[Using geographical detection to analyze responses of vegetation growth to climate change in the Loess Pla-teau, China]. / 基于地理探测的黄土高原植被生长对气候的响应.

In order to explore the responses of different vegetation types to climatic change in the Chinese Loess Plateau (CLP), we analzyed the changes of different vegetation types and their relationships with meteorological factors using trend analysis, Hurst index, and geographical detector model based on normalized difference vegetation index (NDVI). The results showed that NDVI of different vegetation types from 2002 to 2019 was dominated by a growing trend and codirectional moderate persistence. The NDVI of crops in the built-up and adjacent areas decreased significantly. Except for grassland or meadow that was affected by mixed pixels, the spatial variation of NDVI was significant in the growing season (from April to October). The mean NDVI of different vegetation types followed an oder: coniferous forest > broadleaved forest > scrub > meadow > grassland > crop > steppe > desert. The interactions between meteorological factors were synergistic and non-linear enhancement in the CLP. Moreover, the interaction was more prominent under steppe and desert where habitat was fragile. The synergistic effect of precipitation and temperature had a great influence on all vegetation types. Water vapor, relative humidity, sunshine duration, atmospheric pressure, and wind speed had different explanatory powers on NDVI through indirectly affec-ting hydrothermal conditions.

[Spatial and temporal changes of farmland soil acidification and their influencing factors in different regions of Guangdong Province, Chin]. / å¹¿ä¸œçœä¸åŒåŒºåŸŸå†œç”°åœŸå£¤é ¸åŒ–æ—¶ç©ºå˜åŒ–åŠå ¶å½±å“å› ç´ .

Based on the three datasets from 1980s, 2010 and 2015 in Guangdong Province, we analyzed the spatial and temporal variations of soil pH in farmlands in different regions of Guangdong Province and analyzed the driving factors for such variations. The results showed that the spatial distribution of soil pH in Guangdong Province changed significantly in different periods. During 1980s to 2010, soil pH showed an acidification trend with a decline of 0.3, and increased by 0.09 from 2010 to 2015, with more uneven trend and more obvious acid base differentiation. From the perspective of each region, there was generally a trend of acidification from the 1980s to 2010. From 2010 to 2015, the average pH value of farmland soil in the Pearl River Delta increased by 0.27, while that on the east wing and west wing decreased by 0.05 and 0.15 respectively, showing a unapparent change of soil pH in the mountainous area. Our results showed that soil acidification in diffe-rent areas of Guangdong Province was affected by natural factors such as soil itself and precipitation. In addition, anthropogenic factors such as acid rain, unreasonable fertilization and the planting structure of high-yielding crops were also the main causes of soil acidification. Industrialization, urbanization, mining development, and the spread of soil testing formula fertilization increased soil pH in local areas. Our results could provide theoretical guidance for controlling and alleviating soil acidification and improving the quality of cultivated land in different areas.

[Prediction of Distribution of Soil Cd Concentrations in Guangdong Province, China].

Heavy metals are one of the principal soil pollution sources. Contaminated soils affect the quality of agricultural products, and then threaten human health. Prediction of the contaminants distribution in the soil is the foundation of pollution evaluation and risk control. A total of 1000 soil profiles were collected to investigate the spatial variation of soil cadmium (Cd) concentration in Guangdong province. These datasets were divided into two groups, about 900 samples for model training and the other 100 for model validation. Six frequently used GIS spatial interpolation methods including Spline, Natural Neighbor, Ordinary Kriging, Inverse Distance Weighted, Local Polynomial Interpolation and Radial Basis Function, and Cubist which is a type of rule-based model were compared to determine their suitability parameters for estimating soil Cd concentration. Nine different resolutions including 2000, 1500, 1000, 800, 500, 300, 200, 150, and 90 m were selected to calculate, evaluate and compare their accuracy. The results showed that, 1 Quantitative assessment of the continuous surfaces showed that there was a large difference in the accuracy of the seven methods. Cubist was superior to GIS-based spatial interpolation methods at all resolutions. Cubist was the best tool for mapping the spatial distribution of Cd in soils with thirty-seven specific predictors relevant to the source and behavior of Cd (parent material, land use, soil type, soil properties, population density, gross domestic product per capita, and the lengths and classes of the roads surrounding the sampling sites, climatic factors, etc.) at 300 m×300 m resolution. The second was Spline, its accuracy was optimal at the 1500 m×1500 m resolution. 2 Results of Cubist suggested that the soil Cd spatial distribution was primarily dependent on the properties of soil regional parent materials. And soil samples with higher Cd concentration mainly located in Carboniferous and Quaternary areas. 3 Spatially, Cd concentrations were higher in the Pearl River Delta region and north of Guangdong Province. Many hotspots existed throughout the Pearl River Delta region due to transportation and pollution of the river. The major anthropogenic inputs of heavy metals to soils and the environment were metalliferous mining and smelting in the north of Guangdong Province. The soil Cd geometric mean concentration of 0.147 mg·kg-1 was lower than that of China, however it varied from zero to 6.056 mg·kg-1. The areas with soil Cd concentrations greater than 1.0 and 3.0 mg·kg-1 were 160 km2 and 2140 km2 respectively, accounting for 0.09% and 1.18% of the total area of Guangdong Province.

[Hyperspectral inversion of paddy soil iron oxide in typical subtropical area with Pearl River Delta, China as illustration]. / äºšçƒ­å¸¦å ¸åž‹åŒºåŸŸæ°´ç¨»åœŸæ°§åŒ–é“é«˜å ‰è°±åæ¼”­­ä»¥ç 江三角洲为例.

Iron oxide is the main form of iron element existing in the soil. In subtropical areas, the high-content iron oxide constitutes the soil's important coloring components, or its mineral substances, such as goethite and hematite, making the soil color apparently different from that in other climatic zones. The present paper, with the Pearl River Delta, a typical subtropical area, as illustration, and through analysis of the correlation between different spectral forms and the content of soil iron oxide, created inversion models of soil iron oxide by extracting characteristic spectral bands. The findings showed that there was a negative correlation between the content of soil iron oxide and the reflection spectrum, and the sensitive bands were mainly found in such visible near-infrared regions such as 404, 574, 784, 854 and 1204 nm. The correlation between the spectrum through differential processing and the soil iron oxide was significantly improved. On the basis of the correlation-prominent bands, the methods of both multiple linear regression and principal component analysis were adopted so as to remove collinear bands, and finally, characteristic bands were selec-ted to serve as the input parameters of inversion models. A comparison of the results revealed that the best inversion model of soil iron oxide content in the Pearl River Delta was BP artificial neural network (i.e., RMSEC=0.22, RMSEP=0.81, R2=0.93, RPD=12.20). It was applicable with excellent stability to the fast estimation of the iron oxide content in the soil and could hopefully serve as the research basis for the measure of the spatial distribution of the soil iron oxide.

[Statistical characteristic analysis of soil PAHs in a coking contaminated site of China].

Statistical characteristic analysis of pollutants in contaminated sites can help identify the origin, generation, and spatial variation of different pollutants, and can reduce the uncertainty of site survey data. Taking a large and abandoned contaminated coking site of China as the object, 114 surface (0-50 cm) soil samples were collected, with the statistical and spatial characteristics of 16 priority PAHs (sigmaPAHs) analyzed. The descriptive statistical analysis indicated that the sigmaPAH levels varied significantly, and the data were severely skewed. The correlation matrix (CM) and principal component analysis (PCA) showed that the extracted first two principal components (PCs) could effectively represent the whole site pollution data. Four pollutants, i. e., Baa, Bbf&Bkf, Bap, and Inp, were selected for trend analysis and spatial local variance analysis. In the east-west and north-south directions of the site, the pollution showed a low-high-low trend. The variation coefficient of pollution for the site was higher in the central, northwest, and southwest regions, while lower in the other regions.

[Heterogeneous characteristic of PAHs' spatial distribution in a large coking site of China].

A large coking contaminated site was selected to study the PAHs' spatial distribution probability in surface-soil (0-50 cm) through the indicator kriging of the non-parametric geostatistics, and the map of probability distribution with a contaminant target was plotted over the entire site. Results indicated that the indicator semivariograms were stable after the conversion of sample data, but the poor correlation of spatial samples was observed due to the spatial variability. In this site, the distribution of the contamination probability of four PAHs' showed a similar characteristic, and the areas with a probability of more than 45% were mainly distributed in production process workshops for coking, gas purification, tar products etc, of the central, northwest and southeast site with serious contamination, while the areas with a probability of less than 45% were mainly distributed in coal preparation, gas purification workshops of the southwest and northeast site. Based on the above analysis results, we can draw a conclusion that the forecast probability results are consistent with the occurrence and distribution of pollution sources, which is important for defining the remediation boundary and calculating the contaminated soil volume.

[Spatiotemporal variation of land disturbance on large opencast coal mine area].

Based on the Landsat-TM images in 1990 and 2005, and with principal component analysis, this paper studied the land use change on the Antaibu opencast coal mine of Pingshuo mine area in Shanxi Province in 1990-2005. The results showed that the spatial characteristic of spectra on the opencast coal mine varied with land type, area distribution, and landscape pattern. The first and second principal components of the TM images had obvious spatial characteristic, i. e., the first principal component highlighted the characters of excavation and transportation area, slope area, and cumuli and stripping area, while the second principal component highlighted the information of higher and lower vegetation-cover area. According to the land use type, the study area was classified into stripping area, excavated area, land reclamation area, and original landform area. In 1990 -2005, the excavated area had a little change, original landform area reduced by 15.263 km2 reclamation area increased by 8.513 km2, and stripping area increased constantly from 5.522 km2 in 1990 to 11.889 km2 in 2005.