【作者】 梁爽；

【导师】 杨国东；

【作者基本信息】 吉林大学 ， 大地测量学与测量工程， 2018， 硕士

【摘要】 积雪对土壤水热影响的研究具有十分重要的意义,可为农业生产、农业灌溉、水资源调节等提供科学指导,为春季土壤墒情的监测和预报提供理论依据。积雪的保温作用对下垫面土壤的温湿度变化和能量平衡具有重要影响;积雪作为陆地上重要的淡水资源,融化后产生大量水分,能够改变土壤的水文状况,春季土壤湿度也与冬季积雪参数密切相关。本文利用物理过程模型和遥感监测,采用模型模拟和利用遥感数据两种方法开展积雪对土壤温湿度的影响研究。积雪和土壤湿度的微波遥感反演以全天候、全天时以及大范围等优势,被广泛应用。对于地表以下冻土温度,由于冬季雪盖的影响,目前遥感方法却无法探测和获取,而基于积雪热力模型(SNTHERM)的物理过程模型方法则可精准的预测和模拟下垫面土壤温度的变化过程,适合积雪对土壤温度的影响机理研究。本文主要内容分为两个部分,第一部分为基于物理过程模型的方法研究积雪对冻土温度的影响。首先对物理过程模型进行了敏感性分析和精度评价,发现模型对于土壤温度的模拟精度较高,但对于土壤湿度,尤其是春季土壤湿度的模拟效果欠佳。因此,应用模型探究了积雪参数对下垫面冻土温度的影响。第二部分为基于遥感数据产品,采用GIS地学统计方法,分析东北农田区冬季积雪参数(最大雪深、平均雪深、积雪日数)对春季土壤湿度的影响,构建积雪参数与春季土壤湿度之间的回归方程。本文的主要结论如下:(1)SNTHERM模型的气象驱动数据的敏感性分析结果表明,土壤温湿度对大气温度和长波辐射通量最敏感,初始土壤特征参数的敏感性分析表明,土壤温度对初始输入的土壤特性参数并不敏感(RMSE<1℃),土壤湿度对土壤的塑性指数最敏感,因此准确获取气象驱动数据中的大气温度和长波辐射通量,以及土壤的塑性指数,可提高模型模拟精度;(2)积雪覆盖条件下,SNTHERM能够有效的模拟和预测冻土温度。SNTHERM浅层土壤(5cm、10cm、15cm)模拟精度优于深层土壤(20cm、40m),在积雪覆盖条件下,浅层土壤的均方根误差均在2℃之内,结果表明SNTHERM可用来进行雪盖下浅层冻土温度的模拟,为冻土温度的反演提供了新的思路;(3)积雪覆盖对下垫面浅层土壤的温度变化影响显著,对下垫面土壤湿度的影响并不明显,主要是由于冻土在冻结期,液态水较低,液态水迁移量很小。土壤温度的变化与积雪覆盖厚度具有一定的关系,当积雪厚度达到10cm以上时,下垫面土壤温度的变化显著,积雪越厚,积雪保温的效果越好;(4)1980-2009年冬季积雪参数与春季土壤湿度的相关性分析结果表明,东北农田区冬季年际积雪参数与春季土壤湿度(4、5月份)呈现显著正相关(置信度<0.05),显著象元个数占象元总数45%以上,对4月份土壤湿度影响强度大于5月份。最大雪深与土壤湿度的拟合效果好于平均雪深、积雪日数与土壤湿度的拟合效果,12月份最大雪深与土壤湿度的拟合效果好于其它月份的积雪参数,结果表明,最大雪深可作为春季土壤墒情预报的重要积雪参数。更多还原

【Abstract】 It is of great significance to study the influence of snow on soil water and heat.It can provide scientific guidance for agricultural production,agricultural irrigation,water resources regulation and so on.It provides a theoretical basis for monitoring and forecasting soil moisture in spring.The heat preservation effect of snow has an important influence on the temperature and moisture change and energy balance of the soil on the underlying surface.As an important fresh water resource on land,the snow accumulated a large amount of water after melting,which can change the hydrological condition of the soil,and the soil moisture is closely related to the winter snow accumulation parameters in spring.In this paper,using physical process model and remote sensing monitoring,two methods of model simulation and remote sensing data are used to study the influence of snow on soil water and heat.Microwave remote sensing inversion of snow and soil moisture is widely used for all-weather,all day and wide range.For the temperature of frozen soil below ground surface,the remote sensing method cannot be detected and obtained due to the effect of snow cover in winter,and the physical process model based on the snow thermal model(SNTHERM)can accurately predict and simulate the change process of frozen soil temperature in the underlying surface,which is feasible for the study of the influence mechanism of snow cover on soil temperature.The main content of this paper is divided into two parts.The first part is the study of the influence of snow on the frozen soil temperature based on the physical process model.First,the sensitivity analysis and accuracy evaluation of the physical process model have been carried out.It is found that the simulation accuracy of the model for soil temperature is high,but the simulation effect on soil moisture,especially in spring soil moisture,is not good.Therefore,the influence of snow parameters on the temperature of the underlying surface frozen soil is studied by the physical process model.The second part,based on remote sensing data products,uses GIS geoscience statistics to analyze the effects of winter snow cover parameters(maximum snow depth,mean snow depth and duration of snow cover)on soil moisture in spring in the northeast of China,and establish a regression equation between snow parameters and soil moisture in spring.The main conclusions of this paper are as follows:(1)The sensitivity analysis of SNTHERM meteorological driven data shows that soil temperature and moisture are most sensitive to atmospheric temperature and long wave radiation flux,and the sensitivity analysis of initial soil characteristic parameters shows that soil temperature is not sensitive to initial input soil characteristic parameters(RMSE<1 C),and soil moisture is most sensitive to soil plastic index.Therefore,accurate acquisition of atmospheric temperature and long wave radiation flux in meteorological driving data and plasticity index of soil can improve the accuracy of model simulation.(2)Under the condition of snow cover,SNTHERM can effectively simulate and predict the temperature of seasonal frozen soil.The simulation accuracy for shallow soil layers(5cm,10 cm,15cm)is better than the deep soil layers(20cm,40 m).Under the condition of snow cover,the root mean square error of the shallow soil is within 2 ℃.The results show that SNTHERM can be used to simulate the temperature of the shallow frozen soil under the snow cover,which provides a new way of the inversion of the seasonal frozen soil temperature.(3)The effect of snow cover on the temperature changes of the shallow layer of the underlying surface is significant,but the influence on the soil moisture is not obvious,mainly because the frozen soil is frozen,the liquid water is low,and the liquid water migration is very small.The change of soil temperature has a certain relation with the thickness of snow cover.When the thickness of snow is above 5cm,the change of soil temperature in the underlying surface is significant.With the increase of snow thickness,the better the effect of snow insulation.(4)The correlation analysis between the snow parameters and soil moisture of April and May in 1980-2009 year shows that the annual snow parameters have a significant positive correlation with soil moisture(confidence level <0.05)in the Northeast farmland,which is more than 45% of the study area,and the effect on soil moisture in April is better than that in May.The fitting effect of the maximum snow depth and soil moisture is better than the fitting effect of the average snow depth,the number of snow days and the soil moisture.The fitting effect of the maximum snow depth and the soil moisture in December are better than that in the other months.The results show that the maximum snow depth can be used as an important snow parameter for the prediction of soil moisture in spring.更多还原