Hydrological Disasters Research Articles

Hydrological changes in the drought‐prone region of Maharashtra (India): Implications for sustainable water use

Recent climatic changes and anthropogenic activities considerably affect regional water resources, particularly in water‐scarce regions. The present study, therefore, aims to understand the changes in monsoon rainfall, potential evapotranspiration (PET), surface runoff, dam storage and groundwater in the drought‐prone region of Maharashtra. The analyses of trend and step‐change were carried out using standard parametric tests. The results were verified with the application of non‐parametric statistical techniques. Notably, monsoon rainfall and runoff were observed to be declining. Exceptionally, the Yerala Basin witnessed significant increase in rainfall along with heavy rainfall events. Contradictorily, Agrani and Man Basins registered a notable decline in rainfall and runoff. As vulnerability to hydrological disasters is high, these basins should be carefully monitored. Although surface water harvesting brought a positive change in post‐monsoon groundwater level, it was the main reason for a decline in runoff volume at downstream reaches, which led to intra‐basin water disparity. The study area witnessed significant increase in groundwater extraction. Unfortunately, the misuse of government‐subsidized schemes exacerbates groundwater exploitation and stockpiling of water resources. Moreover, the increasing rate of PET augments hydrological losses and amplifies water distress. Interestingly, almost all the hydrological changes were notably experienced during the post‐2000 period. To cope with present and future water scarcity problems, it is therefore necessary to restrict irrational use and allocation of available limited water resources. The findings that emerge from this investigation may help to formulate judicious water policies and plans which could be implemented by the government, agronomists, water resource managers and farmers.

Present and future of the South Asian summer monsoon’s rainy season over Northeast India

An iconic feature of the Indian Summer Monsoon Rainfall (ISMR), a longer than June–September rainy season over Northeast India (NEI), while a much shorter one over northwest India is expected to be altered by climate change but an objective definition of the length of the monsoon rainy season (LRS) over the NEI is lacking. Here, defining the LRS objectively over NEI, we show that the El Niño and Southern Oscillation (ENSO) is a primary driver of LRS, while rainfall during LRS is poorly correlated with the ENSO. In contrast to a significant decreasing trend of LRS and LRS-rainfall during the historical period, the projected LRS under the SSP5–8.5 scenario continues to decrease while the LRS-rainfall acquires a significant increasing trend over NEI. A significant increase in the impact of hydrological disasters is expected over NEI in the future due to the increasing intensity and frequency of extreme rain events within a shorter rainy season.

Built environment from the ancient Bali: The Balinese heritage for sustainable water management

Custom and tradition played essential roles in developing the built environment among the Balinese Hindu society for centuries. The wisdom in managing the environment has passed through generations, as demonstrated by some ancient remnants and Old Balinese inscriptions. We observed that glorifying the mountains in this society has long been a part of protecting the hydrologic cycle. This practice even started in the pre-Hindu era, as shown by sacred megalithic features adjacent to the natural reservoir, water spring, and mountainous forest. At some point, the behavior has successfully passed through times and even developed into a more complex socio-cultural system such as subak. Nevertheless, Balinese society is now facing the threat of hydrological disasters, primarily due to a rapid change in land use. Here, we describe the importance of revealing the value behind the ancient Balinese water-management system to gain people's resilience and sustainability. Multidimensional scaling (MDS) was used to measure the sustainability index and the leverage factors affecting water tradition in Bali. We reveal that the inheritance of the Balinese people's water management tradition is generally weak in each dimension. The socio-cultural dimension score is relatively high, with a sustainability score of 71.11%, which signifies that the society still trusts their culture of water resources management to protect nature. However, the economy and ecology dimensions' score low, with sustainability scores of 56.12 and 63.34%, respectively, indicating the need for improvement in economic matters through the policy strategy. Moreover, our study also suggests that the noble value of the cultural heritage related to the water management system needs to be revitalized and disseminated to the public. Hence, the implication is not limited to conserving the natural environment and cultural heritage. It also provides a reference for the current society regarding a built environment in harmony with sustainability principles.

Temporal and spatial changes in hydrological wet extremes of the largest river basin on the Tibetan Plateau

Global warming accelerates the rate of inter-regional hydrological cycles, leading to a significant increase in the frequency and intensity of hydrological wet extremes. The Tibetan Plateau (TP) has been experiencing a rapid warming and wetting trend for decades. This trend is especially strong for the upper Brahmaputra basin (UBB) in the southern TP. The UBB is the largest river on the TP, and these changes are likely to impact the water security of local and downstream inhabitants. This study explores the spatial-temporal variability of wet extremes in the UBB from 1981–2019 using a water- and energy-budget distributed hydrological model (WEB-DHM) to simulate river discharge. The simulated results were validated against observed discharge from the Ministry of Water Resources at a mid-stream location and our observations downstream. The major findings are as follows: (1) the WEB-DHM model adequately describes land-atmosphere interactions (slight underestimation of −0.26 K in simulated annual mean land surface temperature) and can accurately reproduce daily and monthly discharge (Nash-Sutcliffe efficiency is 0.662 and 0.796 respectively for Nuxia station); (2) although extreme discharge generally occurs in July and is concentrated in the southeastern TP, extreme wet events in the UBB are becoming increasingly frequent (after 1998, the number of extreme days per year increased by 13% compared to before) and intense (maximum daily discharge increased with a significant trend of 444 (m3s−1) yr−1), and are occurring across a wider region; (3) Precipitation is more likely to affect the intensity and spatial distribution of wet extremes, while the air temperature is more correlated with the frequency. Our wet extreme analysis in the UBB provides valuable insight into strategies to manage regional water resources and prevent hydrological disasters.

Data-driven multi-step prediction and analysis of monthly rainfall using explainable deep learning

Monthly rainfall prediction is a crucial topic for the management of water resources and prevention of hydrological disasters. To make a multi-step monthly rainfall prediction and discover the primary factors affecting rainfall, this study presents an explainable deep learning approach that integrates four consecutive modules, namely the Gated-Recurrent-Unit-based (GRU-based) encoder module, attention mechanism module, GRU-based decoder module, and expected-gradient-based explanation module, respectively. The first three modules constitute an encoder-decoder with an attention mechanism which could predict monthly rainfall in multiple continuous months in the future, while the explanation module computes attribution values for the input weather and climate features to quantify their importance. A case study is conducted on monthly rainfall data collected from Darwin and Perth, Australia, from Jan 1921 to Dec 2020. The results mainly indicate that: (1) Compared with the baseline methods, the multi-step monthly rainfall predictions made by the encoder-decoder with an attention mechanism show better agreement with the ground truth. (2) For January rainfall in Darwin, the most important weather factors are last January and February rainfall, while the most critical climate index is last February’s Southern Oscillation Index, with its high values inhibit while low values promote January rainfall in Darwin. (3) The most significant feature for June rainfall in Perth is last June’s solar radiation whose feature value is weakly negatively correlated with its attribution value. The study’s significance lies in enhancing the accuracy of multi-step rainfall prediction and providing interpretability through identification of influential factors, which facilitates long-term planning of water resources and deeper understanding of complex weather systems.

Stratospheric Ozone Loss Enhances Summer Precipitation Over the Southern Slope of the Tibetan Plateau

AbstractHeavy summer precipitation over the southern slope of the Tibetan Plateau has dramatic influences on water resources and hydrological disasters in South Asia. It experienced increasing trends over 1979–1996 and decreasing trends over 1996–2022, which are not yet well understood. Here we show observational and numerical evidence that stratospheric ozone has significant impacts on long‐term trends of summer precipitation in this strong convection area. It is found that stratospheric ozone depletion, by modulating the lower stratospheric temperature and upper‐tropospheric static stability, enhances deep convection and precipitation over the southern slope of the Tibetan Plateau. The results suggest that the ozone recovery in the future may reduce the summer precipitation over the southern slope of the Tibetan Plateau in the first half of the 21st century, which would be imperative for future water resource management in South Asia.

Climate Change and Potential Demise of the Indian Deserts

AbstractIn contrast to the “wet gets wetter and dry gets drier” paradigm, here, using observations and climate model simulations, we show that the mean rainfall over the semi‐arid northwest parts of India and Pakistan has increased by 10%–50% during 1901–2015 and is expected to increase by 50%–200% under moderate greenhouse gas (GHG) scenarios. The GHG forcing primarily drives the westward expansion of the Indian summer monsoon rainfall (ISMR) and is facilitated by a westward expansion of the Indian Ocean (IO) warm pool. Mechanistically, the westward expansion of ISMR is a consequence of the episodic genesis over IO and the northward propagation of an expanded Inter‐Tropical Convergence Zone on a sub‐seasonal time scale. While an adaptation strategy to increased hydrological disasters is a must, harnessing the augmented rainfall would lead to a substantial boost in food productivity, bringing transformative changes in the socio‐economic condition of people in the region.

Historical information sheds new light on the intensification of flooding in the Central Mediterranean

Hydrological disasters, such as floods, can have dire consequences for human societies. Historical information plays a key role in detecting whether particular types of hydrological disasters have increased in frequency and/or magnitude and, if so, they are more likely attributable to natural or human-induced climatic and other environmental changes. The identification of regions with similar flood conditions is essential for the analysis of regional flooding regimes. To this end, we here present the longest existing flood reconstruction for the Eastern Liguria Area (ELA) in northwestern Italy, covering 1582 to 2022 CE, which offers a case study representative of the central Mediterranean region. An Annual Flood Intensification Index was developed to transform the historical data into a continuous annual hydrological time-series contained by a homogeneous data structure for the study-area. We found two change-points (trend breaks) in the reconstructed time-series, in 1787 and 1967, with only occasional heavy floods comparable to present-day disasters occurring before the first change-point, and an increasing intensification of floods after the second change-point up to the present day. The recent intensification of flooding in the ELA, associated with changes in land use and land cover, also appears to coincide with phases in which hydrological hazards have become more changeable and extreme in disaster-affected areas. This is evidenced by river basin responses to human-induced disturbances.

Sea Surface Temperature Anomaly Characteristics Affecting Rainfall in Western Java, Indonesia

Western Java is densely populated with high socio-economic activity. Climate-related disasters can be mitigated with the support of an understanding of systems that produce reliable climate predictions. One of the climate variables included in hydrometeorological disasters is rainfall. The characteristics of rainfall in Western Java cannot be separated from the sea surface temperature (SST) around the area. This study compares the relationship between SST and rainfall with singular value decomposition (SVD) and compares it with Pearson's correlation. SVD Model performance was evaluated using square covariance fraction (SCF) and Pearson correlation. The results showed that rainfall has a higher correlation with SST Anomaly (SSTA) by using SVD, with a correlation of about 0.63 in 6 to 9 months without lag time. Rainfall in western Java was closely related to the positive SSTA anomaly in southern Indonesia, especially the waters south of Java Island, and negative anomalies in other areas. Furthermore, atmospheric dynamic analysis showed that the positive coefficient expansion is followed by warmer SST, lower surface air pressure, higher water vapor, and higher rainfall, all were respective to their normal conditions around western Java. This study concludes that warmer SSTA around Western Java causes increased rainfall in western Java than normal and potentially impacts the hydrological disaster in West Java.

The problem of flooding and analysis of the ways of its solution

An analysis of literary sources and informational materials on flood situations that occurred in different historical periods on different continents and in different countries was carried out. It has been established that floods accompany mankind from the earliest times to the present day. Information about catastrophic floods has been preserved since the third millennium BC, in connection with which they should be considered as a historical category. According to the World Meteorological Organization (WMO), floods account for more than 45% of natural disasters related to water (hydrological disasters), which account for about 70% of all natural disasters on the planet. Taking into account the fact that floods are among the most dangerous natural phenomena in terms of the number of victims and the amount of damage, the problem of floods should be considered global. The global nature of the flood problem is confirmed by its following features: the problem concerns all humanity, affects the interests and fate of all countries and peoples; leads to significant social, ecological and economic losses, which, if they increase, will threaten the existence of human civilization; contributes to the aggravation of other internationally recognized global problems, such as environmental, food, social (in terms of population protection), land degradation. Taking into account the fact that in recent years the incidence of catastrophic floods has increased, their social, ecological and economic consequences have increased, as well as the number of human victims from floods, the article focuses on the need to develop a highly effective scientific and methodological toolkit for solving the problem of floods and its wide use not only in Ukraine, but also in many countries of the world. It has been established that among the many areas of systemic research of the problem of floods, the main and most important area is the fundamental improvement of the conceptual foundations of flood protection. The author's version of the universal concept of solving the flood problem is proposed, which is based on a universal algorithm for achieving the target indicator of flood protection. The maximum value of the ratio of the flood protection effect to the costs of achieving this effect was chosen as the target indicator of flood protection. The protective anti-flood effect is represented by the total damage averted by anti-flood measures, which consists of averted social, ecological and economic damages.

Penerapan Metode Principal Component Analysis (PCA) dan Long Short-Term Memory (LSTM) dalam Memprediksi Prediksi Curah Hujan Harian

Since the last three years North Luwu has experienced frequent hydrological disasters in the form of floods and landslides. The disaster had a negative impact on the availability of clean water, failed to plant and even tended to reduce the quality of the harvest. Cocoa is one of the leading commodities of North Luwu Regency whose productivity has decreased due to the impact of climate change so that it will affect the sustainability of the local population's income. Therefore, the purpose of this research is to anticipate rainfall that will occur to prevent or reduce the risk of failure and loss. Principal Component Analysis (PCA) Method is used as feature extraction to find out the most influential variables and the Long Short-Term Memory (LSTM) method is used as a prediction method. Future rainfall is predicted using meteorological variables such as pressure, evaporation, maximum temperature, average humidity, and sunshine duration from 1 January 2017 to 30 September 2022. Based on the PCA results, 4 variables are obtained that have the most influence on rainfall, namely: variable evaporation, maximum temperature, average humidity, and length of sunlight. These variables are used as input to predict rainfall using LSTM. In this study using trial parameters, namely the number of hidden, batch size, and learn rate drop period. The best prediction results were obtained for MAPE of 0.0018 with the number of hidden, batch size and learn rate drop periods of 100, 32, and 50 respectively. The prediction results show very heavy rainfall occurring on August 28, 2021 of 101.9734 mm, 21 September 2021 of 108.6528 mm, and 5 April 2022 of 116.5510 mm. In this study PCA was able to increase accuracy in considering all parameters and choosing the most effective.

Variability and predictability of the Northeast India summer monsoon rainfall

AbstractThe mesoscale orography over one of the wettest regions of the world makes the Northeast India (NEI) vulnerable to hydrological disasters while sustaining a biodiversity “hotspot.” The monsoon rainy season over the NEI is known to be longer than June–September (JJAS), but an objective definition has been lacking. Understanding the drivers and predictability of rainfall variability over the region is key for sustainable development planning and adaptation to increasing disasters in the backdrop of a warming climate but remained poor due to lack of a working definition of “summer monsoon season” over NEI. Here, using the relationship between rainfall over the region and winds at 850 hPa over North Bay of Bengal (BoB), we provide an objective definition of the monsoon season over NEI and argue that the “summer monsoon rainy season” over the NEI (NEIR) is from May to September (MJJAS). In contrast to a significant negative relationship between JJAS Central India (CI) rainfall and a JJAS El Niño–Southern Oscillation (ENSO) index, the MJJAS rainfall over the NEI has no relationship with MJJAS or JJAS ENSO. Instead, we unravel that (a) the tropical Northwest‐Pacific (TNWP) sea surface temperature (SST), (b) the South Equatorial Indian Ocean Dipole (SEIOD) and (c) the Atlantic Zonal mode (AZM) are potential drivers of the NEIR variability. Using a causal inference algorithm, namely the Peter and Clark Momentary Conditional Independence (PCMCI) method, we show that SEIOD directly influences NEIR while TNWP SST has a two‐way connection with the NEIR. While “internal variability” may be higher over the NEI, significant modulation of the variances of the subseasonal fluctuations by predictable drivers like the Atlantic Niño and North Atlantic water temperature provides optimism for seasonal prediction of the NEIR.

Identifying Strategies for Lessening Hydrological Disaster Vulnerability

This qualitative exploratory single case study identified strategies for lessening population vulnerability during a hydrological disaster. The research occurred in a densely populated county in the South Atlantic region of the United States. Semi-structured interviews and thematic data analysis served to address the research question. Methods involved n = 15 semi-structured interviews with uniformed and sworn personnel working in emergency management and public safety roles from agencies and divisions within the county. The data analysis found four generic strategies for lessening population vulnerability to hydrological disasters, (1) resiliency management, (2) resiliency partnership, (3) safety preparedness, and (4) backup preparedness. These four strategies support previous empirical findings and identified best practices. The four generic strategies offer practical approaches that can become a part of a community's overall emergency management plan to assist in lessening the challenges associated with hydrological disaster vulnerability.

Response of Runoff Change to Extreme Climate Evolution in a Typical Watershed of Karst Trough Valley, SW China

Identifying the response of runoff changes to extreme climate evolution was of great scientific significance for the rational regulation of watershed water resources and the prevention of hydrological disasters. However, the time–frequency response relationships were not clear. The Yinjiang River watershed, a typical watershed with karst trough valley areas, was chosen to identify the impact of different climatic driving factors on runoff changes from 1984 to 2015. Continuous wavelet transform (CWT), cross-wavelet transform (XWT), and wavelet coherence transform (WTC) were performed to study the response relationship and time–frequency effect between runoff changes and extreme climate change at different time scales. The main results showed that: (1) Twelve extreme climate indices (ECIs) were detected to have a significant impact on runoff changes, mainly on a 6-year time scale; (2) The R10 and Rx1day in extreme precipitation index and SU34.4 and TNx in the extreme temperature index were the main driving factors of runoff changes, which had relatively large impacts on runoff changes in high and low energy vibration regions. However, the remaining eight ECIs that passed the 0.05 confidence level showed relatively large impacts on runoff changes only in low energy vibration regions; (3) The transition of the interaction between ECIs and runoff changes in high and low time–frequency scales was related to the abrupt change characteristics of the ECIs. The correlation of abrupt change was an important reason for the emergence of highly correlated regions that trigger high and low energy vibrations; (4) As a whole, the extreme precipitation events were ahead of runoff changes at the high time–frequency scale and exhibited small lag effects at the low time–frequency scale, while extreme temperature events were mainly ahead of runoff changes. This study has effectively revealed the impact of climate factors at different scales on runoff changes, and provides a theoretical understanding for regulating and managing water resources in karst basins.

Evaluation of IMERG for GPM satellite-based precipitation products for extreme precipitation indices over Turkiye

As hydrological disasters become more common, the need for monitoring extreme precipitation events is growing globally. Remote sensing technologies developed in recent years help to meet this demand with satellite-based precipitation products, particularly for ungauged basins. However, due to limitations and uncertainties, these products must be validated before being used in hydrometeorological/climatological applications. Validation studies also help improve retrieval algorithms by identifying errors and biases in the data that can be corrected or adjusted. One of NASA's Global Precipitation Measurement products, called Integrated Multi-satellitE Retrievals for GPM (IMERG), has been extensively validated for its ability to detect extreme precipitation events in drastically different climate regions across the world. However, such studies remain limited for Turkiye. In this study, we examined the ability of IMERG Early (E), Late (L), and Final (F) products to detect extreme precipitation in Turkiye using rain gauge data over an eleven-year period (2010−2020). We used extreme precipitation indices recommended by the Expert Team on Climate Change Detection Indices (ETCCDI) to evaluate the ability of IMERG products to detect extreme precipitation. For the stations selected across Turkiye, the Pearson correlation coefficients (R) between daily ground-based precipitation measurements and the three IMERG precipitation products (IMERG-E, IMERG-L, and IMERG-F) were found to be 0.54, 0.55, and 0.64, respectively. IMERG products performed better in detecting the precipitation extreme based on the 85% percentile compared to the 95% and 99% percentiles. IMERG underestimates consecutive dry days by about 25% while overestimating consecutive wet days at the same rate. The performance of IMERG Early, Late, and Final products is inversely proportional to the daily precipitation totals. The annual intensity of precipitation outperformed in the Final run compared to other products across the country. Light and moderate rainfall was detected well, while the detection capability of IMERG decreased for heavy and heavy torrential precipitation. Geographically, the Aegean, Mediterranean, and Southeastern Anatolia regions generally performed better, while Eastern Anatolia and interior parts of the Black Sea region exhibited poorer performance. The low performance in Eastern Anatolia may be attributed to the region's mountainous terrain, while the low performance in the Black Sea region may be due to orographic forcing and land-sea interaction. Overall, the findings of this study showed that the detection capability of IMERG products performed better with increasing latency.

Calculation of indicators of maximum extreme temperature in Sinaloa state, northwestern Mexico

One of the climate problems that causes the most environmental impact worldwide is the trend of increasing occurrence of events of maximum extreme temperature, signaled by indicators such as hot extremes (HE) and maximum maximorum (highest maximum) temperature (MmT). These events can cause conditions ranging from severe droughts to heat stroke, which can cause death in any population. Indicators of maximum extreme temperature in one of the most important agricultural areas in northwestern Mexico were calculated based on significant trends (ST) and adjusted return periods. To calculate the trends of the maximum extreme temperature, frequency (FR), annual average duration (AAD), annual daily duration (ADD), intensity (IN) of HE, and MmT, the Mann-Kendall and Sen’s slope tests were applied to data obtained for 19 weather stations from the CLImate COMputing database for the period 1982–2014. Adjusted return periods (ARP) were calculated for each indicator of maximum extreme temperature by fitting a probability distribution function. For the study area, the ST and maximum extreme temperature shows a prevailing cooling trend. This can be deduced by observing the proportion of negative ST compared with positive ST. The highest positive magnitudes of ST were recorded at stations CUL (FR = 3.44 HE dec-1), GUT (AAD = 6.15 day HE-1 dec-1 and IN = 13.62 °C dec-1), IXP (ADD = 35.00 day dec-1) and POT (MmT = 2.50 °C day-1 dec-1). For ARP, the estimate of the average occurrence frequency of extreme events per100 years are FR = 6.11 HE dec-1 (1 time), AAD = 6.64 day HE-1 dec-1 (4 times), ADD = 38.68 day dec-1 (1 time), IN = 39.09 °C dec-1 (6 times) and MmT = 41.95 °C day-1 dec-1 (1 time). These findings are of key importance for the economic sectors related to agricultural production in the state known, at least to date, as “the breadbasket of Mexico” (Sinaloa). The results will help to develop adaptation/prevention measures before the coming socioeconomic and hydrological disasters.

Towards an Accurate and Reliable Downscaling Scheme for High-Spatial-Resolution Precipitation Data

Accurate high-spatial-resolution precipitation is significantly important in hydrological and meteorological modelling, especially in rain-gauge-sparse areas. Some methods and strategies have been applied for satellite-based precipitation downscaling, residual correction and precipitation calibration. However, which downscaling scheme can provide reliable high-resolution precipitation efficiently remains unanswered. To address this issue, this study aimed to present a framework combining the machine learning downscaling algorithm and post-process procedures. Firstly, four ML-based models, namely support vector regression, random forest, spatial random forest (SRF) and eXtreme gradient boosting (XGBoost), were tested for downscaling and compared with conventional downscaling methods. Then, the effectiveness of the residual correction process using ordinary Kriging and the calibration process using the geographical difference analysis (GDA) method was investigated. The results showed that the ML-based methods had better performance than the conventional regression and interpolation approaches. The SRF and XGBoost outperformed others in generating accurate precipitation estimation with a high resolution. The GDA calibration process significantly improved the downscaled results. However, the residual correction process decreased the downscaling performance of the ML-based models. Combining the SRF or XGBoost downscaling algorithm with the GDA calibration method could be a promising downscaling scheme for precipitation data. The scheme could be used to generate high-resolution precipitation, especially in areas urgently requiring data, which would benefit regional water resource management and hydrological disaster prevention.

Risk Matrix of Hydrological Disasters Combining Rainfall Thresholds and Social-Environmental Criteria

This paper presents a procedure for risk assessment for hydrological disasters considering the threshold rainfall and environmental and social criteria. A case study was carried out to test its feasibility in the northern region of the State of São Paulo, Brazil. The advantage of this procedure is that it only uses data available on government institutions and websites. For this reason, other regions and countries can easily adapt the procedure to their reality. Initially, the hydrological disasters were obtained including date, type of disaster, geographical coordinates and the number of victims. Next, for each disaster, the daily rains corresponding to the dates of the events were obtained from government websites, to establish the rainfall thresholds. Social criteria weighted the poverty index, population density, and the elderly population. The Environmental criteria weighted hydrology, geomorphology and geology factors. An open-source Geographic Information System (GIS) enabled the spatial distribution of disasters through the characterization of the physical environment in hydrology, geomorphology and geology features. The risk assessment was then obtained by combining the rainfall-triggering event with the environmental Susceptibility with social vulnerability. As a result, 31 of the 138 studied municipalities suffered from hydrological disasters, accounting for 99 occurrences between 2002 and 2017.

Emulating Rainfall-Runoff-Inundation Model using Deep Neural Network with Dimensionality Reduction

Abstract Predicting the spatial distribution of maximum inundation depth (depth-MAP) is important for the mitigation of hydrological disasters induced by extreme precipitation. However, physics-based rainfall-runoff-inundation (RRI) models, which are used operationally to predict hydrological disasters in Japan, require massive computational resources for numerical simulations. Here, we aimed at developing a computationally inexpensive deep learning model (Rain2Depth) that emulates an RRI model. Our study focused on the Omono River (Akita Prefecture, Japan) and predicted the depth-MAP from spatial and temporal rainfall data for individual events. Rain2Depth was developed based on a convolutional neural network (CNN), and predicts depth-MAP from 7-day successive hourly rainfall at 13 rain gauge stations in the basin. For training the Rain2Depth, we simulated the depth-MAP by the RRI model forced by 50-ensembles of 30-year data from large-ensemble weather/climate predictions. Instead of using the input and output data directly, we extracted important features from input and output data with two dimensionality reduction techniques (principal component analysis (PCA) and the CNN approach) prior to training the network. This dimensionality reduction aimed to avoid overfitting caused by insufficient training data. The nonlinear CNN approach was superior to the linear PCA for extracting features. Finally, Rain2Depth was architected by connecting the extracted features between input and output data through a neural network. Rain2Depth-based predictions were more accurate than predictions from our previous model (K20), which used ensemble learning of multiple regularized regressions for a specific station. Whereas the K20 can predict maximum inundation depth only at stations, our study achieved depth-MAP prediction by training only the single model Rain2Depth.

An assessment of factors influencing morphometry of the Himalayan rivers Chenab and Teesta

Water resource is rapidly falling short to meet the needs of the population worldwide. The unjust anthropogenic activities have led countries to water stressed/scarce conditions, higher number of hydrological disasters such as floods, droughts and descending levels of groundwater. Being an alpine region, the trans Himalayan region is among the most susceptible regions of the world to the climate change. In this study, the comparative morphometric assessment was made of the eastern and western regions of the alpine zones of the Himalayas. Morphometry of the river basin is a step towards water resource management which provides baseline statistical data about river water. Both river basins were compared using morphometric parameters and further on the basis of 4th order watersheds. The study revealed significant differences in both the basins where the upper Chenab basin has a comparatively higher water discharge, more powerful geological control and lesser basin lag time. The upper Teesta basin demands more management strategies to cope up with higher soil erosion and floods.