Spatial Variability Of Trace Elements In Soil And Rice Grain In Three Major Landforms, Zhejiang Province

Soil is the pre-eminent source of most biologically active micronutrients. Rice is the most important crop in Southeast China. Consumption of rice is the main source of micronutrients to human. And observations of soil and rice properties are always related to particular locations in space. Knowledge on magnitude and spatial distribution of micronutrient concentrations in both soil and rice grain provides approaches for assessing micronutrient level in both soil and grain and facilitating future site specific management and fertilization.The Survey on Agri-Geological Environment under way in China is conducted on the 1km~2 grid, taking the area within the grid as a mixed sample. However, in Zhejiang Province, the geomorphology is complex, soil types are diverse, and land is intensively cultivated. In an area of 1km~2, soil properties will change greatly. This induces the application difficulty of the Survey on Agri-Geological Environment. Hence, it is necessary to study the spatial variability of trace elemental concentrations in soil and rice grain within a 1km~2 grid.An investigation was conducted is to survey the magnitude and spatial distribution of trace elemental concentrations in soil and rice grain, as well as soil properties in three major landforms of Zhejiang Province, that is, the river valley plain in Jinqu Basin, Hangjiahu water net plain, and coastal plain in the south of Hangzhou Bay. A total of 584 soil samples (0-20cm) and 259 rice grain samples were obtained simultaneously in mid October, 2005 when the rice was ready for harvest. The soil was sampled on grids at 100-m intervals, with additional nested distances of about 60 m, 36 m, 21 m, 12 m, 7 m, 4 m, 2 m, 1 m. And in the valley plain and coastal plain, surveying along transects was used to determine the size and frequency of soil changes. The results obtained can supply information for the agricultural users on the micronutrient soil levels as well as for land use planning. The data will be of use for further detailed studies in the regions as well as complementing the geochemical database.The main conclusions were summarized as follows:(1) The total concentrations of trace elements in soils are close to the background level of Zhejiang Province and China, and are mostly within the limitation of first level of national soil environmental standards. DTPA-extractable trace elemental concentrations in soil usually are within the range of abundance to significant abundance level. And most trace elemental concentrations in rice grain confirm to the limitation of national standards.(2) There is great variability for trace elements concentrations in soil and rice grain, within the grid of 1km~2. Using geostatistics, their spatial structures were analyzed and semivariogram models were determined for kriging estimates. In Hangjiahu water net plain, the spatial variation exists between the east and the west of the plot. In the valley plain, the spatial distribution of trace elemental concentrations show great spatial variability from the bottom to the upland, perpendicular to the river. And in the coastal plain, the trace elements concentrations changed greatly along the transect from the coast to the inland.(3) The main factors influencing the spatial distribution of trace elements were determined from analysis of variance and scale effect. In the valley plain, soil types can explained the most variation of soil trace elemental concentrations. In the coastal plain, the reclaimed area during different historical stage is the main cause of soil trace elemental concentrations. In the Hangjiahu water net plain, the main variation may resulted from the difference of agricultural management measures within the field.(4) Coregionalization usually exists between trace elemental concentrations in rice grain and that of soil, as well as soil properties. The spatial distribution of one trace element in rice grain is similar to that of the same element in soil. Soil factors can indicate the spatial feature of trace elements in rice grain. By virtue of coregionalization theory, soil factors as the secondary variables, can improve the estimate accuracy and reduce the cost on trace elements concentrations measurement.(5) Three principal components were extracted from soil total elemental concentrations and soil properties. Copper, iron, nickel, and lead are combined together in the Hangjiahu water net plain. Copper, iron, manganese, and nickel were grouped together in the valley plain. And iron, manganese, lead showed paragenetic feature in the coastal plain. The kriged maps of the principal components showed obvious similarity to the maps of the variables involved in each group. Thus the principal components can reflect the common resources or distribution processes. Their maps can be used to predict the general spatial distribution pattern of certain elements with similar geochemical behavior.(6) There are great differences in parent materials, soil-forming environment, soil properties, trace elemental concentration in soil and rice grain. This illustrates that the weather condition, topography, and soil-forming process are the main factors controlling the spatial distribution of soil and rice properties at a larger scale.(7) Low variation for total metal concentrations within each soil types indicated that soil types could be used as a basic unit for assessing the metal contents of soils. This result proved the ability of soil classification system to account for the spatial variability of trace elemental concentrations, which is a significant tribute to the usefulness of soil taxonomy to represent soil differences. It seems advisable to combine soil classification with geostatistics in future research and management zones development.(8) The soil sampling strategy integrating grids at 100-m intervals, unbalanced nested distances, and the sampling along transects. Grid sampling was chosen for mapping the distribution of the elements throughout the whole region, and detecting the trends if they existed. Nested sampling is an economical way of determining the spatial scale, knowledge of which can be used to plan future sampling. Surveying along transects is an often used and economical way of determining the size and frequency of soil changes. The strategy combing above sampling method can be used for analyzing the spatial variability of soil properties at different scales and that of multi-variables. The sampling distances in the study regions of the three major landforms for capturing the main variability of trace elements, are 10m for the study area in the water net plan, 100 m for the valley plain, and 500 m for the coastal plain, respectively.