Research On Small And Medium Scale Soil Moisture Monitoring Method Based On Cosmic Ray Neutron Method

Soil moisture is an important carrier in the circulation of environmental geological materials,and an important source and sink of water circulation in the soil-bio-atmosphere.Timely and accurate understanding of soil moisture content is of great significance for research in many fields such as agriculture,environment,hydrology,meteorology and ecology.At present,soil moisture monitoring is mainly based on point scale and large-scale remote sensing methods.There is a gap in the monitoring scale between the two methods.The Cosmic Ray Neutron Probe?CRNP?has a monitoring scale of a few hundred meters in radius,which just compensates for the gap in the monitoring scale between the above methods.In this study,CRNP was used as the technical means to study the summer maize farmland in the Huang-Huai-Hai Plain Compared with distributed sampling and frequency domain reflectance?FDR?,the applicability of this method in typical agricultural areas in northern China was evaluated.In addition,UAV remote sensing was used to accurately calibrate biomass water equivalent in order to further improve the monitoring accuracy of CRNP.The results are as follows:?1?The spatial weight factor of CRNP in the test area was determined.The measurement range of the CRNP is a flat gyro or funnel type with the sensor as the axis.Based on the CRNP spatial weight distribution theory of K?HLI et al.,the measurement range and spatial weighting factor distribution of CRNP in the study area are calculated.After calculation,the radial distance is 233 m,and the vertical measurement depth is negatively correlated with the distance from the measurement point to the CRNP,and the range is 30.1-20.7 cm.?2?Compare the CRNP with the results of distributed sampling and FDR measurements.During the growth period of summer maize?2018.06-2018.10?,the correlation coefficient R² of CRNP and the synchronous distributed sampling data was 0.94?n=6?,and the root mean square error was 0.018 cm³·cm^-3.The correlation coefficient R² between CRNP?24 h mean?and FDR?24 h mean?is 0.55?n=110?,and the RMSE is 0.035 cm³·cm^-3.?3?The biomass water equivalent of CRNP was corrected accurately by using high spatial resolution UAV remote sensing.Compared with the FDR data of the same period,the vegetation correction based on UAV remote sensing can effectively improve the measurement accuracy of CRNP.At the jointing stage?2018.07.21-08.15?,the average error of soil water content decreased from 0.024 cm³·cm^-3 to 0.009 cm³·cm^-3,and the average absolute error declined from 0.027 cm³·cm^-3 to 0.019 cm³·cm^-3.The RMSE drop to 0.024 cm³·cm^-3;the average error of the milk ripening period?2018.08.25-09.15?fell to 0.027 cm³·cm^-3,and the average absolute error was 0.047 cm³·cm^-3 is reduced to 0.034 cm³·cm^-3,and the RMSE is declined from 0.056 cm³·cm^-3 to 0.043 cm³·cm^-3.In summary,CRNP,as a new medium and small scale soil moisture monitoring method,has shown good applicability in the Huang-Huai-Hai farmland ecosystem.Because of its high spatial recognition ability and similar monitoring scale with CRNP,UAV remote sensing can accurately calibrate the impact of CRNP on vegetation and further improve the measurement accuracy of CRNP.This study is a coupling application of two new technologies in the field of soil moisture monitoring.The combination of the two technologies has very broad prospects for development and deserves more in-depth and extensive research.