Estimation Of Pollutant Loads Research Articles

Assessment of Agricultural Non-point Source Pollution Loads Applying Improved Export Coefficient Model (IECM) : A Case Study in Taihu Lake Basin

Accurate and comprehensive estimation of agricultural non-point source pollution (ANPSP) is of vital importance in watershed pollution control, while data inaccessibility and complexity of process-based models may be main restrictions in practice. An estimation methodology framework for ANPSP was established based on improved export coefficient model (IECM) in this research. A series of updated, regional coefficients were collected and/or calculated with multiple pollution sources considered (i.e., animal husbandry, farming and forestry, fishery, and rural domestic sewage) to provide a comprehensive estimation of TN and TP pollution loads in Taihu Lake Basin. According to this model, pollution loads of TN and TP in Taihu Lake Basin were 65791.28 and 11400.38 t in 2016, respectively. Animal husbandry was main source of ANPSP, accounting for 31.05% and 66.70% for TN and TP, respectively, which requires more efficient and clean agricultural production modes and strict pollution control measures. IECM-based estimation framework may provide a reference for ANPSP management practice in large scale watershed.

Pollution load estimation and influencing factor analysis in the Tuhai River Basin in Shandong Province of China based on improved output coefficient method.

Estimating the pollution loads in the Tuhai River is essential for developing a water quality standard scheme. This study utilized the improved output coefficient method to estimate the total pollution loads in the river basin while analyzing the influencing factors based on the STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) model. Findings indicated that the projected point source pollution loads for total phosphorus (TP), chemical oxygen demand (COD), and ammonia nitrogen (AN) would amount to 3937.22 ton, 335,523.25 ton, and 13,946.92 ton in 2021, respectively. Among these, COD pollution would pose the greatest concern. The primary contributors to the pollution loads were rural scattered life, large-scale livestock and poultry breeding, and surface runoff. Per capita GDP emerged as the most influential factor affecting the pollution loads, followed by cultivated land area, while the urbanization rate demonstrated the least impact.

Pollutant load estimation and load reduction target (LRT) projection for total maximum daily load (TMD) allocation on tropical rivers

Pollutant load may be defined as the mass of a substance that passes a particular point of a river in a specified amount of time. Meanwhile, estimation of pollutant loading and identification of their sources is crucial to environmental management and planning. For the first time (in this study), Flow rate measurement was used to estimate daily pollutant loading from Intermittent water quality concentration data, using a 2-dymensional Water Quality Analyser (WQA). Subsequently, Total Maximum Daily Load (TMDL) was determined using the Load duration approach, while Load Reduction Targets were projected for the future, the using regression option of trend analysis available in the WQA. Out of the ten parameters used for the study, BOD, NH3, and TSS have been identified as the most critical pollutants in the area, which require average load reduction of 3898.88 kg-day, 1053.28 kg-day, and 444,716.50 kg-day respectively, to achieve water quality class II, until 2030. Moreover, the study reveals that the load reduction target for BOD and TSS would decrease in the future, while that of NH3 increases (p < 0.001). This is even as significant variability also exists for the projected load reduction target over the months throughout the projected period (p < 0.01). It was concluded that WQA provides a cost and time effective, and a reliable means for estimation of Pollutant load and projection of Load Reduction Target. The study recommends source identification for the critical pollutants into the river and allocation of TMDLs using the dynamic flow approach.

Estimation of Pollutant Load in Typical Drainage Ditches of Ningxia Yellow River Diversion Irrigation Area Based on LOADEST Statistical Model

To comprehensively comprehend the spatiotemporal variations in pollution load within the Sixth Drainage Ditch of the Ningxia Yellow River Diversion Irrigation Area, we employed the LOADEST model. We utilized daily flow data and concentrations of ammonia nitrogen (NH3-N), nitrate nitrogen (NO2-N), total nitrogen (TN), and total phosphorus (TP) to construct regression equations for the pollutant load at four distinct monitoring sections of the Sixth Drainage Ditch. The results unveiled an impressive range of correlation coefficients (R2) for the pollution load regression equations at the four monitoring sections, ranging from 72.42% to 94.4%. This indicates a strong fit for the pollution load regression equations, rendering them suitable for estimating the pollution load of the Sixth Drainage Ditch. Furthermore, the changing patterns of various pollutants in the same monitoring section exhibit a remarkable level of consistency. In each case, they initially experience an upward trajectory followed by a subsequent decrease. Notably, the total nitrogen (TN) load in the drainage area exceeds that of the total phosphorus (TP). The spatial distribution patterns of the total nitrogen (TN) and total phosphorus (TP) load within the Sixth Drainage Ditch exhibit a progressive increase from the upstream to downstream areas. Meanwhile, the spatial distribution characteristics of ammonia nitrogen (NH3-N) and nitrate nitrogen (NO2-N) follow a similar pattern of an initial increase followed by a decrease.

A Process‐Based Model to Track Water Pollutant Generation at High Resolution and Its Pathway to Discharge

AbstractThe estimation of water pollutant loads is of crucial importance as it decides data inputs for both subsequent water quality models and watershed pollutant mitigation strategy. However, the generation of water pollutant loads at high resolution and the life‐cycle pathway of them from sources, through pipelines and wastewater treatment plants, finally to recipient water bodies remain unclear. This study aims to establish a process‐based Water Pollutant Loads Tracking model and applies it to a rapidly urbanizing watershed in Taihu Lake Basin, China at the resolution of 5 m × 5 m in 2017, whose terrain is flat, slowly lowing from west to east. Results show that, of 261.55 tons of the total phosphorus (TP) generation, 71.28% is collected, and 64.32% is treated, with only 26.46% discharged to the water bodies and 1.25% to the target ones. The spatial hotspots of TP generation are mainly concentrated in residential and industrial areas. Direct discharges from point sources and nonpoint runoffs, especially in rainy seasons, are recognized as the main challenges for local water pollution control. The main innovation of this study is to quantify the generation of water pollutant loads at high resolution and to trace the subsequent pathway to collection and discharge, and then to identify the potential substantial gap between generation and discharge, demonstrating the efficacy and broader applicability of this model.

Distribution Characteristics of Carbon, Nitrogen, and Phosphorus and Pollution Load Estimation of Sediments in Danjiangkou Reservoir, China

Danjiangkou Reservoir is a world-famous large artificial freshwater lake that offers water resources for the middle route of the South-to-North Water Diversion Project in China. In this study, the distribution of carbon, nitrogen, and phosphorus in reservoir sediments and their pollution assessments were elucidated at different water periods. The average TN (total nitrogen), TP (total phosphorus), and TOM (total organic matter) contents were 794.8 mg/kg, 807.2 mg/kg, and 8.7% in the nonflood season, respectively. When the time comes to flood season, with the large amount of nitrogen pollution inputted from peripheral nonpoint sources and phosphorus released by the accelerated exchange of water bodies, the average TN concentration increased to 1061.2 mg/kg. In addition, the average TP and TOM contents were reduced to 559.5 mg/kg and 6.3%. Nutrient pollution fluctuated between the safe and lowest level. Reservoir eutrophication risk was low. There was a certain nitrogen enrichment in the Danjiangkou Reservoir, and the flood season was the main period of nitrogen pollution input. Water exchange during flood season might accelerate organic matter degradation. Near the dam, sediment organic matter content increased significantly, reaching severe pollution levels. The results of the simulated sediment nutrient release test showed that the nitrogen and phosphorus in the reservoir would release slowly. Moreover, their annual release flux was calculated as 470.4 t and 87.9 t, respectively. It illustrated that the internal pollution of Danjiangkou Reservoir was light, and the release amount was small, so it was not the main pollution source of the reservoir at present.

Characteristics of water quality and estimation of pollution loads in the main rivers surrounding Erhai Lake

Characteristics of water quality and estimation of pollution loads in the main rivers surrounding Erhai Lake

Study on total phosphorus pollution load estimation and prevention and control countermeasures in Dongting Lake

In order to fight the battle for reducing total phosphorus (TP) in Dongting Lake and continuously improve the water ecological environment of the Yangtze River, this study accounts for the TP pollution load in Dongting Lake. To this end, the sources of the pollutants are analyzed and corresponding prevention and control measures are proposed by comprehensive use of field research and historical data analysis. The obtained results reveal that the annual TP input to the Dongting Lake was 23287 t in 2017, among which the input from four rivers and three lake entrances represent the main source with an annual TP input of 14033 t (including 3681 t from sediment input), accounting for 60.26%. Agricultural pollution is the second largest source of TP input, accounting for 32.69%, while domestic pollution only accounts for 5.93%. Additionally, the contributions of atmospheric deposition and industrial pollution in order are 0.69% and 0.4%. The TP input from moving sources such as birds and waterway shipping is very small, only accounting for 0.03%. Based on the achieved results, this study aims to propose corresponding prevention and control measures in four aspects: interception of river and lake pollution, control of agricultural pollution sources, enhancement of sewage treatment and industrial pollution prevention.

A method for researching the eutrophication and N/P loads of plateau lakes: Lugu Lake as a case

Lugu Lake is one of the best plateau lakes in China in terms of water quality, but in recent years the eutrophication of Lugu Lake has accelerated due to high nitrogen and phosphorus loads. This study aimed to determine the eutrophication state of Lugu Lake. Specifically, the spatio-temporal variations of nitrogen and phosphorus pollution during the wet and dry seasons were investigated in Lianghai and Caohai, and the primary environmental effect factors were defined. Adopting the endogenous static release experiments and the exogenous improved export coefficient model, a novel approach (a combination of internal and external sources) was developed for the estimation of nitrogen and phosphorus pollution loads in Lugu Lake. It was indicated that the order of nitrogen and phosphorus pollution in Lugu Lake was Caohai > Lianghai and dry season > wet season. Dissolved oxygen (DO) and chemical oxygen demand (CODMn) were the main environmental factors causing nitrogen and phosphorus pollution. Endogenous nitrogen and phosphorus release rates in Lugu Lake were 668.7 and 42.0 t/a, respectively, and exogenous nitrogen and phosphorus input rates were 372.7 and 30.8 t/a, respectively. The contributions of pollution sources, in descending order, were sediment > land-use categories > residents and livestock breeding > plant decay, of which sediment nitrogen and phosphorus loads accounted for 64.3 % and 57.4 %, respectively. Regulating the endogenous release of sediment and obstructing the exogenous input from shrubland and woodland are emphasized for the management of nitrogen and phosphorus contamination in Lugu Lake. Thus, this study can serve as a theoretical foundation and technical guide for eutrophication control in plateau lakes.

Contribution Analysis of External Source Pollution Load and Environmental Capacity Estimation of Reservoirs in Central Yunnan Plateau

To explore the source of the pollution load and its contribution rate in the upper reaches of the plateau reservoir and to analyze the water environment capacity of the reservoir, we selected the Chaishitan Reservoir in the Yunnan Plateau as the research object, applied the pollutant discharge coefficient method to estimate the source of external pollution in the upstream basin of the reservoir, used the simultaneous monitoring data of hydrology and water quality to calculate pollution load into the reservoir, and used the eutrophication model to calculate the maximum capacity of TN and TP in the reservoir under different water quality target scenarios. The results showed that:① the main characteristic pollutants in Chaishitan Reservoir and the above basin were TN and TP. ② COD and TP in the upper reaches of the reservoir mainly came from rural non-point source pollution, with contribution rates of 49.40% and 50.11%, respectively; NH4+-N and TN mainly came from urban domestic pollution sources, with contribution rates of 45.76% and 33.77%, respectively. Among the contributions of rural non-point source pollution, the contribution rates of COD and TP in Luliang District were 34.82% and 36.82%, respectively. The contributions of COD, NH4+-N, TN, and TP to urban domestic pollution were the highest in Qilin District, all of which were up to 65%. ③ The inflows of COD, NH4+-N, TN, and TP were 28050.90, 2465.16, 4680.54, and 870.93 t·a-1, respectively. The inflow of TN and TP pollution load was 4637.80 t·a-1 and 125.04 t·a-1, respectively. ④ When the target of water quality was Class Ⅲ, and the requirements of the Water Function Zoning of Yunnan province were met, the environmental capacities of TN and TP were 1102.62 t·a-1 and 54.85 t·a-1, respectively. Rural non-point source pollution and urban domestic pollution sources were the main sources of pollution in the upper reaches of Chaishitan Reservoir, which were priority control sources. These research results can provide a scientific theoretical basis for pollution source treatment in the plateau reservoir basin.

Spatial-temporal characteristics of corrected total phosphorus pollution loads from agricultural non-point sources in Tuojiang River watershed, Sichuan Province of southwestern China.

Traditional method of estimating pollution loads may neglect the internal spatial heterogeneity of socio-economic driving factors, which can result in overestimate and underestimate of pollution loads. In this study, the corrected approach to estimating total phosphorus (TP) pollution load was proposed to explore its future variation to develop effective phosphorus pollution control strategies for water environment management. As the first-class tributary of the Yangtze River, the TP out of limits in the Tuojiang River is serious. Thus, based on the presently related basic datasets related to TP pollution load estimation, we firstly adopted the GM (1,1) model to predict their varied trends from 2021 to 2025. We then used the pollution emission coefficient method to calculate the TP pollution load. Moreover, considering the temporal and spatial heterogeneity of the pollutant generation coefficient, we further introduced population and GDP factors to further modify the pollutant generation coefficient to correct TP pollution load. Finally, we employed the exploratory spatial data analysis (ESDA) method to explore spatial distribution characteristics and spatial autocorrelation of TP pollution load from diverse pollution sources in 2025. The results showed that the total TP pollution load from diverse pollution sources will increase from 12,194.92 t in 2021 to 12,461.26 t in 2025, an increase of 2.18%. More concretely, the TP pollution load from rural domestic sewage, rural domestic waste and livestock, and poultry pollution sources will separately decrease by 94.24 t, 77.9 t, and 86.52 t. However, the TP pollution load from agricultural runoff and agricultural solid wastes pollution sources will increase by 74.52 t and 451.49 t, respectively. The contribution of TP pollution load from diverse pollution sources to total TP pollution load will be as follows: livestock and poultry (63.49%) > agricultural solid wastes (16.72%) > agricultural runoff (12.26%) > rural domestic sewage (4.12%) > rural domestic waste (3.41%). The difference in the spatial distribution of TP pollution load from diverse pollution sources in 2025 will be prominent. TP pollution from rural domestic sewage and rural domestic waste pollution sources is more serious in the Xindu and Longquanyi districts, and that from agricultural runoff and agricultural solid wastes pollution sources is more prominent in the midstream and downstream. TP pollution load from livestock and poultry pollution source is higher in the Renshou, Anyue, Rongxian, Luxian counties, and Jiangyang district. Additionally, TP pollution load from rural domestic sewage, rural domestic waste, agricultural runoff, and agricultural solid wastes pollution sources in 2025 will show a clear spatial correlation, but the spatial correlation of TP pollution load from livestock and poultry pollution source will be weak. The study is effective to eliminate the influence of temporal and spatial variation of pollutants generates coefficients on TP pollution load estimation. The method can reflect the actual condition of pollution loads in watersheds more objectively, which can be applied to estimate other pollution loads of similar watersheds with intensive socio-economic activities. The findings in this study can provide a critical reference for the stakeholders to balance water environment conservation and socio-economic development.

Nitrogen and phosphorus load estimation of inflow rivers to Yuqiao Reservoir based on GWLF model

Watershed model technology was an important tool for non-point source pollution prevention and control. In order to figure out the total nitrogen (TN) and total phosphorus (TP) load of inflow rivers to Yuqiao Reservoir in 2021, GWLF model was used to simulate the hydrologic process in its sub-basin. Firstly, the hydrological parameters of the model were determined by the measured meteorological and hydrological data. Secondly, the hydrological parameters were extended to the whole basin, and the three main river inflows to Yuqiao Reservoir were simulated in 2021. Lastly, the nutrient loads were estimated by multiplying the simulated inflow as mentioned above by the observed TN and TP concentration data at the same site. The results showed that the TN and TP load of inflow rivers was 7 106.9 t and 80.0 t, respectively, of which the flood period from July to August contributed 55.3% and 69.3%, respectively. GWLF model was effective in simulating hydrological processes in the study area. The Nash efficiency coefficients of the calibration and the validation period were 0.89 and 0.91, respectively. GWLF model was suitable for hydrological simulation and estimation of non-point source pollution load in small and medium-sized basins.

Pollution Load Estimation and Control Countermeasures of Zhangze Reservoir

Raising the water pollution control countermeasures on the basis of rational pollution load estimation is significant for improving water quality. Zhangze Reservoir, the largest water body in Changzhi city, China, was selected for this study. Considering the information constraints of the reservoir basin, the pollution load estimation method system covering point sources, non-point sources, and internal sources is systematically constructed using an export coefficient model, an sediment pore water diffusion model, and other methods, with the aim of estimating the discharge of pollutants into the Zhangze Reservoir from domestic sources, industrial sources, agricultural sources, sediment release, and urban runoff. The findings indicate the following: 1) the pollution loads of COD, NH3-N, TN, and TP are 3,157.82t, 177.62t, 760.72t, and 42.29t, respectively; 2) in terms of the distribution of pollution sources, urban domestic sources top the rest, accounting for 65.47% of COD, 82.83% of NH3-N, 79.00% of TN, and 60.65% of TP, followed by the agricultural source; 3) the total discharge of the main water pollutants exceeds the water environmental capacity, characterized as the dominating existence of domestic point source, the coexistence of point sources and non-point sources, and the coexistence of exogenous sources and endogenous sources. Finally, control countermeasures are raised to minimize the total discharge of pollutants for improving the water quality.

An assessment of 2 watershed models to meet watershed planning needs

Hanson R, Hammond Rowan J, Green M. 2022. As assessment of 2 watershed models to meet watershed planning needs. Lake Reserv Manage. 38:16–27. Watershed models provide a quantitative approach for estimating loads, providing source load estimates, and evaluating various management alternatives to determine the best and most efficient actions to reduce pollutant loads. For watershed plans to be approved by the US Environmental Protection Agency (EPA) and state agencies and to obtain Clean Water Act Section 319 funding, a modeling component is required. In this study, we tested 2 watershed quality models, STEPL and SWAT, in a New Hampshire watershed to determine, when there are limited data available, whether the investment of resources required for a more complex model produces a more accurate understanding of phosphorus loads for the purpose of watershed planning. We found that while a simple model such as STEPL does not provide understanding of the complexities and unique nature of different watersheds, it easily provides quantitative estimates of long-term mean pollutant loads that, based on available data and state guidance, are adequate for watershed planning purposes. A more sophisticated model, such as SWAT, can provide more insight into watershed dynamics, but to be used to its full potential it requires far more data and expertise than may exist in watershed groups and municipalities involved with applied watershed management planning. We recommend that watershed plans include goals for implementing studies that produce rich datasets and more sophisticated modeling as a long-term watershed management objective to improve understanding about watershed dynamics and predict how management actions will affect pollutant loads to achieve overall water quality improvement goals.

Review on agricultural non-point source pollution monitoring sections layout and pollutant loading estimation in small watershed

流域尺度面源污染的监测是系统认识农业面源污染的发生、迁移及转化过程, 并对其进行有效控制的重要基础. 当前, 田块尺度的面源污染监测方法比较成熟, 而流域尺度的监测, 尤其是监测断面的布设及采样频率设置等方面的研究较少. 本文详细梳理了小流域采样断面布设、采样频率优化和河流断面通量估算3方面的主要进展. 1)从小流域样点布设来看, 包括常规监测采样设计、针对性监测采样设计和融合前两种类型的监测采样设计3种类型, 点位布设方法上有遗传算法、模糊逻辑法、熵值法和模型法等, 样品采集方式包括随机采样、复合采样、综合采样以及连续采样4种类型, 其中复合采样应用广泛；2)从采样频率来看, 1~2周一次的采样频率即可精准获取污染负荷通量, 若需要进一步提高精度, 可在水文/水质变异大的时期提高监测频率及在特殊断面加密布点；3)在通量估算上, 当前主要的计算方法有平均法、插值法和回归/曲线法3类方法, 其中流量加权的浓度估计法、插值算法和LOADEST法是简便且精确的方法, 方法的选择上也可根据不同时期流域污染源特征进一步优化.;Watershed-scale non-point source pollution monitoring is an important basis for systematically understanding the occurrence, transportation and transformation processes of agricultural non-point source pollution and its effective control. Presently methods monitoring non-point source pollution at field scale are relatively mature, while methods applied at watershed scale, particularly the layout of monitoring section and sampling frequency setting, are rarely studied. This paper reviewed the main progress of sampling section layout, sampling frequency optimization and loading estimation of river sections. From the perspective of sampling layout in small watershed, there are three types of monitoring design including probabilistic design, targeted design and both. Sampling points were located with genetic algorithm, fuzzy logic method, entropy and modelling method, etc. Sample collection methods include random sampling, compound sampling, comprehensive sampling and continuous sampling. Compound sampling is widely used. Concerning the sampling frequency, once every 1-2 weeks is sufficient to obtain accurate pollution loading. If higher accuracy is necessary, the monitoring frequency and special sections can be increased when hydrology/water quality variation is large. For loading estimation, average method, interpolation method and regression or curve method were widely used. Concentration estimation algorithm by flow-weighted, interpolation and LOADEST are simple and exact. Selection of methods could be further optimized based on the characteristics of pollution sources in different periods of the basin.

A pollution assessment of grossly polluting industries in India

The unprecedented growth of industrialisation and urbanisation produced various environmental problems in India. In this paper we quantified the pollution potential and their reduction from grossly polluting industries. Pollution load estimation has been done according to their performance and treatment. We identified 17 grossly polluting industries and their total numbers for sugar, distilleries, caustic soda, fertiliser, oil refinery, iron and steel, pulp and paper, pharmaceuticals, pesticides, petrochemicals, thermal power plants, dye stuffs, tanneries, cement, aluminium smelter, copper smelter, and zinc smelter. The paper revealed the volume of wastewater and air pollution generated and pollution potential in terms of important environmental pollutants. The total contribution of major organic pollutant in effluent load from 17 sectors of industries and overall reduction due to control measures and percent reduction was estimated. The contribution of emission load in terms of particulate matter (PM) by industries is also estimated and percent reduction due to control strategies.

Estimation of Basic Unit Pollution Load of an Urban Area in the Lake Biwa Basin

Estimation of Basic Unit Pollution Load of an Urban Area in the Lake Biwa Basin

Modelled estimates of dissolved inorganic nitrogen exported to the Great Barrier Reef lagoon

Measuring stream pollutant loads across the Great Barrier Reef (GBR) catchment area (GBRCA) is challenging due to the spatial extent, climate variability, changing land use and evolving land management practices, and cost. Thus, models are used to estimate baseline pollutant loads. The eWater Source modelling framework is coupled with agricultural paddock scale models and the GBR Dynamic SedNet plugin to simulate dissolved inorganic nitrogen (DIN) generation and transport processes across the GBRCA. Catchment scale monitoring of flow and loads are used to calibrate the models, and performance is assessed qualitatively and quantitatively. Modelling indicates almost half (47%) of the total modelled DIN load exported to the GBR lagoon is from the Wet Tropics, and almost half of the total modelled DIN load is from sugarcane areas. We demonstrate that using locally developed, customised models coupled with a complementary monitoring program can produce reliable estimates of pollutant loads.

Modelled estimates of fine sediment and particulate nutrients delivered from the Great Barrier Reef catchments

The eWater Source modelling framework has been modified to support the Great Barrier Reef (GBR) Dynamic SedNet catchment modelling concept, which is used to simulate fine sediment and particulate nutrient generation, loss, and transport processes across GBR catchments. Catchment scale monitored data sets are used to calibrate and evaluate models. Model performance is assessed qualitatively and quantitatively. Modelling predicts that approximately half of generated sediment is delivered to the GBR lagoon; the remainder is deposited on floodplains, trapped in reservoirs or lost through other minor processes (e.g. irrigation extractions). Gullies are the major source of sediment, with comparable contributions from hillslopes and streambanks. Hillslope sources are considered the major source of particulate nutrients across the GBR catchments. We demonstrate that using locally developed, customised models coupled with a complementary monitoring program can produce credible modelled estimates of pollutant loads and provide a platform for testing catchment scale assumptions and scenarios.

Application of Convolution Neural Networks and Hydrological Images for the Estimation of Pollutant Loads in Ungauged Watersheds

River monitoring and predicting analysis for establishing pollutant loads management require numerous budgets and human resources. However, it is general that the number of government officials in charge of these tasks is few. Although the government has been commissioning a study related to river management to experts, it has been inevitable to avoid the consumption of a massive budget because the characteristics of pollutant loads present various patterns according to topographic of the watershed, such as topology like South Korea. To address this, previous studies have used conceptual and empirical models and have recently used artificial neural network models. The conceptual model has a shortcoming in which it required massive data and has vexatious that has to enforce the sensitivity and uncertain analysis. The empirical model and artificial neural network (ANN) need lower data than a conceptual model; however, these models have a flaw that could not reflect the topographical characteristic. To this end, this study has used a convolution neural network (CNN), one of the deep learning algorithms, to reflect the topographical characteristic and had estimated the pollutant loads of ungauged watersheds. The estimation results for the biochemical oxygen demand (BOD) and total phosphorus (TP) loads for three ungauged watersheds were all excellent. However, prediction results with low accuracy were obtained when the hydrological images of a watershed with a land cover status different from the ungauged watersheds were used as training data for the CNN model.