[Effects of soil data and map scale on assessment of total phosphorus storage in upland soils.] / ä¸åŒåˆ¶å›¾å°ºåº¦åŠåœŸå£¤æ•°æ®å¯¹åœŸå£¤å ¨ç£·å‚¨é‡ä¼°ç®—çš„å½±å“.

Accurate assessment of total phosphorus storage in farmland soils is of great significance to sustainable agricultural and non-point source pollution control. However, previous studies haven't considered the estimation errors from mapping scales and various databases with different sources of soil profile data. In this study, a total of 393×104 hm2 of upland in the 29 counties (or cities) of North Jiangsu was cited as a case for study. Analysis was performed of how the four sources of soil profile data, namely, "Soils of County", "Soils of Prefecture", "Soils of Province" and "Soils of China", and the six scales, i.e. 1:50000, 1:250000, 1:500000, 1:1000000, 1:4000000 and1:10000000, used in the 24 soil databases established for the four soil journals, affected assessment of soil total phosphorus. Compared with the most detailed 1:50000 soil database established with 983 upland soil profiles, relative deviation of the estimates of soil total phosphorus density (STPD) and soil total phosphorus storage (STPS) from the other soil databases varied from 4.8% to 48.9% and from 1.6% to 48.4%, respectively. The estimated STPD and STPS based on the 1:50000 database of "Soils of County" and most of the estimates based on the databases of each scale in "Soils of County" and "Soils of Prefecture" were different, with the significance levels of P＜0.001 or P＜0.05. Extremely significant differences (P＜0.001) existed between the estimates based on the 1:50000 database of "Soils of County" and the estimates based on the databases of each scale in "Soils of Province" and "Soils of China". This study demonstrated the significance of appropriate soil data sources and appropriate mapping scales in estimating STPS.

[Study on canopy spectral characteristics of paddy polluted by heavy metals].

Because of frequent mining, heavy metals are brought into environment like soils, water and atmosphere, resulting heavy metal contamination in the agricultural region beside mines. Heavy metals contamination causes vegetation stress like destruction of chloroplast structure, chlorophyll content decrease, blunt photosynthesis, etc. Spectral responses to changes in chlorophyll content and photosynthesis make it possible that remote sensing is applied in monitoring heavy metals stress on paddy plants. Field spectroradiometer was used to acquire canopy reflectance spectra of paddy plants contaminated by heavy metals released from local mining. The present study was conducted to (1) investigate discrimination of canopy reflectance spectra of heavy metal polluted and normal paddy plants; (2) extract spectral characteristics of contaminated paddy plants and compare them. By means of correlation analysis, sensitive bands (SB) were firstly picked out from canopy spectra. Secondly, on the basis of these sensitive bands, normalized difference vegetation indices (NDVI) were established, and then red edge position (REP) was extracted from canopy spectra via curve fitting of inverted Gaussian model. As a result of correlation analysis, 460, 560, 660 and 1 100 nm were considered respectively as sensitive band for Pb, Zn, Cu and As concentration in paddy leaves. Furthermore, heavy metal concentrations (Pb, Zn, Cu and As) were significantly correlated with NDVIs (Pb, NDV(510, 810); Zn, NDVI(510, 870; Cu, NDVI(660, 870); As, NDVI(510, 810)). Heavy metals were also significantly correlated with REP, however, the inflexion termed as spectral critical value (SCV) between low and high heavy metals concentrations should be considered during applying REP in remote sensing monitoring. Moreover, NDVI and REP are much better than SB in terms of capability of expressing spectral information. Therefore, heavy metals contamination in paddy plants can be remotely monitored via ground spectroradiometer when NDVI and REP are selected as spectral characteristics.

[Dynamics of recent cultivated land in Zhejiang Province and relevant driving factors].

Through the human-computer interactive interpretation of the 2000, 2005, and 2008 remote sensing images of Zhejiang Province with the help of RS and GIS techniques, the dynamic database of cultivated land change in the province in, 2000-2008 was established, and the driving factors of the cultivated land change were analyzed by ridge regression analysis. There was a notable cultivated land change in the province in 2000-2008. In 2000-2005 and 2005-2008, the annual cultivated land change in the province arrived -1.42% and -1.46%, respectively, and most of the cultivated land was changed into residential and industrial land. Non-agricultural population rate, real estate investment, urban green area, and orchard area were thought to be the main driving factors of the cultivated land change in Zhejiang Province, and even, in the developed areas of east China.

[Simulation of methane emissions from rice fields in the Taihu Lake region, China by using different unit of soil database with the DNDC model].

Application of a biogeochemical model, DeNitrification and DeComposition or DNDC, was discussed to assess the impact of CH4 emissions on different soil database from rice fields in Taihu Lake region of China. The results showed that CH4 emissions of the polygon-based soil database of 1:50000, which contained 52034 polygons of paddy soils representing 1107 paddy soil profiles extracted from the latest national soil map (1:50000), were located within the ranges produced by the county-based soil database of 1:50000. However, total emissions of the whole area differed by about 1680 Gg CH4-C. Moreover, CH4 emissions of the polygon-based soil database of 1:50000 and the county-based soil database of 14,000,000, which was the most popular data source when DNDC model was applied in China, have a big estimation discrepancy among each county-based unit in spite of total emissions of the whole area by a difference of 180 Gg CH4-C. This indicated that the more precise soil database was necessary to better simulate CH4 emissions from rice fields in Taihu Lake region using the DNDC model.

[Estimates and affecting factors of soil organic carbon storages in Yunnan-Guizhou-Guangxi Region of China].

Soil organic carbon (SOC) plays a key role in the terrestrial eco-systems. However, there is a large variation in SOC estimates at regional and global scales. In order to improve the accuracy of SOC estimates, the SOC storage in Yunnan-Guizhou-Guangxi Region of China (include Yunnan Province, Guizhou Province and Guangxi Zhuang Municipality) was estimated using 798 soil profiles and 1:500 000 digitized soil map, and the dominant affecting factors on SOC density were also discussed employing stepwise regression and path analysis. Results showed that the SOC storages estimated in the 0-20 cm and 0-100 cm layers are 4.39 Pg and 10.91 Pg, respectively; and the corresponding SOC density are 56.2 Mg x hm(-2) and 139.8 Mg x hm(-2), respectively. The mean SOC density of Yunnan-Guizhou-Guangxi Region is higher than that of China. The environmental factors (including altitude, longitude, latitude, annual mean precipitation and annual mean temperature), soil parent materials and land use could explain 37.9% and 30.7% of the variability of SOC density to the upper 20 cm and 100 cm, respectively. The environmental factors are the dominant affecting factors of SOC density. The effect of temperature is more important than that of precipitation, and the temperature and precipitation mainly vary with altitude and latitude, respectively. Except for temperature and precipitation, there are also other factors varying with altitude, longitude and latitude significantly affect SOC density. And the effects of other factors are more important than that of precipitation.

[Estimation models of vegetation fractional coverage (VFC) based on remote sensing image at different radiometric correction levels].

The images of post atmospheric correction reflectance (PAC), top of atmosphere reflectance (TOA), and digital number (DN) of a SPOT5 HRG remote sensing image of Nanjing, China were used to derive four vegetation indices (VIs), i. e., normalized difference vegetation index (NDVI), transformed vegetation index (TVI), soil-adjusted vegetation index (SAVI), and modified soil-adjusted vegetation index (MSAVI), and 36 VI-VFC relationship models were established based on these VIs and the VFC data obtained from ground measurement. The results showed that among the models established, the cubic polynomial models based on NDVI and TVI from PAC were the best, followed by those based on SAVI and MSAVI from DN, with the accuracy being slightly higher than that of the former two models when VFC > 0.8. The accuracy of these four models was higher in middle-densely vegetated areas (VFC = 0.4-0.8) than in sparsely vegetated areas (VFC = 0-0.4). All the established models could be used in other places via the introduction of calibration models. In VI-VFC modeling, using VIs derived from different radiometric correction levels of remote sensing image could help mining valuable information from remote sensing image, and thus, improving the accuracy of VFC estimation.