The Assessment Of Soil Quality On Cotton Field Based On Remote Sensing Technology

Xinjiang is the major region of cotton production in China, It has great significance toevaluate the soil quality of cotton fields for producers and managers.This research take theadvantages of grid data management by remote sensing technology, combined with evaluationindex such as NDVI, which extracted by remote sensing technology, and a small amount ofphysical and chemical index of measuring point data. Research a quick, convenient andlossless method to obtain the soil attribute information of the cotton fields, using themulti-spectral remote sensing data to construct cotton soil quality evaluation index system,explore the application of remote sensing technology on cotton soil quality evaluation.From2009to2011, TM data, HJ satellite data, digital maps, field measured data wereused; years study about soil evaluation indexes, include physical indicators, nutrientindicators, biological indicators in the study area were referenced; and remote sensing imageprocessing methods such as: supervised classification, correlation analysis, principalcomponent analysis, vector and raster conversion and other technical means were utilized,using geographical information system and remote sensing image analysis softwareprocessing platform to get information for cotton fields in soil quality evaluation index system,and to evaluate soil quality, the main research results are as follows:(1) Achieved basic information of cotton field by different kinds of RS image analysismethods, establishes the foundation for monitoring cotton field’s soil quality. Study thedifference spectral information between various land use patterns and growing characteristics,and extract information about position and area of cotton field in study area; use real samplesto classified the soil texture; obtain the topography information by DEM data; and gain thesoil type classification by layer-stacking the National Soil map and remote sensing image ofstudy area. The conclusion is that it is feasible and effective to extract physical indicators ofcotton soil quality evaluation by indigenous satellite.(2) Study the space distribution of soil fertility and the growing condition of crops, findout that: the normalized difference vegetation index NDVI classification map is obtained bydensity slicing can reveal the corresponding region of cotton growing conditions, and soilorganic matter, alkali-hydro nitrogen, rapidly-available potassium content, content of51-63mg/kg on the range of available P2O5in the spatial layout are basically identical.Multi-temporal NDVI data can reveal the growing situation of cotton field and it can be aaccording to evaluate the soil quality, meanwhile, it laying a good theory foundation formonitoring and evaluating soil quality.(3) It is concluded that: the key period of cotton growing for production can implyimportant information to predicted yield. Using Multi-temporal vegetation index and somesample data to achieve inversion model. Results shows that: vegetation index, especially theratio vegetation index(RVI) and normalization vegetation index (NDVI), are not only haverelationships with soil nutrient such as available nitrogen, soil organic matter, but also hassignificant correlation with cotton single boll weight, yield and total output of residential area.It will become one of the most important indexes to evaluate cotton field soil quality, andimprove the scientificalness and veracity of the evaluation. (4) Structure the cotton Soil quality evaluation indicator system, tentatively. The systemmake use of the relationship between multi-temporal remote sensing, cotton yield, and soilphysical and chemical properties, comprehensive utilization of soil fertility index, use variedanalysis method evaluated the cotton soil quality. The evaluation index system including:1,physical indexes: land use type, soil texture, soil type and topography conditions;2, nutrientindexes: soil organic matter, alkali solution nitrogen, available phosphorus, availablepotassium content;3, biological indicators: key growth period long, vegetation index, thecotton yield. It also test and verify that the indicator system is rapid, accurate and manageablewhen evaluating and it will be appropriate for informatization management of cotton field.