices, clustering analysis, generating the thematic map and issuing the information with remote sensing technology. Several functions of the system were tested using the RS images at Fangqiang Farm, Jiangsu Province. The results showed that the system could effectively read general remote sensing images, invert the crop growth indices, classify the crop growth information based on the cluster models, interact with users for generating the thematic map of crop growth status, and issue the RS image information rapidly via internet. The present system has overcome the previous weakness that the ordinary users could not directly participate in the process of RS images analysis, and can help to monitor the crop growth condition and provide decision support for precision crop management at regional scale.Model-and3S-based digital rice farming system was developed based on plentitude combination of rice growth simulation model, rice management knowledge model, rice monitoring model and3S technologies (GPS, GIS, and RS). Guided by the methodology of software component design and the technology of object oriented program, the system was constructed using the Browser/Server mode and based on the4-tier (the data tier, data access tier, business logic tier and presentation tier) architecture under the Microsoft.NET framework. By inputting the parameters of varieties, climate, soil and production management condition, the system provided users the information services of rice farming zone, cultivation strategy design, growth prediction, strategy evaluation, dynamic regulation, growth monitoring, precision management, productivity evaluation, management of disease, insect and weed, intelligent learning, system management and system help. The study expanded the application scale of models and was helpful to promote the informationization and modernization of the rice production management.