Research And Application On The Satellite Altimetry Data To Monitor The Water Level Of Lakes And Reservoirs In China

Lakes and reservoirs are important surface water resources,which are closely related to our production and life.It is of great significance to obtain the water level change of lakes and reservoirs in time.Satellite altimetry technique used to monitor the lake water level,it has the unique advantages of high precision,all-weather,periodical revisiting,low monitoring cost and the coverage of the no-data area.To some extent,the use of this technology to fill ground monitoring has become a development trend.But the current study,especially in China it is also mainly in the study stage of a single lake or a few lakes in a particular area,also failed to rich altimetry data applications across the country,from the business of monitoring the goal of lake water level in our country also has the very big disparity.To this end,this paper carried out the application research on the monitoring of the water level of the lakes and reservoirs in China,and the main research work and conclusions are as follows:(1)The method of extracting water level based on the evaluation of satellite altimeter data quality.For solving satellite altimetry data quality is not stable,A method for extracting water level based on high data quality is proposed.Using the Jason-3 test data,we selected hongze lake area with unstable data quality as an example.The results show that this method can quickly and effectively remove abnormal data,the method to improve data accuracy obviously superior to the traditional method.The correlation coefficient between the satellite altimetry water level and the measured water level was increased from 0.11 in the traditional method to 0.59,and the root mean square error decreased from 2 m to 0.5 m.It is verified that the jason-3 data is highly reliable for water level monitoring in lakes and reservoirs.This paper solves the problem that the traditional method is not applicable to the data of the lake area with more noise and different quality,and improves the availability and monitoring accuracy of the altimetry data.In addition,for the data quality is bad in some cycles,the extracted water level is generally not high precision,based on the data quality evaluation results,they can be discarded,can further improve the monitoring precision of the whole,the correlation coefficient can be increased to 0.9,the root mean square error is reduced to 0.19 m,is it no data lakes and reservoirs to build accurate capacity curve is of great significance.(2)Analysis of the general monitoring capability of the satellite altimeter data on the water level of China's lakes and reservoirs.Based on satellite altimetry characteristics and The basic principle of altimeter data monitoring lake water level,combined with the satellite ground track file,the national water distribution information,a series of spatial analysis with GIS technology experiment,obtain our national lakes and reservoirs.specific number,name list and basic monitoring conditions who can be monitored by the short cycle revisited the altimeter satellite(Jason – 2/ 3,Sentinel – 3a/3b).And on the basis of the five great lakes(Tibetan lake area,the southwest and the east lake area northwest,northeast lakes and lake area)based on the surrounding terrain,the surface size,shape features,systematic selection of multiple lakes and reservoirs as the characteristic lake library for experimental study.(3)Satellite altimetry data monitoring water level experiment and applicability analysis of lakes and reservoirs in China.The water level monitoring experiment of 44 characteristic lakes and reservoirs was conducted using two series of altimetry data by Jason and sentinel-3.The monitoring evaluation system is constructed and the monitoring results are verified by the measured data.Both data have good monitoring results when minimum effective monitoring conditions are met(Jason data is in the mountain type lakes and reservoirs,with the intercept > 1.6km,the narrowest surface width > 0.5 km.When the terrain is flat,it needs to intercept >,1.4 km,and the narrowest water surface is about 0.4 km.Sentinel data need to intercept > 0.6km,the narrowest surface width of > 0.2km.).Jason data monitoring lake library mean absolute error(MAE)concentration in the range of 15-36 cm,root mean square error(RMSE)concentration in the 17th-43 cm range,correlation coefficient(R2)concentrated in the 0.5 0.96 range.For sentinel-3 data,the MAE is concentrated in the range of 5-29 cm,the RMSE concentration in the range of 7-38 cm and R2 in the range of 0.7-0.99.Furthermore,the influence of lake area,water surface size and topography on the monitoring accuracy is analyzed.In the lake district,the latitude and elevation differences of different lakes have no obvious influence on the monitoring accuracy.On the terrain difference,Jason data under the same monitoring condition,the more flat terrain monitoring accuracy is higher.The sentinel-3 data were less affected by surrounding topography,and the monitoring accuracy was higher,and no significant difference was observed.This improves the tracking mode and greatly reduces the orbit resolution.Made in different types of lake library monitoring,overall were superior to Jason data,especially small and medium-sized lakes and reservoirs,and can be detected to a smaller lakes and reservoirs,enhance the monitoring capabilities.