| Scientific management of soil nutrients and rational use of fertilizer are strategic issues affecting sustainability of Chinese agriculture. Soil properties not only have spatial variability, but also change with the time. However, traditional agricultural management has being based on field unit. The same quantities of agricultural materials were applied to the whole field at the certain zone. It is common that supplement of soil nutrients was imbalanced and that utilization ratio of fertilizer was low under the condition of the agricultural model. In response, it is important to fully understand the spatio-temporal variability of soil nutrient, define proper managerial units, and meet the wants of fertilization.The aim of precision agriculture is to match resource application and agronomic practices with soil and crop requirements as they vary in space and time within every management zones of farmland. In present, much of the research of precise management zones has concentrated on small scale farmland. Agriculture of Xinjiang BINGTUAN has the characteristics of wide farmland, high level mechanization, unionized management, similar weather in zone of every corps, and different spatial distribution of soil nutrition. The study on precise management zone at the 125th Corps not only provides conference model suitable to local precision agriculture , but also provides basic information variable fertilization and soil precise management on large scale fartmland.In this study, conventional statistics, Geostatistics, the spatial analysis technique of geographic information system were used to analyze spatio-temporal variability of soil properties in the study area, based on which, to design the classify precise management zones. Some satisfied results were obtained as follows.1 GIS and Geostatistics were used for the analysis of the spatio-temporal variability of organic matter, total nitrogen, available phosphorus and available potassium estimated in topsoil samples. These samples were collected from Oasis cotton field in northern Xinjiang in 1996 and 2005. The contents of soil organic matter increased from 1996 to 2005, whereas, that of total nitrogen, available phosphorus and available potassium decreased. After long different land management, the content of soil nutrients has become more uniform than before. The CV value for soil nutrients was all decreased during the past 10 years. The spatial autocorrelation of soil organic matter and available phosphorus decreased from 1996 to 2005. However, that for total nitrogen and available potassium increased. The range of spatially dependent of soil organic matter increased from 1996 to 2005, whereas, that of total nitrogen, available phosphorus and available potassium decreased sharply. Temporal stability map of soil nutrients indicated the stable and unstable zone. Soil organic matter had the highest degree of temporal instability. In contrast, total nitrogen, available phosphorus and available potassium were temporally stable. The temporal stability map indicated a decreasing trend from north to south across the C.V. of oil nutrients. Large scale mechanized operation and artificial fertilization were the main reasons of the contents and soil spatial variability of soil nutrient changed. As far as spatio-temporal variability of soil nutrition is concerned, stabilizing nitrogen, strengthening phosphorus supplementing potassium taken into consideration during the course of fertilization. Therefore, it will be possible to develop a regionalized soil nutrients management program for soil nutrients in experimental region.2 GIS and Geostatistics were used for the analysis of the spatial variability of organic matter, total nitrogen, available phosphorus and available potassium estimated in topsoil samples with three scales. These samples were collected from oasis cotton field in northern Xinjiang Province. The C.V. of different soil nutrients ranged from 0.14-0.52 at three scales, showing a medium spatial variability. When on large scale, soil spatial variability is high and when on medium or small scale, the variability is low. Results indicated the spatial autocorrelation of soil OM, TN and AP was medium with three scales, and AK have strong spatial autocorrelation on large or medium scale, but it showed pure nugget effect with weak spatial dependence on small scale. The range of spatially dependent is high on large scale, while the range is medium on medium scale, and the range is low on small scale. The soil variability and the spatial autocorrelation increased with the enlarging of study scale and more complexity of soil. Spatial variability of soil nutrient on small scale was heavily dependent on random factors, while spatial variability of soil nutrient on large scale was heavily dependent on structural random factors. On large scale, the distribution maps of soil nutrients would ignore the spatial variability on small scale, but on medium scale, the distribution maps of soil nutrients could indicate the spatial distribution on both large and small scale clearly. It was an appropriate way that soil nutrients in experimental region might be managed at medium scale based on the clearly defined management zones.3 Based on analyzing spatio-temporal variability of soil properties in the study area, Fuzzy c-means clustering algorithm was used to delineate management zones. In order to determine the optimum fuzzy control parameters at the 125th Corps. Organic matter, total nitrogen, available phosphorus, and salinity were collected from cotton field in northern Xinjiang Province. In order to determine the optimum fuzzy control parameters, the fuzziness performance index (FPI), c-φcombinations and the multiple regression based on external variable were used in this study. Meanwhile, the cotton yield was chosen as the external variable. The average confusion index was low in all management zones. Thus, the overlapping of fuzzy classes at points was low and the spatial distribution of membership grades was unambiguous. To estimate the validity of zoning result, the general statistics analysis on the data was carried out. The zoning statistics showed that variation coefficients of soil properties decreased, while the means of the soil properties differed sharply between management zones. These results indicated the classified management zones can carry out variable input and precise management decision-making according to spatial variability of different management units, and to provide basic information and effective means for precise field management.
|
PDF Full Text Request