Estimating soil moisture using remote sensing data: A machine learning approach

Soil moisture is an integral quantity in hydrology that represents the average conditions in a finite volume of soil. In this paper, a novel regression technique called Support Vector Machine (SVM) is presented and applied to soil moisture estimation using remote sensing data. SVM is based on statistical learning theory that uses a hypothesis space of linear functions based on Kernel approach. SVM has been used to predict a quantity forward in time based on training from past data. The strength of SVM lies in minimizing the empirical classification error and maximizing the geometric margin by solving inverse problem. SVM model is applied to 10 sites for soil moisture estimation in the Lower Colorado River Basin (LCRB) in the western United States. The sites comprise low to dense vegetation. Remote sensing data that includes backscatter and incidence angle from Tropical Rainfall Measuring Mission (TRMM), and Normalized Difference Vegetation Index (NDVI) from Advanced Very High Resolution Radiometer (AVHRR) are used to estimate soil water content (SM). Simulated SM (%) time series for the study sites are available from the Variable Infiltration Capacity Three Layer (VIC) model for top 10 cm layer of soil for the years 1998–2005. SVM model is trained on 5 years of data, i.e. 1998–2002 and tested on 3 years of data, i.e. 2003–2005. Two models are developed to evaluate the strength of SVM modeling in estimating soil moisture. In model I, training and testing are done on six sites, this results in six separate SVM models – one for each site. Model II comprises of two subparts: (a) data from all six sites used in model I is combined and a single SVM model is developed and tested on same sites and (b) a single model is developed using data from six sites (same as model II-A) but this model is tested on four separate sites not used to train the model. Model I shows satisfactory results, and the SM estimates are in good agreement with the estimates from VIC model. The SM estimate correlation coefficients range from 0.34 to 0.77 with RMSE less than 2% at all the selected sites. A probabilistic absolute error between the VIC SM and modeled SM is computed for all models. For model I, the results indicate that 80% of the SM estimates have an absolute error of less than 5%, whereas for model II-A and II-B, 80% and 60% of the SM estimates have an error less than 10% and 15%, respectively. SVM model is also trained and tested for measured soil moisture in the LCRB. Results with RMSE, MAE and R of 2.01, 1.97, and 0.57, respectively show that the SVM model is able to capture the variability in measured soil moisture. Results from the SVM modeling are compared with the estimates obtained from feed forward-back propagation Artificial Neural Network model (ANN) and Multivariate Linear Regression model (MLR); and show that SVM model performs better for soil moisture estimation than ANN and MLR models.     

The contribution of tropical cyclones to rainfall in Mexico

Investigating the contribution of tropical cyclones to the terrestrial water cycle can help quantify the benefits and hazards caused by the rainfall generated from this type of hydro-meteorological event. Rainfall induced by tropical cyclones can enhance both flood risk and groundwater recharge, and it is therefore important to characterise its minimum, mean and maximum contributions to a region or country’s water balance. This work evaluates the rainfall contribution of tropical depressions, storms and hurricanes across Mexico from 1998 to 2013 using the satellite-derived precipitation dataset TMPA 3B42. Additionally, the sensitivity of rainfall to other datasets was assessed: the national rain gauge observation network, real-time satellite rainfall and a merged product that combines rain gauges with non-calibrated space-borne rainfall measurements. The lower Baja California peninsula had the highest contribution from cyclonic rainfall in relative terms (∼40% of its total annual rainfall), whereas the contributions in the rest of the country showed a low-to-medium dependence on tropical cyclones, with mean values ranging from 0% to 20%. In quantitative terms, southern regions of Mexico can receive more than 2400 mm of cyclonic rainfall during years with significant TC activity. Moreover, (a) the number of tropical cyclones impacting Mexico has been significantly increasing since 1998, but cyclonic contributions in relative and quantitative terms have not been increasing, and (b) wind speed and rainfall intensity during cyclones are not highly correlated. Future work should evaluate the impacts of such contributions on surface and groundwater hydrological processes and connect the knowledge gaps between the magnitude of tropical cyclones, flood hazards, and economic losses.     

Evaluating the suitability of TRMM satellite rainfall data for hydrological simulation using a distributed hydrological model in the Weihe River catchment in China

The objective of this study is to quantitatively evaluate Tropical Rainfall Measuring Mission (TRMM) data with rain gauge data and further to use this TRMM data to drive a Distributed Time-Variant Gain Model (DTVGM) to perform hydrological simulations in the semi-humid Weihe River catchment in China. Before the simulations, a comparison with a 10-year (2001-2010) daily rain gauge data set reveals that, at daily time step, TRMM rainfall data are better at capturing rain occurrence and mean values than rainfall extremes. On a monthly time scale, good linear relationships between TRMM and rain gauge rainfall data are found, with determination coefficients R² varying between 0.78 and 0.89 for the individual stations. Subsequent simulation results of seven years (2001-2007) of data on daily hydrological processes confirm that the DTVGM when calibrated by rain gauge data performs better than when calibrated by TRMM data, but the performance of the simulation driven by TRMM data is better than that driven by gauge data on a monthly time scale. The results thus suggest that TRMM rainfall data are more suitable for monthly streamflow simulation in the study area, and that, when the effects of recalibration and the results for water balance components are also taken into account, the TRMM 3B42-V7 product has the potential to perform well in similar basins.     

How do GPM and TRMM precipitation products perform in alpine regions? A case study in northwestern China’s Qilian Mountains

Journal of Geographical Sciences - Satellite technologies provide valuable areal precipitation datasets in alpine mountains. However, coarse resolution still limits the use of satellite...     

Comparison of TRMM and water district rain rates over New Mexico

This paper compares monthly and seasonal rain rates derived from the Version 5 (V5) and Version 6 (V6) TRMM Precipitation Radar (TPR, TSDIS reference 2A25), TRMM Microwave Imager (TMI, 2A12), TRMM Combined Instrument (TCI, 2B31), TRMM calibrated IR rain estimates (3B42) and TRMM merged gauge and satellite analysis (3B43) algorithms over New Mexico (NM) with rain gauge analyses provided by the New Mexico water districts (WD). The average rain rates over the NM region for 1998–2002 are 0.91mmd?1 for WD and 0.75, 1.38, 1.49, 1.27, and 1.07mmd?1 for V5 3B43, 3B42, TMI, PR and TCA; and 0.74, 1.38, 0.87 and 0.97 mm d?1 for V6 3B43, TMI, TPR and TCA, respectively. Comparison of V5 3B43 with WD rain rates and the daily TRMM mission index (TPR and TMI) suggests that the low bias of V5 3B43 for the wet months (summer to early fall) may be due to the non-inclusion of some rain events in the operational gauge analyses that are used in the production of V5 3B43. Correlation analyses show that the WD rain rates vary in phase, with higher correlation between neighboring WDs. High temporal correlations (>0.8) exist between WD and the combined algorithms (3B42, 3B43 and TCA for both V5 and V6) while satellite instrument algorithms (PR, TMI and TCI) are correlated best among themselves at the monthly scale. Paired t-tests of the monthly time series show that V5 3B42 and TMI are statistically different from the WD rain rates while no significant difference exists between WD and the other products. The agreements between the TRMM satellite and WD gauge estimates are best for the spring and fall and worst for winter and summer. The reduction in V6 TMI (?7.4%) and TPR (?31%) rain rates (compared to V5) results in better agreement between WD estimates and TMI in winter and TPR during summer.     

利用TRMM卫星资料对青藏高原地区强对流天气特征分析

利用热带测雨卫星TRMM(Tropical Rainfall Measure Mission)多种探测结果,结合NCEP再分析资料,研究了发生在青藏高原地区的一次强对流天气特征,综合分析了高原地区对流云特殊的水平、垂直结构特征。结果表明:(1)该强对流降水系统由几个孤立、零散的块状降水云团组成,以深厚弱对流降水为主,微波亮温的低值区也呈孤立、零散的块状分布,并且整个对流系统的云顶高度一致偏高,深厚强对流降水的雨谱主要集中在1～20mm.h-1的范围内,90%以上的深厚弱对流降水样本数和降水量都集中在0～5mm.h-1范围内,在垂直方向上呈被"挤压"状态。除云冰粒子集中在6～18km高度外,可降冰、可降水和云水粒子都集中在低层8km以下,冰雹天气表现为可降冰粒子在低层含量偏高。(2)高原地区强对流天气的特征与其他地方的不同,表现为雨强较小,比平原地区明显偏弱,且对流云降雨样本在不同降雨率范围内分布不均匀,降水云团雨顶高度也远低于平原地区的对流云,地表降水率大值区与微波辐射亮温低值区呈不完全对称分布,潜热释放呈单峰型。(3)高原地区强对流系统发生时,垂直上升运动在400hPa达到最大,水汽主要集中在400hPa高度以下的范围内。     

TRMM和CMORPH卫星资料对三峡库区降水的评估分析

利用1998—2016年TRMM和CMORPH两种遥感卫星资料降水和同期三峡库区气象观测站数据,通过比较干、支流和远、近库区气象站点的降水变化和蓄水前后降水量、雨日、降水强度和频率等变化特征,分析评估了基于两种卫星遥感降水和测站降水的三峡库区局地降水变化。结果表明:库区TRMM和CMORPH卫星降水年际变化特征总体上与气象观测站相符,反演效果在日尺度TRMM略逊于CMORPH,在季尺度CMORPH略逊于TRMM;两种卫星资料对冬季降水反演效果都偏弱。三峡库区干流和支流站点的降水变化总体一致,干、支流各站点降水量均具有较强的年际变化特征。TRMM相比CMORPH更能重现干、支流测站降水的年际变化特征,CMORPH降水年际波动振幅总体上比测站偏大。蓄水前后时段(1998—2003年与2004—2016年)对比,从不同等级降水的强度和雨日、季节降水频率和总量等变化反演效果来看,CMORPH资料分布相比TRMM更接近测站的变化趋势,反演效果略优于TRMM;但两种卫星资料的降水频率和降水量分布与测站的误差在蓄水前后变化都不明显。此外,气象测站、TRMM、CMORPH资料都表现出蓄水后三峡远、近库区年降水量的比值呈平稳波动状态,表明三峡水库蓄水后附近地区降水没有明显变化。     

CMORPH和TRMM 3B42降水估计产品的评估检验

文章利用2007-2010年中国2447个站点的逐小时降水观测数据对同期CMORPH和TRMM 3B42卫星降水估计产品进行了检验和评估。经过对比分析得出:CMORPH和TRMM 3B42卫星降水资料与地面台站资料的日平均降水量空间分布具有较好的相似性,3 h降水量的相关系数在大部分地区分别在0.5和0.4以上;偏差均在±0.25 mm之间,但CMORPH存在显著的南北差异;平均绝对误差、相对误差和均方根误差均存在明显的季节周期性变化;两种卫星资料能够较好反映我国大部分区域夏季降水日变化特征,但在部分地区存在大的偏差;CMORPH和TRMM 3B42总空演率分别为7.23%和2.63%,总漏演率分别为3.25%和5.50%。     

TRMM/PR资料在中尺度模式MM5中的应用

将中尺度非静力模式MM5中的积云参数化Grell方案作了改进，使它含有降水云的云水，即雨水含量，并用该模式对1998年6月29日08时-30日08时(北京时)和1998年7月1日08时-2日08时(北京时)发生在淮河流域的两次特大暴雨进行数值模拟研究，同时通过采用R-qr关系将TRMM／PR(Tropical Rainfall Measuring Mission/Precipitation Radar)得到的降水强度资料月计算出比含水量qr，然后用qv′＝qv qr取代原模式中的比湿qv。结果表明将TRMM／PR资料加入模式后，由于PR雷达具有较高的空间分辨率，能够很好地反映中小尺度系统的空间结构，能够使湿度值接近实际，缩短了中尺度系统的发展时间，使得模拟出来的降雨强度、雨量中心位置以及雨带形状更接近实况。     

基于热带测雨卫星探测的东亚降水云结构特征的研究

利用热带测雨卫星的测雨雷达(TRMMPR)、微波成像仪(TMI)、可见光和红外辐射计(VIRS)、闪电成像仪(LIS)对降水云的综合探测结果,结合全球降水气候计划降水资料(GPCP)和中国气象台站雨量计观测资料,分析了东亚降水分布特点,并比较了TRMMPR与GPCP及地面雨量计观测结果的差异;揭示了中国中东部大陆、东海和南海对流降水和层云降水平均降水廓线的季节变化特征及物理意义,以及TMI高频和低频微波信号对地表降水率变化的响应特点;通过对中尺度强降水系统、锋面气旋降水系统和热对流降水系统的个例分析,探明了降水结构及其与闪电活动的关系、降水云顶部信息与地表雨强之间的关系。