Xie De-ti

A hybrid approach of support vector machine with particle swarm optimization for water quality prediction

Water quality prediction is an important and widely studied topic since it has significant impact on national or regional ecological and water resources management. Due to water quality indicators series nonlinearity and non-stationary, the accuracy of conventional mostly used methods including regression analysis, ARIMA and neural network has been limited. The use of support vector machine has been shown to be an effective technology to solve classification, prediction problem of nonlinearity and small sample. However, the practicability of SVM is effected due to the difficulty of selecting appropriate SVM parameters. This paper presents a hybrid approach of support vector machine with particle swarm optimization to determine svm free parameters for developing the accuracy of predictions. The approach is applied to predict Heishui river water quality of the Beibei, Chongqing. Traditional ARIAM model and BP neural network are investigated as comparison basis. The experiment results show that the proposed approach can achieve better prediction performance.

An insight on drivers of land use change at regional scale

The studies of driving forces of regional land use change (LUC) are to reveal the real motivation behind the LUC and its interacting mechanism, so as to simulate and predict the process of LUC. Presently, studies rooting from different natural and socio-economic backgrounds and from different scales have deepened the people’s understanding and cognition to driving forces of regional LUC. Biophysical driving forces are relatively stable, and have the cumulating effects. Human driving forces are relatively active, and are main driving forces of short-term regional LUC. Existing regional LUC models can answer the three main problems: which contribution (why), which location (where) and what rate (when). But, regional land use system is defined as the self-organized system, usually affected by the critical value area and sudden change, and controlled by different stages. To reduce the impact of critical threshold and break on land use system, the studies of LUC driving forces will aim at following priority areas: data linkage between remote sensing and no-remote sensing data; underlying driving force identification; driving factor quantification; driving factor scale dependence; and driving process integration simulation.

Effects of conservation tillage on soil organic matter in paddy rice cultivation

Abstract In this study, the effect of conservation tillage on soil organic matter (SOM) in paddy rice cultivation after 10 yr was investigated. Four treatments, disk till-fallow (DTF), disk till-wheat (DTW), conservation till-fallow (CTF) and conservation till-wheat (CTW) were used. The results indicated that the combinative application of no-tillage, ridge culture and wheat cultivation was a sound conservation practice in paddy rice cultivation. It not only significantly increased the concentration of SOM in the topsoil, but also affected optical and pyrolysis characteristics of humic acids (HA) through changing the composition and structure of SOM. At 0–10 cm, the greatest SOM content was in CTW, but declined sharply with depth, while in DTF, DTW and CTF the SOM content was not as high at the surface as in CTW, but did not decline as fast as in CTW. The oxidation stabilization of SOM was generally greater in no-tillage and ridge culture than that in disk till. The HA optical density in CTW at wavelength 665 nm and 465 nm was 0.122 and 0.705, while in DTF was 0.062 and 0.321, respectively. E 4 /E 6 ratio in CTW was higher than that in the other treatments. The enthalpy capacity of the exothermal peak (360–365°C) for the HA DTA curve in no-tillage and ridge culture was lower than that in disk till, while the HA absorption peaks in 1000–1050 cm −1 presented the reverse trend. The oxidation stabilization coefficient of HA in no-tillage and ridge culture was higher than that in disk till, indicating that the polycondensation degree and aromatization of HA were stronger. These findings suggest that it may be possible to manipulate paddy soil through conservational tillage and crop practices, and thereby maintain adequate SOM concentrations, and mitigate soil organic carbon loss from soil to atmosphere.

The Virtual Area of Arable Land Based on Food Consumption in China

The virtual area is the base line value of the per capita arable area to ensure food security. As a reference, it is to understand the key reasons why the governments of the land management must reinforce the control to protect the cultivated land and the populaces must have the awareness of cultivated land security to participate the cultivated land conservation. To calculate the virtual area for per capita cultivated land area based on Chinas food consumption structure, then the yield of food including grain, plant oil, sugar and vegetable, the virtual area per capita of cultivated land and the per capita actual arable area in China from 2011 to 2020 were calculated by the line regression analysis. At last, the ratio between the per capita virtual area of cultivated land and the per capita actual arable area was calculated too. This result showed that cultivated land pressure is less after 2011. The findings of this analysis suggested that cultivated land conservation is critical in China were consistent with the demand for improvement of Chinese consciousness, implement of land policy and development of economy. Because the cultivated land security is the base of food security.

Integrated Indicators-based Gradation of Cultivated Land Quality in Mountainous Region of Southwestern China

The gradation of cultivated land is to assess the suitability of cultivated land for agricultural production in terms of natural and economic properties of land. It can be used to evaluate sustainability of land use and soil management practices. Formal and informal surveys, Delphi and comprehensive index method are adopted to identify appropriate integrated indicators, and measure the gradation of cultivated land quality. The determination of integrated indicators presents three basic features of cultivated land quality: stable plant productivity, social acceptability and maintenance of environmental quality. Pronounced concentration-dissipation law of total function score value occurs in paddy field units, with three ranges, i.e., 79∼68.5, 68.5∼59 and 59∼51. Total function score value in dryland units mainly ranges from 40 to 70, but slight concentration-dissipation law of each unit is still observed, with four ranges, i.e., 79∼68, 68∼51, 51∼37 and 37∼15. Paddy field quality is divided into three gradations, and the scales are 18 220.9 ha, 5410.6 ha and 2890.9 ha. Dryland quality is classified into four gradations, and the scales are 1548.6 ha, 8153.8 ha, 3316.3 ha and 685.2 ha. The total function score value of every gradation unit is conformed to the results of farmers’ assessment. Research results meet with the needs of accurate degree of the gradation of cultivated land quality. Integrated indicators-based gradation of cultivated land assessment accords with the characteristics of land resources and human preference in the mountains of southwestern China. This study will be useful to promote the monitoring of cultivated land quality, and to supply fine ground and knowledge for establishing appropriate cultivated land management practices in Southwestern China.