Runoff Modeling in Ungauged Catchments Using Machine Learning Algorithm-Based Model Parameters Regionalization Methodology

Model parameters estimation is a pivotal issue for runoff modeling in ungauged catchments.The nonlinear relationship between model parameters and catchment descriptors is a major obstacle for parameter regionalization,which is the most widely used approach.Runoff modeling was studied in 38 catchments located in the Yellow–Huai–Hai River Basin(YHHRB).The values of the Nash–Sutcliffe efficiency coefficient(NSE),coefficient of determination(R2),and percent bias(PBIAS)indicated the acceptable performance of the soil and water assessment tool(SWAT)model in the YHHRB.Nine descriptors belonging to the categories of climate,soil,vegetation,and topography were used to express the catchment characteristics related to the hydrological processes.The quantitative relationships between the parameters of the SWAT model and the catchment descriptors were analyzed by six regression-based models,including linear regression(LR)equations,support vector regression(SVR),random forest(RF),k-nearest neighbor(kNN),decision tree(DT),and radial basis function(RBF).Each of the 38 catchments was assumed to be an ungauged catchment in turn.Then,the parameters in each target catchment were estimated by the constructed regression models based on the remaining 37 donor catchments.Furthermore,the similaritybased regionalization scheme was used for comparison with the regression-based approach.The results indicated that the runoff with the highest accuracy was modeled by the SVR-based scheme in ungauged catchments.Compared with the traditional LR-based approach,the accuracy of the runoff modeling in ungauged catchments was improved by the machine learning algorithms because of the outstanding capability to deal with nonlinear relationships.The performances of different approaches were similar in humid regions,while the advantages of the machine learning techniques were more evident in arid regions.When the study area contained nested catchments,the best result was calculated with the similarity-based parameter regionalization scheme because of the high catchment density and short spatial distance.The new findings could improve flood forecasting and water resources planning in regions that lack observed data.

Soil erosion differences in paired grassland and forestland catchments on the Chinese Loess Plateau

In this study,two adjacent gauged catchments on the Chinese Loess Plateau were selected,in which one catchment was afforested and one was restored with natural vegetation in 1954.The distributions of soil erosion rates were estimated between 2010 and 2020 with a high spatial resolution of 2 m in the paired catchments based on the Revised Universal Soil Loss Equation model(RUSLE)and Geographic Information Systems(GIS).The results showed that the simulated soil erosion rates in 2010-2020 averaged 12.58 and 8.56 t ha^(-1)a^(-1)for the grassland and forestland catchment,respectively.Moreover,areas with high soil erosion rates(>80t ha^(-1)a^(-1))were mainly distributed in the topography with steep slope gradients(>45°).Comparisons between simulated soil erosion rates and observed annual sediment loads indicated that the simulation results of the grassland catchment were lower than the observed values,while it was reversed in the forestland catchment.We conclude that the RUSLE model cannot simulate the gravity erosion induced by extreme rainfall events.For the forestland catchment,insufficient streamflow and dense vegetation coverage are crucial factors resulting in hindering the movement of sediments.

Modelling Land Use/Land Cover Change of River Rwizi Catchment, South-Western Uganda Using GIS and Markov Chain Model

Analysis of catchment Land use/Land cover (LULC) change is a vital tool in ensuring sustainable catchment management. The study analyzed land use/land cover changes in the Rwizi catchment, south western Uganda from 1989-2019 and projected the trend by 2040. Landsat images, field observations, key informant interviews and focus group discussions were used to collect data. Changes in cropland, forestland, built up area, grazing land, wetland and open water bodies were analyzed in ArcGIS version 10.2.2 and ERDAS IMAGINE 14 software and a Markov chain model. All the LULC classes increased in area except grazing land. Forest land and builtup area between 2009-2019 increased by 370.03% and 229.53% respectively. Projections revealed an increase in forest land and builtup area by 2030 and only built up area by 2040. LULCC in the catchment results from population pressure, reduced soil fertility and high value of agricultural products.

Application of HEC-HMS for flood forecasting in Misai and Wan'an catchments in China

The hydrologic model HEC-HMS （Hydrologic Engineering Center, Hydrologic Modeling System）, used in combination with the Geospatial Hydrologic Modeling Extension, HEC-GeoHMS, is not a site-specific hydrologic model. Although China has seen the applications of many hydrologic and hydraulic models, HEC-HMS is seldom applied in China, and where it is applied, it is not applied holistically. This paper presents a holistic application of HEC-HMS. Its applicability, capability and suitability for flood forecasting in catchments were examined. The DEMs （digital elevation models） of the study areas were processed using HEC-GeoHMS, an ArcView GIS extension for catchment delineation, terrain pre-processing, and basin processing. The model was calibrated and verified using historical observed data. The determination coefficients and coefficients of agreement for all the flood events were above 0.9, and the relative errors in peak discharges were all within the acceptable range.

Comparison of water quality in two catchments with different forest types in the headwater region of the Hun River, Northeast China

In the headwater catchments of the Hun River,Northeast China, secondary forests(SF) have been replaced by plantations since the 1960 s. Concern has been growing over this loss and the decline in water quality caused by the plantations. To test the effects of plantations on water quality, we selected two separate catchments covered by SF and Pinus koraiensis plantations(KP) to monitor physical and chemical properties of various hydrological variables including throughfall, stemflow,through-litterfall and runoff(flowing out of outlets of the catchments). The physical properties of water declined after water flowed through the two catchments as compared with rainwater. The pH of runoff in both catchments also dramatically decreased. The concentrations of Cl^-, NO_3^- and NH_4^+ in the runoff from the two catchments were similar(concentrations of Cl-and NH_4^+ in both catchments were similar to those in rainwater). Total P concentration in runoff of the SF catchment was higher than that of the KP catchment(P concentrations in both catchments were also higher than in rainwater) because P concentrations in litter and soil of the SF catchment were higher than those in the KP catchment. In summary, the rainwater became acidic in both catchments, but the responses of most water quality variables were similar in the two catchments, suggesting that appropriate ratios of KP in SF are feasible for secondary forest recovery and for preserving water quality(KP did not cause a decline in quality) in the headstream regions in Northeast of China.

Application of GIS-based SCS-CN method in West Bank catchments,Palestine

Among the most basic challenges of hydrology are the prediction and quantification of catchment surface runoff. The runoff curve number （CN） is a key factor in determining runoff in the SCS （Soil Conservation Service） based hydrologic modeling method. The traditional SCS-CN method for calculating the composite curve number is very tedious and consumes a major portion of the hydrologic modeling time. Therefore, geographic information systems （G/S） are now being used in combination with the SCS-CN method. This paper assesses the modeling of flow in West Bank catchments using the GIS-based SCS-CN method. The West Bank, Palestine, is characterized as an arid to semi-arid region with annual rainfall depths ranging between 100 mm in the vicinity of the Jordan River to 700 mm in the mountains extending across the central parts of the region. The estimated composite curve number for the entire West Bank is about 50 assuming dry conditions. This paper clearly demonstrates that the integration of GIS with the SCS-CN method provides a powerful tool for estimating runoff volumes in West Bank catchments, representing arid to semi-arid catchments of Palestine.

Using Artificial Neural Network to Estimate Sediment Load in Ungauged Catchments of the Tonle Sap River Basin, Cambodia

Concern on alteration of sediment natural flow caused by developments of water resources system, has been addressed in many river basins around the world especially in developing and remote regions where sediment data are poorly gauged or ungauged. Since suspended sediment load (SSL) is predominant, the objectives of this research are to: 1) simulate monthly average SSL (SSLm) of four catchments using artificial neural network (ANN);2) assess the application of the calibrated ANN (Cal-ANN) models in three ungauged catchment representatives (UCR) before using them to predict SSLm of three actual ungauged catchments (AUC) in the Tonle Sap River Basin;and 3) estimate annual SSL (SSLA) of each AUC for the case of with and without dam-reservoirs. The model performance for total load (SSLT) prediction was also investigated because it is important for dam-reservoir management. For model simulation, ANN yielded very satisfactory results with determination coefficient (R2) ranging from 0.81 to 0.94 in calibration stage and 0.63 to 0.87 in validation stage. The Cal-ANN models also performed well in UCRs with R2 ranging from 0.59 to 0.64. From the result of this study, one can estimate SSLm and SSLT of ungauged catchments with an accuracy of 0.61 in term of R2 and 34.06% in term of absolute percentage bias, respectively. SSLA of the AUCs was found between 159,281 and 723,580 t/year. In combination with Brune’s method, the impact of dam-reservoirs could reduce SSLA between 47% and 68%. This result is key information for sustainable development of such infrastructures.

Regional Design Storm and Flood Modelling—Risk Implications in Ungauged Catchments

Most planned developments in a catchment for control of excess water using a culvert, bridge or dam spillway are located at a site in a stream where there are no discharge measurements. Even though, for gauged catchments a number of established flood frequency models and rainfall-runoff models do exist, for ungauged catchments mostly regional flood frequency and event-based rainfall-runoff models are used, which depend on regional parameters. In this paper, a regional approach for design floods is presented and risk implication for design of drainage structures assessed. A case study in light of the above has been considered at four ungauged sites in the Limpopo Drainage Basin in north-eastern Botswana.

Hydrological Processes of Storm Runoff from Catchments of Different Land Uses

In this paper, a novel and efficient study on the hydrological processes of storm runoff from catchments of different land uses is conducted. The motivation is to precisely simulate the hydrological processes of storm runoff in the agricultural catchments with different patterns of land uses, i.e., forest, paddy, and upland, respectively. As it is discussed in this paper, different land use leads to different characteristics of storm runoff. In order to understand the changes in the hydrological processes of storm runoff from catchments of different land uses, the effects of rainfall intensity, initial soil moisture deficit, evapotranspiration rate, percolation rate, and retention capacity on hydrological processes of the catchments are taken into consideration. According to the principle of water balance, a general model to connect the separate hydrological processes is developed; then, the individual hydrological process is studied in detail： Firstly, the daily evaporation is calculated according to the relation between the actual evapotranspiration and the potential evapotranspiration rate; Secondly, the retention of storm runoff is plotted against the total rainfall, and the maximum storage is calculated; Thirdly, the percolation rate is calculated for each catchment.

Winter Rain versus Snow in Headwater Catchments: Responses of an Unconfined Pumice Aquifer, South-Central Oregon, USA

Winter precipitation in two headwaters catchments (elevation ~1600 m) in the rain shadow of the Cascades volcanic arc in south-central Oregon normally falls as snow. However, in water year 2015, winter precipitation fell mainly as rain. An eight year study of the unconfined pumice aquifer allowed inter-annual comparison of groundwater recharge during the freshet and discharge during the growing season. During these water years precipitation ranged from 67% (WY2014) to 132% (WY2017) of the 30 year average, and included the rain dominated winter of WY2015 when precipitation during the water year was 98% of the 30 year average. Change in storage in the pumice aquifer was estimated from thickness of the pumice deposit and depth to water table from the ground surface. Measurements were made where 1) the pumice aquifer was exposed at the surface;2) where the aquifer was partially eroded and overlain by either alluvium or lacustrine glassy silt to fine sand;3) fens where the partially eroded aquifer was overlain by peat;and 4) monitoring wells drilled through the pumice aquifer into bedrock. In all settings, groundwater storage in the pumice aquifer following the rain-dominated winter of WY2015 was similar or less than storage following the drought of WY2014 when winter precipitation fell as snow. Storage at the end of WY2014 and WY2015 was the least observed in the eight year study. Winter-time rain during WY2015 produced runoff rather than storage in snow pack. Runoff conveyed from the catchments by flow in stream reaches normally dry from late summer through the winter months. Rain-dominated winter precipitation stresses the perched pumice aquifer. Winter storms starting as rain and turning late to snow and ground-freezing temperatures lead to runoff during the next rain-dominated precipitation event. These patterns produced stream flow in channels that are commonly dry during the winter, reduced near-surface groundwater storage in the pumice aquifer, muted springtime freshet, and stressing of groundwater-dependent ecosystems, forage in meadows, and forest health.

Slope-gully erosion relationship in small catchments of the middle reaches of the Yellow River

The slope-gully erosion relationship in small catchments of the middle reaches of the Yellow River has long been a topic concerned by relevant departments in China Slope-gully relationship in typical small catchment is determined determined on the concept of net increase of sediment yield by using analytical method of sediment formation at different positions in the catchment The result shows that sediments in a small catchment in the middle reaches of the Yellow River mainly come from slopes. ms paper indicated that the sediment sources from slopes are roughly 55, 60, 78 and 85 % of the total sediment yield of a small catchment in Yangdaogou. Wangjiagou. Jiuyuangou and Nanxiaohegou, respectively, due to impacts of varying degress from slope runoff.

Isotope hydrograph separation in alpine catchments:a review

Isotope hydrograph separation （IHS） is a basic tool in applied hydrology. Its application has expanded to surface water and groundwater interaction, and eco-hydrological processes from runoff generation processes. This paper reviews the progress made in IHS for alpine catchments, with emphasis on its significance in reflecting the impact of global change on water resources. Also, the principle of IHS and its uncertainties are explained in detail. The mechanism of variation of stable isotopes in snow-melt water is discussed, and then methods are presented to improve the separation during snow-melt such as volume weighted average method （VWA）, current melt-water method （CMW） and runoff-corrected event water approach （RunCE）, with their advantages and disadvantages explained. New approaches may extend the applications of IllS, for example, large basin studies combined with GIS, and develop new theories of runoff generation combined with other pararneters such as deuterium excess and DOC.

Assessing Water Availability and Unmet Water Demand Using the WEAP Model in the Semi-Arid Bweengwa, Kasaka and Magoye Sub-Catchments of Southern Zambia

Located in the semi-arid zone of Zambia, the Mutama-Bweengwa, Kasaka and Magoye sub-catchments have witnessed a high demand for water due to increase in population and socio-economic activities putting more pressure on water resources. This study assesses the hydrological components and ascertains the available water resources and unmet demand in the sub-catchments using the Water Evaluation And Planning (WEAP) Model and hydrometeorological data collected between 1951 and 2018. The model was calibrated and validated on 1971-1981 and 2008-2018 data respectively. The results reveal that the sub-catchments have transitioned from positive to negative water balance with -164.295 Mm3/year for Mutama-Bweengwa, -19.021 Mm3/year for Kasaka and -86.368 Mm3/year for Magoye. Evaporation was 1815.259 Mm3/year for Mutama-Bweengwa, 1162.655 Mm3/year for Kasaka and 1505.664 Mm3/year for Magoye. The demand for water has been increasing over time for various purposes such as irrigation, domestic, urban/rural water supply and livestock. The overall water storage in the sub-catchments showed a negative water balance for the year 2018. The observed and simulated peak streamflow were 8.16 m3/s and 7.7 m3/s occurring during the month of January and February respectively. The WEAP model performance achieved R2 of 0.98 during calibration and 0.95 for validation, and an NSE of 0.83 for calibration and 0.85 during validation. The values of objective functions show that the hydrology of the Mutama-Bweengwa, Kasaka and Magoye sub-catchments as predicted by the WEAP model provides satisfactory confidence for prediction of future streamflow and hence projection based on future scenarios.

Assessment of debris flow magnitude in small catchments of the lombardy alps: the val gola case study

Debris flows are among the most destructive of all water-related disasters. They mainly affect mountain areas in a wide range of morpho- climatic environments. Therefore, accurate pre-diction of their run out distances, magnitudes and velocities plays a role of paramount impor-tance, in order to plan and design appropriate structural and non-structural defence measures. In this context, a number of Authors have de-veloped methods feasible to evaluate the ten-dency of a catchment to generate debris flow, without giving an estimation of the magnitude. Other empirical procedures are based on the analysis of historical series of debris flow, oc-curred in similar environments, to assess the relationship between the catchment character-istics and the maximum movable debris vol-umes. In this paper, and with reference to Val Gola—a small catchment in the North-East Lom- bardy where debris flows frequently occur—a number of methods, belonging to each of the above mentioned categories, have been briefly reviewed and applied in order to evaluate their effectiveness and consistency.

Hydrological Controls on Nutrient Exportation from Old-Growth Evergreen Rainforests and <i>Eucalyptus nitens</i>Plantation in Headwater Catchments at Southern Chile

Soil cover disturbances have a direct effect on biogeochemistry, potentially enhancing nutrient loss, land degradation and associated changes in ecosystem services and livelihood support. The objective of this study was to assess how canopy affected throughfall chemistry and how hydrology affected stream nutrient load responses in two watersheds dominated by native old-growth evergreen rainforest (NF) and exotic plantation of Eucalyptus nitens (EP), located at the Coastal mountain range of southern Chile (40&deg;S). We measured nitrogen (NO3-N, NH4-N, Organic-N, Total-N) and total phosphorus (Total-P) at catchment discharge, and δ18O in throughfall precipitation and stream discharge in both catchments, in order to separate throughfall (or new water) contributions during storm events. It was hypothesized that all nutrients showed an increase in concentration as discharge increased (or enhanced hydrological access), in EP;but not in NF. Our results indicated that Organic-N, Total-N and Total-P concentrations were positively related to discharge. However, NO3<sup style="margin-left:-7px;">--N showed a negative correlation with catchment discharge. Organic-N and Total-P showed a flush during storm events;the opposite was observed for NO3<sup style="margin-left:-7px;">--N. However, this behavior suggested that NO3<sup style="margin-left:-7px;">--N was being retained by charged particles or soil micro biota, whether Organic-N was flushed as it was more concentrated in big pore water that was not tightly attached, compared with NO3<sup style="margin-left:-7px;">--N.

Large Wood Volume and Longitudinal Distribution in Channel Segments Draining Catchments with Different Land Use, Chile

The storage, longitudinal distribution and recruitment processes of in-stream large wood (LW) were studied comparing channel segments draining four Chilean mountain catchments with different land use. The segments were divided into relatively uniform reaches of different lengths and surveyed for LW (piece dimensions, position in the channel, orientation to flow and aggregation) and stream morphology (slope and bank full channel width and depth) characterizations. LW volume stored in the Pichun, El Toro and Vuelta de Zorra study channels are within the range informed in international researches from streams draining catchments with similar forest covers. However, the 1057 m3/ha of LW stored in Tres Arroyos is extremely high and in the same order of magnitude than the reports from old-growth forests in the Pacific Northwest of USA. The size of the area that can potentially provide wood to streams depends on the wood supply mechanisms within any catchment, and the LW stored in the study segments increases as the size of this area increases. This study aims to contribute to the knowledge on the effects of LW in mountain channels, gathering new information and expanding investigations developed in Chile since 2008. This research was carried out within the framework of Project FONDECYT 11106209.

Temporal Dynamics of Land Use and Water Quality in Three Sub-Catchments of the Rur River, Germany

The Rur catchment has over time undergone land use change which could have affected the biogeochemical processes of the river. Three sub-catchments in the Rur, Upper Rur, Inde and Wurm have different kinds of land use. Upper Rur is more natural catchment;Inde is mixed type and Wurm is highly modified by anthropogenic activities. This study investigated how land use changes from 2000 to 2018 have influenced SO4 and Cl dynamics in the Rur catchment. Land use maps were developed in QGIS environment for land use change calculation. Historical water quality data were collected from the online public source by Ministry of Environment and Nature Conservation in Germany. R-software was used for statistical analysis and graphical presentation. Less land use change was observed in the Upper Rur between 2000 to 2018. But in the Inde and Wurm decrease in agricultural land and associated increase in industrial, commercial and urban land were observed. Increase in mining area inside the catchment has enhanced the level of SO4 and EC in the Inde river. Conversion rates of natural to human dominated land use could be quantified in this study through land use change mapping, which will further help in making water management plan for these and comparable German and European catchments. However, high quality historical data set is a key requirement to maximize the output in process of relating impact of land use change in water quality.

A Method of Catchments Health Assessment under Value-pressure Model and Its Application in Urbanized River Network Area:A Case Study in Shanghai,China

Catchments health assessment is fundamental to effective catchments management. Generally, an assessment method should be selected to reflect both the purpose of assessment and local characteristics. A trial in Shanghai was conducted to test the method for catchments health assessment in urbanized fiver network area. Seven indicators that described four dimensions of river, river network, land use and function, and local feature were used to assess catchments values; while possible change rate of urbanization and industrialization in the next 3 years were chosen for catchments pressure assessment in the value-pressure model. Factors related to catchments classification, indicators measurement and protection priority have been considered in the development strategies for catchments health management. The results showed that value-pressure assessment was applicable in urbanized catchments health management, particularly when both human and catchments had multiple demands. As a result of over 30-year rapid urbanization, more than 70% of Shanghai fiver network area was still in a healthy condition with high catchments values, among them, 39.3% was under high pressure. Poor water quality, simplified river system and weakened local feature of fiver pattern had largely affected catchments health in Shanghai. Lack of long-term monitoring data would seriously restrict the development and validity of catchments health assessment.

Hydrological functioning of forested catchments,Central Himalayan Region,India

Background:Central Himalayan forested catchments provide fresh water supply and innumerable ecosystem services to millions of people.Hence,the understanding of linkages between forests and water is very crucial for availability and quality of water at catchment scale.Therefore,the present study aims to understand the hydrological response of two forested catchments(namely,Arnigad and Bansigad)in the Central Himalayan Region.Methods:Threeyears’data(March,2008 to February,2011)were collected from meteorological and hydrological stations in Arnigad and Bansigad catchments.The present paper describes the mean hydrological response of these forested catchments investigated through detailed field investigation.Results:The annual hyetograph analysis revealed that the rainfall at both the catchments was highly seasonal,and wetperiod(June–September)plays a key role in catchment functioning.Exceedance of rainfall threshold of^200 mm(~10%of annual rainfall)significantly increased streamflow generation in both catchments.In Arnigad,the stream was perennial with a mean baseflow of^83mm per month(~6%of annual baseflow)whereas,Bansigad had greater seasonality due to lack of streamflow during the prewetperiod(March–May).Separation of hydrographs in Arnigad and Bansigad catchments i.e.stormflow(6%and 31%,respectively)and baseflow(50%and 32%,respectively)helped to understand the probability of flooding during wetperiod and drought during dryperiod.The forest ecosystem in Arnigad displayed healthier hydrological functioning in terms of reduced stormflow(82%),and enhanced baseflow(52%),soil moisture(13%),steady infiltration rate(22%)and lag time(~15 min)relative to Bansigad.These enhanced values indicated soil capability to store water in the forested catchment(Arnigad)and helped to understand the volume of water(discharge)that was available during dryperiod.The lower denudation rate at Arnigad by 41%resulted in decreased suspended sediment(18%)and bed load(75%)compared to Bansigad.Further,the enhanced dissolved solids in the Arnigad stream resulted from the higher organic matter generated in the forest floor.Conclusion:This study shows that rainfall during the wetperiod was the main driver of hydrological functioning,whereas,forests provided substantial services by regulating water balance,soil moisture and sediment budget through different mechanisms of forest components at catchmentscale in the Central Himalayan Region.

Assessing discharge periodicity in mountain catchments using classified environmental conditions(Tatra Mountains, Poland)

The periodicity of a river expressed in cycles of various lengths(monthly, seasonal,multiannual) is a result of climatic factors and overlapping environmental conditions within its catchment. In uncontrolled or poorly surveyed catchments, it is very difficult to determine the duration of a stream’s hydrological activity. This is especially relevant for catchments with complicated water circulation in karstic rocks. The present study concerns the small catchment of the Str??yski Potok river located in the area of the Tatra National Park, in the Western Tatras. The observation period covered the 2015 hydrological year, which differed hydrologically from average conditions. This study aims to develop a simple method to explain the processes shaping the mountain stream discharge periodicity. The research employed periodic field observations linked with climatic and non-climatic factors. Environmental conditions were assessed as four classes reflecting their influence on appearance or disappearance of mountain stream water. Class boundaries were the values of quartiles. The degree of correspondence between environmental factors and stream field observations was described via the Index DC(Degree Correspondence Index) approach.Complete correspondence(Index DC =0) was found in 23% catchments, a weak relationship between conditions favouring discharge and actual condition(Index DC=-1, +1) was noted within 11 catchments,while in 9 catchments, no such relationship was found(Index DC =-2, +2). The obtained results indicate a correspondence or lack thereof between the environmental potential of the catchment and its discharge periodicity. The discrepancies between the assessment of the influence of climatic and nonclimatic factors and the data collected during field observations provide a basis for more detailed studies.Continuation of these studies based on the proposed classifications will allow for a more complete explanation of water disappearance in river channels and the determination of their short-and long-term discharge periodicity.