Since the drought characteristics that start and ending time is not easy to be clearly defined, the drought events’ quantitative statistical analysis has been limited. The univariate analysis and meteorological frequency analysis have been tried. For capturing complex interrelations among drought event characteristics, the multivariate drought frequency analysis is an effective tool for quantifying water deficit during drought event. In this study, the duration and severity, i.e., accumulated water deficit of dam inflow for each drought event were used for Copula frequency analysis. The analysis was conducted under various threshold levels of daily dam inflow regimes giving different duration and severity for the same event. The Soyangang Dam (1974-2019) and Chungju Dam (1986-2019) basins were selected for the test area. The trend analysis was conducted using the t-test, Hotelling-Pabst test, and Mann-Kendall test. Before the Copula analysis the univariate frequency analysis was tested for the severity and frequency fitting to GEV model. Then the Copula model was established with the Clayton Copula function and the parameters through the Canonical Maximum Likelihood Method (CML). Based on the derived Copula curves for various return periods, the historical major drought events return periods were evaluated. The return periods of the Copula analysis turn to be higher than the individual univariate analysis and the normal flow threshold gives most feasible return periods considering sensible frequency. The meteorological analysis results show far less frequencies than hydrological analysis.

Stochastically generated streamflow time-series have been employed for diverse water management applications. However, they are often limited to preserving multiscale low-frequency variability in observed streamflow time-series. We present a new, three-stage stochastic model by coupling annual, monthly, and daily simulation models. The proposed model has several components, including (1) a wavelet decomposition coupled to an autoregressive–moving-average (ARMA) model to account for multiscale low‐frequency streamflow oscillations, (2) K‐nearest‐neighbor (KNN) resampling and Z-score based jittering techniques to simulate daily streamflow simulation. The proposed model is applied to the Geum River Basin in South Korea to explore the generation of realistic time series of inflows that exhibit changes to both lower‐order and higher‐order statistics at long‐term (interannual), mid‐term (monthly), and short‐term (daily) timescales. To be specific, our results illustrate that the flexible, continuous simulation approach is substantially valuable for evaluating drought mitigation strategies of water supply infrastructures.

Data quality assurance has been receiving increasing attention in the field of hydrology in the last decade. Only high-quality data ensures data-driven risk analysis and decision-making strategies of hydrology applications. The Central Weather Bureau in Taiwan manages an automated rain gauge network system of hundreds of stations to obtain real-time precipitation observations. Problems arise with such a large volume of data. Occasionally, a station delays returning rainfall observations because of malfunction; moreover, rainfall observations made by one station are markedly higher or lower than neighboring stations, suggesting the existence of anomalies. The anomaly in data should be detected in advance to acquire reliable research results; however, there is a lack of definite criteria for effectively identifying anomalies. In this study, an automated anomaly detection system is developed. The criteria for screening out rainfall anomalies vary according to the four fundamental storm types in Taiwan (frontal rain, Meiyu, convective storms, and typhoons). The K-means clustering analysis is adopted to classify all rain gauge stations of interest by their geographical location and rainfall characteristics. For each cluster, the principal component analysis (PCA) is conducted to derive the first few principal components, aiming to construct an index representing the extent of anomalies in hourly rainfall patterns. Once the criteria are determined, identifying anomalies is straightforward. Ultimately, a dependable anomaly detection system, which is presented as an online interactive web page, was created for identifying possible anomalies to achieve hourly rainfall data quality control.

Hydrological and hydraulic numerical models play crucial roles for water resource management and disaster warning. To properly simulate hydrological characteristics of watershed and river, sufficient monitoring data are essential, especially discharge. In Taiwan, discharges during flood are usually estimated by head-discharge rating curves which are created based on measured water level and discharge data. However understaffed situation, improper equipment and unworkable standard operation procedure make discharge measurement during flood very difficult. Therefore, most of measured discharges are low that cause reliability of head-discharge rating curves low as well. To enhance measurement ability and monitor discharge history during flood, in this study, index surface velocity method is developed based on the flow chart of index velocity method developed by USGS. Acoustic Doppler Current Profiler is used to measure discharges during flood and mean velocities are then calculated. Continuous surface velocities are measured by fixed surface velocity radar and surface velocity data are then treated by filtering and smooth methods to enhance data quality. The Locally Weighted Scatterplot Smoothing is applied in this study. Relationship between surface and mean velocity is therefore built.  The relationship is applied to convert surface velocity into mean velocity and discharge history is then calculated by multiplying corresponding cross sectional area. In this study, the Ximem Bridge in the Yilan River basin, in Taiwan, is a study site to measure discharges and surface velocities by Acoustic Doppler Current Profiler and surface velocity radar, respectively, from 2012 to 2018. Discharge histories of flood events, Typhoon Megi (2016) and Yutu (2018), are estimated using index surface velocity method. The measured, estimated and officially announced discharges are also compared and discussed.

Rainfall infiltration is the primary source of groundwater recharge in Taiwan, and thus identifying the amount of rainfall infiltration is important for groundwater resource management. However, an effective infiltration monitoring system has not been developed yet. In view of this issue, this study proposes a new method to estimate rainfall infiltration based on the temperature profiles in subsurface media (geotemperature profiles). The method is composed of an analytical solution for the geotemperature profiles, a model for estimating the infiltration rate, and a geotemperature monitoring system. The method is verified by a field lysimeter experiment at the National Taiwan University. We installed groundwater monitoring wells in the lysimeter. Thermometers were installed inside the wells and in the subsurface near the wells to measure the geotemperature profiles. Furthermore, tensiometers and water content meters were set up in the field soil to measure the hydraulic pressure and water content, respectively. The infiltration estimation by means of the data obtained from the two instruments could be used to validate the estimation by the proposed method. The study results show that the estimations of the infiltration amount from the geotemperature profiles, tensiometers, and water content meters are consistent. In summary, the proposed method has the potential to quantify the infiltration using geotemperataure profiles and deserves to develop an advanced method to improve the infiltration estimation.

Accurate estimation of snow water equivalent (SWE) has been a pressing issue for managing water supply and forecasting snowmelt-driven floods in a warmer climate. Data assimilation (DA) is a promising approach to improve model estimates of SWE at a large spatial scale by merging remote sensing (RS) observations into model predictions within a unified framework. In densely forested regions, however, the use of most RS techniques (e.g. passive microwave brightness temperature and Lidar) is hampered by the effects (i.e., attenuation and/or scattering of radiation signals) of the forest canopy, resulting in large uncertainties in DA outputs. A well-established, but little-known airborne gamma radiation technique has provided a strong potential to estimate snowpack conditions in forest regions because the gamma technique uses a difference in gamma-ray particles between snow-off and snow-on conditions by limiting canopy effects. In this study, long-term (> 30 years) airborne gamma-based SWE retrievals are assimilated into the Noah land surface model with multiparameterization options (Noah-MP) in mixed forest regions in the northeastern U.S., using the ensemble Kalman filter scheme via the NASA Land Information System. The results emphasize that assimilation of the airborne gamma SWE observations considerably enhanced the model SWE estimates despite the limited number of SWE observations (up to only four observations during the winter period) even in forested regions.

The Global Energy and Water EXchanges Project (GEWEX) of the World Climate Research Programme focuses on in particular process understanding of how water and energy manifest themselves in the earth system. To address the science at a more regional level the Regional Hydroclimate Projects (RHPs) provide a unique opportunity to bring better the scientific community and rally around those issues that are specific and relevant to the region. High mountainous regions of the world provide a unique challenge both in terms of observations as well as modeling. At the same time these regions are of crucial importance to our water resources and the livestream to the major part of the global population. Better understanding of the processes in these regions under climate and environmental change is of the essence to provide sustainable solutions. These mountainous regions act not only as physical barriers but also link communities as they provide the (water) resources they depend upon. At the same time these regions are also more vulnerable and strongly impacted by climate change.  The GEWEX Regional Hydroclimate Projects in the various mountain regions of the world can help bridge the knowledge gap and expertise and as such translate global environmental change to local change and impacts. We focus here on how the Third Pole Enviroment plays a bridging role in both scientific and community aspects.

Downward longwave radiation at the surface (DLR) is a key variable of interest in fields such as hydrology and climate research. However, existing DLR estimation methods and DLR products are still problematic in terms of both accuracy and spatiotemporal resolution, especially on the Tibetan Plateau (TP) with harsh environment. In this paper, we propose a deep convolutional neural network (DCNN) based method to estimate hourly DLR at 5-km spatial resolution from top of atmospheric (TOA) brightness temperature (BT) of the Himawari-8/Advanced Himawari Imager (AHI) thermal channels, combined with near-surface air temperature and dew point temperature of ERA5 and elevation data. Validation results show that the DCNN-based method outperforms popular Random Forest and Multilayer Perceptron based methods, and that our proposed scheme integrating multi-source data outperforms that only using remote sensing TOA observations or surface meteorological data. Compared with state-of-the-art CERES-SYN and ERA5-land DLR products, the estimated DLR by our proposed DCNN-based method has higher spatiotemporal resolution and accuracy, with correlation coefficient of 0.95, root mean square error of 17.2 , and mean bias of -0.8  in the testing period on the Tibetan Plateau. The estimated DLR will provide data support for TP hydrometeorological research community.

Vegetation is always exposed to environmental changes, which are mainly caused by water in arid lands. Water intake by terrestrial plants is mainly carried out by rooting systems; however, in-situ data are generally limited and spatially sparse. We mainly investigated the relationship between water availability and the growth of vegetation in west-central Africa, by using satellite observations including Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) Soil Moisture Content (SMC), i.e. reflecting soil moisture in the near-surface layer, and Gravity Recovery and Climate Experiment (GRACE) Terrestrial Water Storage (TWS), i.e. reflecting terrestrial water change.  In the multiple regression analysis, the relationship between the SMC and Leaf Area Index (LAI) is vital in the region covered by annual plants, utilizing only near-surface soil water because of the shallow rooting system. TWS contains soil water that cannot be utilized by plants because of too deep. Therefore the relationship between LAI and TWS is weak. We found that the decline in TWS during the primary dry season becomes more pronounced toward the equatorial region. Thus, we assessed the relationship between TWS fluctuation and LAI in the primary dry season. We found that this relationship was strong in regions that receive very little rainfall during the considered period, which may be caused by clarifying the difference in vegetation activities by a scant supply of water.  The above findings suggest that it can be possible to investigate the relationship between groundwater and vegetation by focusing on the decline of TWS. Following this, it is necessary to consider the water availability and investigate the relationship in the equatorial region and other regions during the rainy season to grasp the water stress of vegetation due to the effects of cloud cover.

The soil Freeze/Thaw process (F/T) is sensitive to climate change, accurately monitoring the soil F/T has important scientific significance. The brightness temperature data observed by ELBARA-II, as well as soil temperature and near-surface air temperature at the Maqu site of the of the Zoige Plateau Wetland Ecosystem Research Station, Chinese Academy of Sciences, from June 2017 to June 2018 are deployed in this study. Four soil Frost Factors (FF) of which are the Normalized Polarization Ratio, Polarization Difference , the Combined Horizontal Polarization Difference , and the Combined Vertical Polarization Difference were constructed with the ELBARA-II data. The characteristics of the Freeze-Thaw process were evaluated by these four Frost Factors. The results show: There are agreements between FFs and the F/T estimated from two reference data sets (in situ measurements of soil temperature and air temperature). Frost Factor of the Normalized Polarization Ratio and Frost Factor of Polarization Difference  get the highest accuracy of 83.6% and 82.8%  at the incident angle of 50.0°respectively. All Frost Factor have seasonal changes, the four frost factors were higher in spring than in summer, autumn and winter. The standard deviation of the Normalized Relative Frost Factor is the largest in autumn, and the maximum can reach 0.3; it is the smallest in winter, and the minimum is only 0.0. The brightness temperature difference between H-polarization and V-polarization decreases in freeze/thaw transitions, and the difference in the polarization difference during the freezing-thawing transition period is more significant than that during full frozen or complete thawed period. This study finds that the Relative Frost Factor has high accuracy in soil F/T monitoring, and Frost Factors of Polarization Difference has the highest accuracy at 50.0° incident angle, which can reach 83.6%.

To meet the need of real time global soil moisture (SM) data sets, a Soil Moisture Operational Product System (SMOPS) has been developed at National Oceanic and Atmospheric Administration (NOAA) to produce a one stop shop for SM observations. What makes the SMOPS unique is its near real time global blended SM product. Since the first version SMOPS publicly released in 2012, the SMOPS has been updated twice based on the users’ feedbacks through improving retrieval algorithms and including observations from new satellite sensors. The SMOPS V3.0 developed in 2016 and officially released in 2017 merges ASCATA, ASCATB, SMAP, SMOS and AMSR2 soil moisture retrievals. Relative to the older versions, it has better stability in the global domain-averaged spatial coverage. This study reveals SMOPS V3.0 has overwhelming advantages of reduced data uncertainties and increased correlations with respect to the quality controlled in situ measurements. Compared to the 5 individual satellite SM retrievals, SMOPS V3.0 SM product presents a significant advantage in data availability. Significant differences in data availability, climatology and dynamic range of SM values between the bias-corrected SMOPS and individual SM data lead to remarkable distinctions in Noah model SM simulations. Significant improvements of assimilating individual and blended satellite SM retrievals on model SM simulations versus the open loop in both surface and root zone soil layers are evident with reducing the Soil Climate Analysis Network (SCAN) measurements-based root-mean-square errors and raising the correlations with respect to the Enhanced Vegetation Index. Compared to the individual SM assimilations, model SM estimations with benefits of assimilating the SMOPS data provide the more remarkable improvements in surface soil layer. With the higher accuracy, the blended data product from the new version SMOPS is expected to benefit the hydrological, meteorological and climatological researches as well as numerical weather, climate and water prediction operations. 

The Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS), which incorporates satellite imagery and in situ station information, is a new high-resolution long-term precipitation dataset available since 1981. This study aims to understand the performance of the latest version of CHIRPS in depicting the multiple timescale precipitation variation over Taiwan. Analyses focus on examining whether CHIRPS is better than another satellite precipitation product－the Integrated Multi-satellitE Retrievals for Global Precipitation Mission (GPM) final run (hereafter IMERG)－which is known to effectively capture the precipitation variation over Taiwan. We carried out the evaluations made for annual cycle, seasonal cycle, interannual variation, and daily variation during 2001–2019. Our results show that IMERG is slightly better than CHIRPS in most examined features, except that CHIRPS performs better than that of IMERG in representing the (1) magnitude of the annual cycle of precipitation, (2) spatial distribution of the seasonal mean precipitation for all four seasons, (3) quantitative precipitation estimation of the interannual variation of area-averaged winter precipitation in Taiwan, and (4) occurrence frequency of the non-rainy grids in winter. Notably, despite the fact that CHIRPS is not better than IMERG for many examined features, the ability of CHIRPS to depict the temporal phase of precipitation variation over Taiwan for the annual, seasonal, and interannual timescales all pass the 95% significance test. This highlights the potential use of CHIRPS in studying the multiple timescale precipitation variation in Taiwan during the years 1981–2000, for which there are no data available in the IMERG database.

Attaining an adequate water supply considering both increasing human water demand and climate change is one of the challenges faced by many countries. Therefore, assessing the changes in water resource availability and sustainability is crucial for the long-term planning and adaptation options to global climate changes. The objective of this study was to evaluate the spatial and temporal changes in the amount of terrestrial water storage (TWS) in Afghanistan by using remote sensing data, derived from the Gravity Recovery and Climate Experiment (GRACE) satellite between 2003 and 2016 with a spatial resolution of 1° × 1°. Sen’s slope and Modified Mann-Kendall tests have been used for assessing the rate of change and significant trend in TWS respectively. Finally, for assessing the changes in the spatial and temporal distribution of TWS, the concept of reliability, resiliency, and vulnerability (RRV) was used for this purpose. The results show that the spatial distribution of water storage change in Afghanistan has experienced more negative changes in southeast and central regions where most of the population and the temporal pattern in the mean of TWS found more negative changes in the northeast (Himalayan region) and southeast desert during the study period. In general, it is found that the terrestrial water storage of the country is more sustainable in the northeast and southwest and less sustainable in the south and central region.

Most of the types of slope landslides are highly related by soil moisture. Observation of soil moisture by remote sensing approach can provide data with a wide coverage and fine temporal resolution. However, the soil condition of the localized landslide in a fine resolution cannot be detected by the satellite. This research aims to simulate the soil moisture using hydrological model in landslide-prone sub-catchment and to use the simulated runoff as a proxy for downscaling the satellite soil moisture. The study is conducted in upper Brantas River Basin Indonesia, which is vulnerable for flood and landslide. The soil moisture is derived from Soil Water Index from Sentinel product with 11 km resolution which has been corrected from the biases using ground sensors and soil testing. Physically distributed hydrological model is applied with 30 m spatial resolution in a daily scale. The soil water content retrieved from the model with the same resolution can be reproduced dynamically. The downscaling follows Linear Regression, with the regression parameters derived from simulated soil moisture using Linear Least Square method. The verification by comparing the model output with ground sensors observation shows the plausible results. The proposed method integrated with soil stability model will be useful for further application in landslide early warning system.

Actual evapotranspiration (AET) is a significant component of the water cycle on the Tibetan Plateau. Based on the process-based land surface ET/heat fluxes algorithm (P-LSH), the impacts of shortwave radiation and root-zone soil moisture on canopy conductance were considered in this study. Based on data from three FLUXNET towers on the Tibetan Plateau, six key parameters in the P-LSH algorithm were determined by Sobol′ sensitivity analysis. Among them, the parameter, which define the relationship of g₀ versus Normalized Difference Vegetation Index (NDVI), was identified as the most sensitive parameter. Then we utilized the Differential Evolution Markov Chain (DE-MC) method to analyze uncertainty of model parameters. Posterior distributions of parameters were used to correct the response curve of g₀ to NDVI in the P-LSH algorithm and simulate daily AET. The results show that uncertainty of six parameters are greatly reduced after the Markov process. Comparing to the original algorithm, the simulation accuracy of g₀ algorithm by taking into account the incoming shortwave radiation and root-zone soil moisture is greatly improved, specifically reflects in a higher R² (0.85 versus 0.80) and a lower RMSE (14.94 W m^-2 versus 19.07 W m^-2). Finally, although reanalysis-based  VPD was the major source of uncertainty of reanalysis-driven AET estimates, reanalysis-driven AET estimates did not cause excessive errors compared to tower-driven AET estimates (with R² 0.83 versus 0.85, RMSE 16.45 W/m² versus 14.94 W/m²), which provided a reasonable basis for further application of the improved P-LSH algorithm to larger watershed scales.