Research on the Evaluation and Forecast of Spatio-Temporal Variability of Soil Fertility in Black Soil Area of SongLiao Plain

摘要

In order to find out the spatio-temporal variation law of soil fertility in black soil area and the mutual restriction relation between each factor, this article analyses and evaluates the spatio-temporal variation of soil fertility in black soil area and predicts the spatio-temporal developing trend by using the spatio-temporal sequence algorithm and the Kriging interpolation method of the neural networks. Experiments have shown that the spatio-temporal sequence algorithm has fully taken into account the multidimensionality and timeliness of spatial data and been suitable for multi-year comprehensive evaluation of soil fertility change in the same land. Based on the Kriging interpolation method of the neural networks, this article considers fully kinds of environmental factors in the prediction, combines the mechanism of factors acting on the spatial variability of soil nutrients and improves the precision and accuracy of predicting the spatial distribution of soil nutrients in complex environment areas. The result of the study can provide theory and technology support for comprehensive evaluation and forecast of the soil fertility.