BACKGROUND Selenium (Se) is one of the essential trace elements in the human body, which has many biological functions. Insufficient or excessive intake of Se will cause a series of diseases. The trace element Se in the human or animal body cannot be synthesized by itself, but can only be supplemented from external food. Se in food, especially in plant food, mainly comes from soil. Therefore, Se in soil is closely related to human health and animal growth. China is a country lacking in soil Se content, especially in Tibet. The background value of soil Se in Tibet is significantly lower than that of surface soil in China. Therefore, local diseases such as Kashin-Beck disease are common in some areas of Tibet due to long-term insufficient intake of selenium. It is of great significance to investigate the distribution characteristic of soil Se content, delineate the distribution area of selenium-enriched soil resources and determine the influencing factors of soil Se content for promoting the development and utilization of selenium-enriched land resources, develop Se-enriched industries and prevent local diseases. This will also provide reference data for the research of soil Se background value.

OBJECTIVES In recent years, it has been a hot topic to investigate the content of Se in soil, to delineate selenium-enriched soil resources and to develop and utilize them. Tibet is the main part of the Qinghai—Tibet Plateau, which has complex and diverse soil parent materials and soil forming processes, forming unique alpine soil types. In addition, Tibet is one of the areas with the least influence of human activities and is the ideal place for environmental geochemistry research. However, due to many factors such as natural geographical location and climate, the research data of soil element geochemistry in Tibet is very limited, and research data of soil Se is rare. Thus, the characteristics, distribution and influencing factors of soil Se content in the study area were studied, to provide a basis for the exploitation and utilization of selenium-enriched land resources, the development of selenium-enriched industry and the prevention of endemic diseases in the frontier ethnic areas of the plateau.

METHODS The collection, processing and analysis of samples of surface soil, vertical profile and rock profile were carried out. The samples were collected from the key farming area of Longzi County, Shannan, Tibet Autonomous Region. The surface soil samples were collected in a grid pattern from the third national land survey map spot. The soil sampling points were mainly arranged on agricultural plots, with an average sampling density of 7.9 points/km². A total of 1587 surface soil samples were collected, with a study area of 200km². The sampling method for surface soil samples was determined according to the actual plot shape. When the plot was square, "X" type sampling was adopted, and when the plot was rectangular, "S" type sampling was adopted. When sampling cultivated land, 5 sub-sampling points were equally combined into 1 sample; for grassland and woodland sampling, 3-4 sub-sampling points were equally combined into one sample. The samples collected at each sub-sampling point were crushed, small stones, roots and other sundries picked out, and after fully mixing, more than 1000g samples reserved and put into sample bags by quartering method. In the study area, 10 vertical soil profiles were set up, and the sampling interval was 1 sample/20cm. The depth of all the profiles was 160cm except for the profile CM01, which was 140cm deep. In addition, a rock profile was set in the study area, and fresh rock samples were collected. The same kind of rock was collected in a multi-point mode and combined into a sample, with the sample weight of 300g. The surface soil samples, and vertical profile samples collected were naturally dried without pollution, and sieved by -10 mesh nylon sieve, then divided by quartering method, weighed and put into sample bottles and sent to laboratory for analysis. Soil samples were analyzed for Se, available Se, organic matter, pH, N, P, available N, available P, available K, etc. Rock samples were analyzed for Se. The contents of Se and available Se were determined by atomic fluorescence spectrometry (AFS), organic matter. Available nitrogen and cation exchange capacity (CEC) were determined by volumetric method (VOL), pH value was determined by ion selective electrode method (ISE), N content was determined by elemental analyzer method (EA), and available P, available K, P and TFe₂O₃ were determined by inductively coupled plasma-optical emission spectrometry (ICP-OES). The detection limit, accuracy, precision and reporting rate of the analytical method adopted all met the specification requirements, and the sample analysis quality was reliable.

RESULTS The results of the content of Se in soil and its influence factors, showed that: (1) The Se content in the topsoil of the study area ranged from 0.14 to 1.51mg/kg, with a median of 0.44mg/kg, which was 2.9 times as high as the average value of Tibet (0.15mg/kg) and 1.5 times as high as the average value of China (0.26mg/kg). The content of available Se in the topsoil ranged from 0.8 to 26.8μg/kg, with a median of 9.2μg/kg. The content of available Se in topsoil was 0.21%-5.79% of total Se. (2) Se-enriched (Se≥0.4mg/kg) soil resource area was 154.53km², which accounted for 77.25% of the total area. Se-enriched soil was mainly distributed in Longzi Town and Ridang Town. There was no excess or deficiency of soil Se in the study area, which indicated that Se-enriched soil was continuous and had the potential to develop Se-enriched soil resources. (3) The geological background was closely related to the Se content in the soil. The Se rich soil was mainly controlled by the distribution of the Nieru Formation (T₃n) and the Ridang Formation (J₁r). The median Se content in the soil developed from the Nieru Formation (T₃n) and the Ridang Formation (J₁r) was 0.44mg/kg and 0.41mg/kg, respectively. Analysis of Se content in rock samples showed that Se content ranged from 0.07 to 11.00mg/kg, with an average of 1.65mg/kg. Se content was high in sericite slate and shale, which further proved that Se-enriched soil was closely related to its parent rock. (4) Soil physical and chemical properties including organic matter, pH, TFe₂O₃ had no significant effect on soil Se content, but soil available Se was positively correlated with organic matter, pH, N, P, alkali-hydrolyzable N, available P, available K, CEC content. There was a positive correlation between the content of organic matter and the content of available Se (R²=0.2792, P < 0.01), and between the content of organic matter and the ratio of available Se to total Se (R²=0.2597, P < 0.01). There was a positive correlation between soil available Se and pH (R²=0.103, P < 0.01). According to the soil pH grading standard, the availability of Se increased gradually from acid to alkaline soil, but in strong alkaline soil, the availability began to decrease due to the methylation reaction of Se. There was a negative correlation between available Se and TFe₂O₃ (R²=-0.346, P < 0.01). In the study area, the content of soil available Se had significantly positive correlation with the content of N, P, alkali-hydrolyzable N, available P and available K, which indicated that the increase of N, P and K content could significantly improve the bioavailability of soil selenium, which had a certain theoretical significance for the artificial control of soil Se content. (5) 10 vertical soil profiles were constructed in different areas of the study area, and the depth of the other profiles was 160cm except for the CM01 profile, which was 140cm. In that vertical soil profile, the content of Se and available Se decreased with the increase of soil depth. The content of Se in the soil below 100cm was less than 0.4mg/kg, and the content of available Se in the soil at 160cm was 60% less than that in the surface soil.

CONCLUSIONS The content of Se in the topsoil of the study area is high, and 77.25% of the study area is in line with the standard of Se-enriched soil. In that soil, the Se content is mainly affected by the parent materials, especially the sericite slate and shale in the Nieru Formation (T₃n) and Ridang Formation (J₁r). The land use type has little effect on Se content and distribution. Physical and chemical properties of the soil, such as organic matter, pH, N, P, available N, available P, available K and CEC, have little effect on total Se content, but soil available Se is significantly positively correlated with organic matter, pH, N, P, alkali-hydrolyzable N, available P, available K and CEC. Soil nutrient management can further improve bioavailability of soil selenium. Only the characteristics and influencing factors of soil Se content in Longzi County, Tibet Autonomous Region, were discussed, in order to provide a geological basis for the development and utilization of Se-enriched land resources. However, the process of Se uptake by crops is a very complex biogeochemical process, and is affected by many factors. Therefore, it is necessary to further strengthen research on the characteristics of Se content and its migration and transformation in soil-crop system.