Fully distributed rainfall-runoff modeling using spatial-temporal graph neural network

Recent studies using latest deep learning algorithms such as LSTM (Long Short-Term Memory) have shown great promise in time-series modeling. There are many studies focusing on the watershed-scale rainfall-runoff modeling or streamflow forecasting, often considering a single watershed with limited generalization capabilities. To improve the model performance, several studies explored an integrated approach by decomposing a large watershed into multiple sub-watersheds with semi-distributed structure. In this study, we propose an innovative physics-informed fully-distributed rainfall-runoff model, NRM-Graph (Neural Runoff Model-Graph), using Graph Neural Networks (GNN) to make full use of spatial information including the flow direction and geographic data. Specifically, we applied a time-series model on each grid cell for its runoff production. The output of each grid cell is then aggregated by a GNN as the final runoff at the watershed outlet. The case study shows that our GNN based model successfully represents the spatial information in predictions. NRM-Graph network has shown less over-fitting and a significant improvement on the model performance compared to the baselines with spatial information. Our research further confirms the importance of spatially distributed hydrological information in rainfall-runoff modeling using deep learning, and we encourage researchers to incorporate more domain knowledge in modeling.

Formulation of Parsimonious Urban Flash Flood Predictive Model with Inferential Statistics

The curve number (CN) rainfall–runoff model is widely adopted. However, it had been reported to repeatedly fail in consistently predicting runoff results worldwide. Unlike the existing antecedent moisture condition concept, this study preserved its parsimonious model structure for calibration according to different ground saturation conditions under guidance from inferential statistics. The existing CN model was not statistically significant without calibration. The calibrated model did not rely on the return period data and included rainfall depths less than 25.4 mm to formulate statistically significant urban runoff predictive models, and it derived CN directly. Contrarily, the linear regression runoff model and the asymptotic fitting method failed to model hydrological conditions when runoff coefficient was greater than 50%. Although the land-use and land cover remained the same throughout this study, the calculated CN value of this urban watershed increased from 93.35 to 96.50 as the watershed became more saturated. On average, a 3.4% increase in CN value would affect runoff by 44% (178,000 m3). This proves that the CN value cannot be selected according to the land-use and land cover of the watershed only. Urban flash flood modelling should be formulated with rainfall–runoff data pairs with a runoff coefficient > 50%.

Modelling of snowmelt runoff across the Himalayan Region

The Snowmelt-Runoff Model (SRM) was used to evaluate the impact of climate change on hydrological aspects of Lidder River Catchment of the Himalayan Region. It was observed that the river has an average discharge of 1082.49 cusecs. The coefficient of determination (R2) was varies in the range 0.90-0.95 during model validation period (2013-2018).The average coefficient of determination 0.926 and average seasonal volume difference (Dv) was obtained (-) 0.83%. The snow melt runoff harvested water can be used to bring 10 per cent more area under irrigation and water use efficiency which can be increased to an extent of 12-15 per cent for sustainable agriculture production in the Himalayan Region.

Floods and their problems: Land uses and soil types perspectives

The phenomenon of flooding that occurs in almost all regions of the earth causes loss of property and damage to public facilities and causes the loss of many human lives. There are many reports related to the causes of flooding with various solutions offered to overcome the flood problem. However, it seems that these efforts have not been able to eliminate the flood problem. Hydrologists have widely reported various factors that are the cause of flooding with an extensive scope. Therefore, this paper is limited to discussing flooding and its problems, specifically the river flood, from the perspective of land use and soil types. Changes in land use in a watershed can cause an increase in the runoff coefficient. Likewise, different types of soil have different abilities in passing water into the ground. Open land (without land cover) tends to be prone to erosion, reducing the soil’s infiltration capacity and increased surface runoff. Increasing the runoff coefficient will increase the peak discharge in a watershed. The decrease in the river capacity due to sediment can cause a river flood. To support this argument, a rainfall-runoff model, particularly the tank model, is also discussed, taking into account the various uses and types of soil in a watershed. Efforts to anticipate the river flood are also considered for formulating flood disaster control policies in a watershed.

Using NDII patterns to constrain semi-distributed rainfall-runoff models in tropical nested catchments

A parsimonious semi-distributed rainfall-runoff model has been developed for flow prediction. In distribution, attention is paid to both timing of runoff and heterogeneity of moisture storage capacities within sub-catchments. This model is based on the lumped FLEXL model structure, which has proven its value in a wide range of catchments. To test the value of distribution, the gauged Upper Ping catchment in Thailand has been divided into 32 sub-catchments, which can be grouped into 5 gauged sub-catchments where internal performance is evaluated. To test the effect of timing, firstly excess rainfall was calculated for each sub-catchment, using the model structure of FLEXL. The excess rainfall was then routed to its outlet using the lag time from storm to peak flow (TlagF) and the lag time of recharge from the root zone to the groundwater (TlagS), as a function of catchment size. Subsequently, the Muskingum equation was used to route sub-catchment runoff to the downstream sub-catchment, with the delay time parameter of the Muskingum equation being a function of channel length. Other model parameters of this semi-distributed FLEX-SD model were kept the same as in the calibrated FLEXL model of the entire Upper Ping basin, controlled by station P.1 located at the centre of Chiang Mai Province. The outcome of FLEX-SD was compared to: 1) observations at the internal stations; 2) the calibrated FLEXL model; and 3) the semi-distributed URBS model - another established semi-distributed rainfall-runoff model. FLEX-SD showed better or similar performance both during calibration and especially in validation. Subsequently, we tried to distribute the moisture storage capacity by constraining FLEX-SD on patterns of the NDII (normalized difference infrared index). The readily available NDII appears to be a good proxy for moisture stress in the root zone during dry periods. The maximum moisture holding capacity in the root zone is assumed to be a function of the maximum seasonal range of NDII values, and the annual average NDII values to construct 2 alternative models: FLEX-SD-NDIIMax-Min and FLEX-SD-NDIIAvg, respectively. The additional constraint on the moisture holding capacity by the NDII improved both model performance and the realism of the distribution. Distribution of Sumax using annual average NDII values was found to be well correlated with the percentage of evergreen forest in 31 sub-catchments. Spatial average NDII values were proved to be highly corresponded with the root zone soil moisture of the river basin, not only in the dry season but also in the water limited ecosystem. To check how well the model represents root zone soil moisture, the performance of the FLEX-SD-NDII model was compared to time series of the soil wetness index (SWI). The correlation between the root zone storage and the daily SWI appeared to be very good, even better than the correlation with the NDII, because NDII does not provide good estimates during wet periods. The SWI, which is partly model-based, was not used for calibration, but appeared to be an appropriate index for validation.

Multiagent System and Rainfall-Runoff Model in Hydrological Problems: A Systematic Literature Review

Typically, hydrological problems require approaches capable of describing and simulating part of the hydrological system, or the environmental consequences of natural or anthropic actions. Tools such as Multiagent System (MAS) and Rainfall-Runoff Model (RRM) have been used to help researchers to develop and better understand water systems. Thus, this study presents a Systematic Literature Review (SLR) on the joint use of MAS and RRM tools, in the context of hydrological problems. SLR was performed based on a protocol defined from the research question. Initially, 79 papers were found among six bibliographic databases. This total was reduced over four stages of selection, according to exclusion criteria. In the end, three papers were considered satisfactory within the scope of the research, where they were summarized, analyzed, and compared. While the MAS and RRM tools can interact with their results in a coupled model, SLR showed that there are still major challenges to be explored concerning the dynamics between them, as the steps of scales and validation. However, the coupling of MAS and RRM can provide an interesting alternative tool to analyse decision-making about water resources management systems.

Evaluation of water resources changes of the Mountain Dniester in 20th century following the RCP8.5 scenario and based on the “climate-runoff” model

The relevance of the research consists in the need for evaluating the water resources changes of the Dniester due to global warming. The mountain part of the Dniester Basin is a zone of the river's runoff formation that determines its water content. The subject of research includes a process of climate changes and their impact on the water resources of the Mountain Dniester’s catchments. The research focuses on determining the water resources changes under current and possible future climatic conditions represented by climatic scenarios. The research aims at evaluating the water resources changes of the mountain part of the Dniester’s catchment area at the present and in the future by the mid-21 st century (2021-2050) based on the “climate-runoff” model using meteorological observations data (up to 2018 inclusive) and scenario data (averaged data based on 14 mathematical models of the CORDEX project, RCP8.5 scenario). During the research the resources of humidification, heat (heat equivalent) and water content for modern (1989-2018) and scenario (RCP8.5, 2021-2050) climatic conditions based on application of the "climate-runoff" model were evaluated. The theoretical basis for estimating the natural (undisturbed by water management) annual runoff in this model is represented by the water-heat balance equation. The meteorological characteristics (average monthly air temperatures and precipitation) serve as input data. The runoff calculated using the water-heat balance equation is called a climatic runoff. One of the peculiarities of the research consists in the use of the vertical zoning law with respect to distribution of runoff and climatic factors of its formation. During the comparative analysis the dependence of annual runoff norms on height of the Mountain Dniester’s terrain specified in modern regulatory documents served as a basic dependence. Such dependence reflects an altitude-dependant distribution of runoff for the climatic conditions that preceded the significant impact of global warming on air temperature (until 1989). The analysis of the dependences of average long-term values of the annual runoff depending on the terrain altitude showed that the runoff changes for two studied periods (before and after 1989) are within ±12,3%. The analysis of the graphs of chronological course of annual water flow of the mountain tributaries of the Dniester made it possible to confirm the absence of statistically significant trends in their fluctuations. According to the RCP8.5 climate scenario over the period of 2021-2050 and following the results of calculations based on the “climate-runoff” model, the dependences of the average long-term altitude-related values of climatic factors and climatic runoff were retrieved. It was found that the effects of global warming decrease with increasing altitude. In the foothills (up to 200 m) the annual precipitation decreases (up to 11%), the maximum possible evaporation increases (up to 17%) and water resources decrease (up to 46%). Heat resources cease to increase and water resources cease to reduce at the altitudes over 800 m. The average deviation of the scenario and baseline values for precipitation over the estimated period will amount to 2.41% for precipitation, 5.79% for maximum possible evaporation and 8.87% for water resources. Thus, reduction of water resources in the mountainous part of the Dniester by the mid-21 st century will be insignificant. When evaluating the current state of water resources of the Mountain Dniester no significant changes were discovered, thereby not contradicting the other authors’ data.

Understanding the effect of changing land use on floods and soil erosion in the Cagayan de Oro catchment

<p>The destructive capability of typhoons affects lives and infrastructure around the world. Spatial analysis of historical typhoon records reveal an area of intense storm activity within the Southeast Asian (SEA) region. Within SEA is the Philippines, an archipelagic tropical country regularly struck by storms that often cause severe landslides, erosion and floods. Annually, ˜20 cyclones enter the Philippine Area of Responsibility, with about nine making landfall, causing high winds and intense rainfall. Thus, significant research in the Philippines has focused on increasing the resilience of ecosystems and communities through real-time disaster forecasting, structural protections, and programmes for sustainable watershed management (e.g. rehabilitation and conservation agriculture). This dissertation focused on the third aspect through computer modelling and scenario analysis. The study area is the Cagayan de Oro (CDO) catchment (˜1400km²) located in the Southern Philippines. The catchment experienced heavy flooding in 2012 from Typhoon Bopha and has major erosion problems due to mountainous slopes and heavy rainfall. Communities derive ecosystem services (ES) including agricultural production, water supply, recreation, mining resources, flood mitigation, etc. Since changes to the supply or distribution of these ES affects livelihoods and the hydrological response of the catchment to typhoon events, this research used the Land Utilisation and Capability Indicator (LUCI) model to understand the baseline ES and potential changes associated with basin management plans. This was the first detailed tropical application of LUCI, including parameterising it for Philippine soil and land cover datasets in CDO and extending its capability to be applied in future tropical areas. Aside from applying LUCI in a new geoclimatic region, this research contributed to the general development of LUCI through testing and improving its sediment delivery and inundation modelling. The sediment delivery was enhanced using the Revised Universal Soil Loss Equation (RUSLE) model that allows LUCI for the first time to account for impacts of specific land management such as agroforestry and contour cropping on erosion and sediment delivery. Progress was made in updating a flatwater inundation model for use with LUCI, including converting it to Python but this requires further development and testing before it is suitable for application in the Philippines. The development and rehabilitation scenarios showed improved flood mitigation, lower surficial soil erosion rates, and lower loads of nutrients compared to the baseline scenario. Additionally, ES mapping under different land cover scenarios has not been previously accomplished in CDO, and this research provides useful information to guide local decision-making and management planning. The rainfall-runoff model of LUCI was tested against the Hydrologic Engineering Center’s Hydrological Modelling System (HEC-HMS), showing good agreement with observed flow. Since the rainfall-runoff model of LUCI has been minimally utilised in past applications, this CDO application elucidated directions for future work around further testing under extreme rainfall events and climate change. Overall, this novel application of LUCI creates a framework to assist decision-making around land cover changes in the CDO, provides guidance around data requirements and parameterisation procedures to guide future international applications, and has significantly contributed to development and improvement of the LUCI framework to extend its modelling capabilities in the future.</p>

Understanding the effect of changing land use on floods and soil erosion in the Cagayan de Oro catchment

<p>The destructive capability of typhoons affects lives and infrastructure around the world. Spatial analysis of historical typhoon records reveal an area of intense storm activity within the Southeast Asian (SEA) region. Within SEA is the Philippines, an archipelagic tropical country regularly struck by storms that often cause severe landslides, erosion and floods. Annually, ˜20 cyclones enter the Philippine Area of Responsibility, with about nine making landfall, causing high winds and intense rainfall. Thus, significant research in the Philippines has focused on increasing the resilience of ecosystems and communities through real-time disaster forecasting, structural protections, and programmes for sustainable watershed management (e.g. rehabilitation and conservation agriculture). This dissertation focused on the third aspect through computer modelling and scenario analysis. The study area is the Cagayan de Oro (CDO) catchment (˜1400km²) located in the Southern Philippines. The catchment experienced heavy flooding in 2012 from Typhoon Bopha and has major erosion problems due to mountainous slopes and heavy rainfall. Communities derive ecosystem services (ES) including agricultural production, water supply, recreation, mining resources, flood mitigation, etc. Since changes to the supply or distribution of these ES affects livelihoods and the hydrological response of the catchment to typhoon events, this research used the Land Utilisation and Capability Indicator (LUCI) model to understand the baseline ES and potential changes associated with basin management plans. This was the first detailed tropical application of LUCI, including parameterising it for Philippine soil and land cover datasets in CDO and extending its capability to be applied in future tropical areas. Aside from applying LUCI in a new geoclimatic region, this research contributed to the general development of LUCI through testing and improving its sediment delivery and inundation modelling. The sediment delivery was enhanced using the Revised Universal Soil Loss Equation (RUSLE) model that allows LUCI for the first time to account for impacts of specific land management such as agroforestry and contour cropping on erosion and sediment delivery. Progress was made in updating a flatwater inundation model for use with LUCI, including converting it to Python but this requires further development and testing before it is suitable for application in the Philippines. The development and rehabilitation scenarios showed improved flood mitigation, lower surficial soil erosion rates, and lower loads of nutrients compared to the baseline scenario. Additionally, ES mapping under different land cover scenarios has not been previously accomplished in CDO, and this research provides useful information to guide local decision-making and management planning. The rainfall-runoff model of LUCI was tested against the Hydrologic Engineering Center’s Hydrological Modelling System (HEC-HMS), showing good agreement with observed flow. Since the rainfall-runoff model of LUCI has been minimally utilised in past applications, this CDO application elucidated directions for future work around further testing under extreme rainfall events and climate change. Overall, this novel application of LUCI creates a framework to assist decision-making around land cover changes in the CDO, provides guidance around data requirements and parameterisation procedures to guide future international applications, and has significantly contributed to development and improvement of the LUCI framework to extend its modelling capabilities in the future.</p>