Search

Division of accurate management area can make a great contribution to the ecological conservation in precision agriculture, particularly on preventing excessive fertilization and pesticides. There is a regularly uniform distribution of management zoning in different years in the same field. It is necessary to quantitatively analyze the spatial-temporal changes of zoning for better field accurate management according to local conditions. Most previous studies focused on the accurate partition, with emphasis on the selection of input quantity for high accuracy of zoning. The innovation of this study lies in the quantitative expression of a multiyear pattern after the division, together with the influencing factors of zoning. A corn field was selected as the research area at the Youyi Farm in Heilongjiang Province of northeastern China. The remote sensing images were captured from the Sentinel-2A satellite under the European Space Agency (ESA) during the corn seedling stage in the first ten days of June 2017-2020. The normalized difference vegetation index (NDVI) was extracted from the preprocessed images on the ArcGIS software. The object-oriented segmentation was used to segment NDVI images, where the coefficient of variation was selected to evaluate each segmentation. The coefficients of variation in the NDVI were reduced by more than 70% after segmentation. In the partition, a natural breakpoint was used to classify the NDVI. A superposition analysis was utilized to calculate the spatial transfer matrix. The results showed that the patterns of maize in the seedling stage were similar in the study area. The coefficients of variation for the elevation and slope after segmentation were reduced by 42.857%-57.143% and30.723%-34.940%, respectively, indicating that the growth of crops was affected by terrain factors. There was also a significant correlation between the NDVI and terrain factors, such as the elevation and slope. The sensors of soil temperature and humidity were embedded in four different positions on the slope along the ridge line with obvious topographic relief. A line chart or histogram was obtained for the four NDVI, the average values of soil moisture and temperature on the same ridge line according to the elevation changes in June. The highest NDVI values in the four stages were observed on the sunny slope, while gradually decreased from the top to the bottom of a slope. The reason was that the soil temperature of the sunny slope was the highest, while the soil moisture was sufficient, suitable for the emergence conditions of crops, where the emergence was faster with the high NDVI. On the shady slope, the soil temperature was relatively low, unsuitable for the emergence conditions with a low NDVI. In a specific field, the precise management pattern was similar to the same crop in the growth period over many years. The growth of crops at the early stage and the zonings of precise management depended mainly on the distribution of soil moisture and temperature under the different topographies. The findings can provide a spatiotemporal framework to integrate accurate management with variable fertilization and pesticides in precision agriculture. [ABSTRACT FROM AUTHOR]

: Soil organic matter (SOM) can improve the physical, chemical and biological properties of the soil through a variety of functions. An important role of SOM is performed on the soil function and quality, further to prevent emission of greenhouse gas in global carbon circulation. Spectral characteristics of SOM depend mainly on types of soils, as well as different physical and chemical properties. Previous models that constructed by the hyperspectral reflectance or spectral absorption characteristics often lead to the low accuracy in SOM prediction, due mainly to the input type structure was single. In order to improve the accuracy and speed of the prediction model, specific characteristic variables can be selected to reduce the high collinearity between spectral bands, where there is a large amount of hyperspectral data in the presence of redundancy and overlap. The spectral index is set to minimize the influence of independent wavelengths on iterative calculation. Furthermore, the topography significantly determines the surface microclimate, the movement of water on the surface and in the soil, as well as the process of material redistribution. In this study, taking the Hailun City, Heilongjiang Province as the research area, a SOM prediction random forest (RF) model was established for the different types of soil, in order to improve the accuracy of the SOM hyperspectral model. The characteristic bands were selected by a Competitive Adaptive Reweighted Sampling (CARS), while, the Digital Elevation Model (DEM) data and spectral index were data sources. The results showed that: 1) In CARS screening, the characteristic bands of each soil type were compressed to less than 16% of the total wavelength number, which greatly reduced the dimension of soil hyperspectral variables and computational complexity, thereby to improve the prediction ability of the calibration model. The CARS was suitable for the extraction of characteristic key wavelength variables, further to optimize model structure. 2) Three types of input variable that extracted by the grouping experiment were then utilized the prediction of different types of SOM. After grouping, the SOM prediction accuracy depended mainly on the type of soil. Specifically, the maximum prediction accuracy achieved in the Boggy soil of 0.768, where the Ratio of Performance to Interquartile distance (RPIQ) was 3.568. Black soil was the second most accurate. The prediction accuracy of meadow soil was the lowest, only 0.674, and RPIQ was 1.848. The RPIQ for the three types of soil was above 1.8, indicating good prediction ability of the model. 3) Local regression was conducted to improve the prediction accuracy of SOM. The local regression prediction accuracy was the best. The adjusted coefficient of determination, RMSE and RPIQ of the validation set were 0.777, 0.581%, and 2.689, respectively, indicating the model was highly stable. The proposed prediction factors can be used to realize the rapid prediction of RF-SOM, where the traditional complex program can be simplified. The findings can provide a promising basis for the selection of input variables, thereby to predict the types of SOM in different regions. [ABSTRACT FROM AUTHOR]

Soybean is widely cultivated in farmlands with irregular terrain in the northeast China. In this study, the influence of terrain on crop yield was investigated. Dongxing Cooperative of Hailun city in Heilongjiang Province was selected as the area of research. The study area was 10.20 hm2. The iRTK data were measured and utilized to produce the Digital Elevation Model (DEM) used for division of the sampling points’ slope position. The soil data and daily meteorological data were obtained. The soybean yield data along with the field management data at every sampling point during the growing season of soybean were also determined. The decision support system for agrotechnology transfer (DSSAT) model was used to simulate the soybean growth and development process. Seven sampling points with distinct terrain features were randomly selected from sixteen sampling points, and the data from the other sampling points were used for verification of the DSSAT model. Then the adjusted DSSAT model was used to simulate the development of soybean at each sampling point and its yield. In regards to temperature, slope, solar radiation, soil nutrients, etc., DSSAT model was put into use for yield results simulation under the condition of single factor discrepancy, and then the contribution degree of individual factor to the yield difference was examined. Data such as biomass, soil water content and soil alkali-hydrolyzed nitrogen content of each sampling point were collected during the soybean growth period, and the total solar radiation data of the study area was generated by DEM combined with the Angstrom equation. The results showed that the normalized root mean square error(nRMSE) for the simulated yield of the DSSAT model and the measured yield was 7.9%, indicating an excellent simulation. The nRMSE for leaf area index (LAI) and above-ground biomass simulation were 16.2% and 18.7%, respectively. It indicated that the DSSAT model was reliable for simulation of soybean yield, biomass and growth. The growing environment of soybean was affected by the terrain, which determined the spatial change of soybean yield. In comparison with the initial soil condition, the difference of soybean yield was more affected by temperature and slope. Compared with the other positions of slope, the soybean yield at the top and bottom of slope was higher and the yield variability was smaller. The water requirement for soybean increased during the pod-filling period of soybean and the soil water content was low in the sunny slope, which resulted in the decrease of yield. The high amount of alkali-hydrolyzed nitrogen content and soil water content were found at the bottom of the slope that could meet the needs of crop growth. Therefore, the appropriate irrigation measures should be taken to improve the soybean yield during this period. The analysis for crop yield gap could reveal the factors that restrict crop production. The outcomes of the research are conducive to the farmland’s fine operation during the precise management and can provide valuable information for field management. [ABSTRACT FROM AUTHOR]