Freeze-thaw basin nutrient loss comprehensive monitoring method
By combining remote sensing inversion and SWAT modeling, the layout of sampling points and field sample analysis were optimized, solving the resource limitation problem of monitoring nutrient loss in freeze-thaw watersheds and achieving efficient and low-cost scientific monitoring results.
Patent Information
- Application Number
- CN202510951314.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-10
- Publication Date
- 2025-11-21
AI Technical Summary
Existing technologies are insufficient for efficiently monitoring nutrient loss in mid-to-high latitude freeze-thaw basins under conditions of limited resources. Traditional methods are difficult to implement routinely in remote areas, and simulation results lack effective observational data support, leading to monitoring results that deviate from reality.
A comprehensive monitoring method combining remote sensing inversion, model simulation, and field sampling was adopted. Key periods were identified through remote sensing data, sub-basins were divided by combining the SWAT model, sampling point layout was optimized, field samples were collected and laboratory analysis was conducted, and peak pollution periods and key areas were systematically identified.
It enables efficient, low-cost, and scientifically reliable monitoring of nutrient loss in freeze-thaw watersheds in resource-limited areas, improving the understanding and management of loss patterns, and is applicable to high-altitude and high-latitude regions.
Smart Images
Figure CN120992502A_ABST
Abstract
Description
[Technical Field]
[0001] This invention relates to a comprehensive monitoring method for nutrient loss in freeze-thaw watersheds. It adopts a combined approach based on remote sensing inversion, model simulation, and field sampling coupled analysis. By combining long-term remote sensing data, hydrological and water quality models, and field sample analysis, it achieves high-efficiency, low-cost, and highly adaptable monitoring of nutrient loss in freeze-thaw watersheds, belonging to the field of water environment monitoring and pollution identification. [Background Technology]
[0002] Non-point source pollution caused by nutrient loss has become a critical environmental problem threatening global freshwater security. Under the backdrop of intensive agricultural use, the large-scale input of elements such as nitrogen and phosphorus has led to eutrophication and frequent algal blooms, severely disrupting ecosystem balance and drawing significant attention from global management agencies and the academic community. The formation mechanisms and migration pathways of non-point source pollution vary significantly across different geographical environments. Taking mid-to-high latitude regions as an example, seasonal freeze-thaw cycles profoundly influence the spatiotemporal transport processes of water and pollutants, which is key to understanding the pollution characteristics and water environment security of such watersheds.
[0003] Compared to low-latitude regions, mid-to-high latitude freeze-thaw basins exhibit more pronounced hydrological temporal variability. Particularly during the spring snowmelt phase, large amounts of snow melt rapidly within a short period, while the permafrost begins to thaw, releasing moisture from the "active layer." The combined effect of these two factors significantly enhances surface runoff. At this time, pollutants such as nitrogen and phosphorus accumulated in the soil over a long period (e.g., fertilizer residues, organic mineralization products, and livestock excrement) are easily washed into the river network system by meltwater, forming sudden peak pollution loads. This phase is not only a period of rapid water volume increase but also a period of large-scale pollutant migration. In summer, with increased rainfall intensity and frequency, precipitation further erodes exposed areas and sloping agricultural land, leading to non-point source pollution of nutrients. This dual-driven hydrological mechanism of "snowmelt + rainfall" gives non-point source pollution in freeze-thaw basins the significant characteristics of "short-term high intensity, strong spatial heterogeneity, and complex pathways."
[0004] However, long-term continuous monitoring of freeze-thaw watersheds faces numerous challenges due to factors such as remote geographical location, harsh natural conditions, and limited resource investment. Current research mainly relies on two technical approaches: first, deploying continuous automatic observation equipment (such as water level gauges, flow meters, and telemetry samplers) or relying on existing long-term hydrological station networks to acquire high-frequency, long-term series of water quality and hydrological data; second, simulating the spatial distribution of runoff and pollution loads through remote sensing inversion and hydrological and water quality models (such as SWAT, HBV, and TOPMODEL). While both methods have their advantages, their applicability in freeze-thaw areas remains significantly limited. The former incurs high construction costs, making routine operation and maintenance difficult in remote areas; while the latter, although possessing good regional adaptability, relies heavily on the quality of input data and validation with field samples, and model predictions lacking effective observation data often deviate from reality.
[0005] Therefore, it is necessary to construct a new technical framework that integrates the advantages of remote sensing inversion, model simulation, and field monitoring to achieve efficient identification and scientific monitoring of non-point source pollutant loss patterns in freeze-thaw watersheds under limited resource conditions. This method should possess the following advantages: (1) Based on readily available remote sensing data, analyze the hydrological and meteorological characteristics of the watershed to determine key loss times; (2) Avoid the need for model validation based on long-term monitoring samples, and utilize hydrological models to divide river networks and sub-watersheds, supporting scientific site selection; (3) Combine a small number of representative field samples for pollutant concentration measurement analysis, and determine typical loss areas and loss levels based on the distribution relationship between sub-watersheds and river networks, achieving a balance between monitoring costs and data representativeness. Through this integrated observation framework, the understanding and assessment capabilities of material loss patterns in cold-region watersheds can be significantly improved, providing feasible and scientifically reliable pollution monitoring and management support for areas lacking long-term monitoring systems. [Summary of the Invention]
[0006] This invention aims to provide a method for monitoring nutrient loss in freeze-thaw watersheds based on a coupled analysis of remote sensing inversion, model simulation, and field sampling. By integrating remote sensing inversion, model simulation, and field sampling analysis, the method scientifically sets the sampling period and locations, acquires key field observation data, and systematically identifies critical pollution periods and high-risk areas, thus providing a scientific basis for water environment protection and pollution control in data-scarce freeze-thaw watersheds.
[0007] This invention discloses a comprehensive monitoring method for nutrient loss in freeze-thaw watersheds, which specifically includes the following steps:
[0008] Step 1: Identification of Critical Periods of Nutrient Loss Based on Remote Sensing Data
[0009] Nutrient loss is primarily driven by surface runoff from rainfall or snowmelt. Analysis of regional snowmelt and rainfall runoff characteristics helps reveal the occurrence and evolution patterns of typical hydrological events, thereby identifying key periods for nutrient migration. The specific steps are as follows:
[0010] 1) Determining the key periods for snowmelt runoff: MODIS' MOD10A1 product provides Normalized Differential Snow Index (NDSI) raster data, with values ranging from 0 to 1 and a spatial resolution of 500m. This invention acquires NDSI data within the study area based on the vector boundary and the Google Earth Engine (GEE) platform, and processes it using reclassification tools in ArcGIS: when NDSI > 0.1, a value of 1 is assigned (indicating snow cover); when NDSI ≤ 0.1, a value of 0 is assigned (indicating no snow cover), thus generating a binary raster map of snow cover in the study area. Simultaneously, diurnal snowmelt raster data (spatial resolution of 0.1°) for the study area from 1980 to 2023 is downloaded based on the ERA5-Land dataset. To ensure spatial consistency, the aforementioned binary snow cover map is first resampled to the same 0.1° resolution as the ERA5-Land snowmelt data using resampling tools in ArcGIS. Subsequently, using the raster calculator tool in ArcGIS, the resampled snow cover map was used as a mask to perform multiplication calculations with the snowmelt data, retaining only the effective snowmelt values within the snow-covered areas. Further, the Zonal Statistics as Table tool in ArcGIS was applied to statistically analyze the daily effective snowmelt raster data, extracting the total daily snowmelt amount for the study area. Based on this, the daily totals were summarized monthly to obtain the total snowmelt amount for each month, and the multi-year average monthly snowmelt distribution was calculated. Based on the temporal characteristics of the multi-year average monthly snowmelt, the months with significantly concentrated snowmelt were identified, thus defining the critical period for snowmelt runoff in the study area.
[0011] 2) Identifying Key Periods for Rainfall-Runoff: Based on the vector boundary of the study area, daily-scale temperature and precipitation raster data (resolution 0.1°) from the ERA5-Land dataset were acquired using the GEE platform. The daily temperature and precipitation data were statistically analyzed using the Zonal Statistics as Table tool in ArcGIS, extracting the total daily precipitation and average daily temperature for the study area. Based on this, the daily data were aggregated monthly to calculate the total monthly precipitation and average monthly temperature, and further, the time series of multi-year average monthly precipitation and average monthly temperature were calculated. Based on the temporal characteristics of the multi-year average monthly precipitation and average monthly temperature, months with concentrated precipitation above 0°C (when temperatures are above 0°C, precipitation mainly manifests as rainfall, i.e., months with concentrated rainfall) were identified, defining the key periods for rainfall-runoff in the study area.
[0012] Step 2: Sub-basin delineation and sampling point layout based on the SWAT model
[0013] First, digital elevation data (DEM) of the watershed area was acquired based on a geospatial data cloud platform, and high-precision river network information was extracted using the HydroRIVERS database. Then, in the ArcGIS environment, a SWAT model of the study area was constructed using the ArcSWAT module. Specifically, using the sub-watershed delineation function in the ArcSWAT module, based on the DEM and using the HydroRIVERS river network as a river mask, hydrological analysis (including flow direction, flow accumulation, river network extraction, and watershed boundary delineation) was used to complete the sub-watershed delineation and river network structure generation. Based on the sub-watershed and river network distribution, the sampling point locations were determined according to the following principles:
[0014] 1) Systematic deployment along the main stream from upstream to downstream: Sampling points are deployed sequentially from upstream to downstream according to the main river flow path to ensure coverage of all sub-basins outlets within the basin. This helps to identify the changing trends of longitudinal water quality processes in the basin and realize the spatial continuity of the basin as a whole.
[0015] 2) Adjust sampling density based on land use intensity: Increase sampling point density in areas with high levels of human disturbance (such as densely cultivated land, urban agglomerations, or grazing areas) to more sensitively capture the impact of non-point source pollution input caused by human activities; and appropriately reduce the number of sampling points in areas with relatively low disturbance, such as original grasslands and woodlands, to optimize resource allocation while maintaining representativeness.
[0016] 3) Prioritize the confluence of tributaries and main streams: Prioritize the placement of sampling points at the locations where major tributaries flow into the main stream. This can clarify the contribution of each sub-basin (especially tributaries with different pollution backgrounds) to the pollutant flux of the main river, thereby revealing the structural characteristics of spatial heterogeneous input.
[0017] 4) Ensure the feasibility of sample collection: Select a location with a stable riverbed, smooth water flow, wide water surface, no rapids or shallows, and convenient for sampling.
[0018] Step 3: Field sample collection and laboratory analysis
[0019] Based on a clear understanding of the time window and spatial locations for surface water sampling, standardized surface water sampling was conducted in accordance with the "Technical Specification for Surface Water Environmental Quality Monitoring" (HJ 91.2—2022). Pre-washed glass bottles were used to collect surface water samples. The bottles were rinsed three times with the sample to be tested beforehand. After sampling, the samples were immediately labeled, stored in a cool, dark place (4℃), and transported to the laboratory for analysis within 24 hours. The main indicators measured included total nitrogen and total phosphorus.
[0020] The specific method is as follows:
[0021] Total nitrogen determination method: Alkaline potassium persulfate digestion ultraviolet spectrophotometry (HJ 636—2012). Steps: ① Digestion: Take 10 mL of water sample into a 25 mL stoppered graduated tube, add 5 mL of alkaline potassium persulfate solution (weigh 40.0 g of potassium persulfate dissolved in 600 mL of water; separately weigh 15.0 g of sodium hydroxide dissolved in 300 mL of water. After the sodium hydroxide solution cools to room temperature, mix the two solutions and dilute to 1000 mL); seal and place in a high-temperature, high-pressure digestion apparatus, digesting at 120 °C for 30 min; remove after natural cooling. ② Color development: Then add 1.0 mL of hydrochloric acid solution (1+9), dilute with water to the 25 mL mark, stopper and mix well. ③ Using a 10 mm quartz cuvette, measure the absorbance at wavelengths of 220 nm and 275 nm on an ultraviolet spectrophotometer, using water as a reference. Calculate the sample nitrogen concentration based on the linear relationship between absorbance and nitrogen concentration in the nitrogen standard curve. ④ Method for plotting the nitrogen standard curve: Take six 25ml stoppered graduated tubes and add 0.00, 0.20, 0.50, 1.00, 3.00, and 7.00ml of potassium nitrate standard working solution respectively (preparation of potassium nitrate standard working solution: weigh 0.7218g of potassium nitrate, dissolve in an appropriate amount of water, transfer to a 1000ml volumetric flask, dilute with water to the mark, and mix well. Measure 10.00ml of the above solution into a 100ml volumetric flask, dilute with water to the mark, and mix well; this is the potassium nitrate standard working solution, concentration 10.0mg / L), dilute with water to 10ml, and then proceed with the determination steps ①②③. (Using A...) r A standard curve is plotted with the ordinate as the vertical axis and the corresponding nitrogen content as the horizontal axis. The calculation formula is as follows:
[0022] A b =A b220 -2A b275
[0023] A s =A s220 -2A s275
[0024] A r =A s -A b
[0025] In the formula, A b Corrected absorbance for blank solution; A b220 and A b275 The absorbance of the blank solution at wavelengths of 220 nm and 275 nm are respectively; A s Corrected absorbance for standard solutions; A s220 and A s275 These are the absorbances of the standard-transferred solution at wavelengths of 220 nm and 275 nm, respectively; Ar The difference between the absorbance of the standard solution and the absorbance of the blank solution is used to correct the absorbance.
[0026] Total phosphorus determination method: Ammonium molybdate spectrophotometric method (GB 11893-89). Procedure: ① Digestion: Take 25 ml of sample into a 50 ml stoppered graduated tube, add 4 ml of potassium persulfate (50 g / L), tighten the stopper, and secure it with a small piece of cloth and thread. Place the tube in a large beaker and put it in a high-temperature, high-pressure digester. Heat until the pressure reaches 1.1 kg / cm², corresponding to a temperature of 120℃, maintain this temperature for 30 minutes, then stop heating. After the pressure gauge reading drops to zero, remove the tube and allow it to cool. Then dilute with water to the mark. ② Color Development: Add 1 mL of ascorbic acid solution (100 g / L) to the diluted digestion solution and mix well. After 30 seconds, add 2 mL of molybdate solution (molybdate solution preparation method: dissolve 13 g of ammonium molybdate in 100 mL of water. Dissolve 0.35 g of potassium antimony tartrate in 100 mL of water. While stirring continuously, slowly add the ammonium molybdate solution to 300 mL of sulfuric acid (1+1), add the potassium antimony tartrate solution and mix thoroughly). ③ Absorbance Measurement: After standing at room temperature for 15 minutes, use a 30 mm path length cuvette to measure the absorbance at a wavelength of 700 nm, using water as a reference. After subtracting the absorbance of the blank test, calculate the phosphorus concentration of the sample based on the linear relationship between absorbance and phosphorus concentration in the phosphorus standard curve. ④ Method for constructing the phosphorus standard curve: Take seven 50ml stoppered graduated tubes and add 0.0, 0.50, 1.00, 3.00, 5.00, 10.0, and 15.0mL of phosphate standard solution, respectively. (Preparation of phosphate standard solution: Weigh 0.2197±0.001g of potassium dihydrogen phosphate, dried at 110℃ for 2h and cooled in a desiccator, dissolve in water, transfer to a 1000mL volumetric flask, add 800mL of water and 5mL of sulfuric acid (1+1), dilute with water to the mark and mix well. After mixing, take 10.0mL of the solution to a 250mL volumetric flask, dilute with water to the mark and mix well; this is the phosphate standard solution with a phosphorus content of 2.0μg / ml). Add water to 25mL. Then process according to steps ①②③. Use water as a reference and measure the absorbance. After subtracting the absorbance of the blank test, construct a standard curve with the corresponding phosphorus content.
[0027] Step 4: Monitoring and Identification of Pollutant Loss Intensity in Freeze-Thaw Watersheds
[0028] To systematically reveal the temporal variation characteristics and spatial distribution patterns of nutrient (total nitrogen and total phosphorus) loss in the watershed during the freeze-thaw period, a spatiotemporal comparative analysis was conducted based on the laboratory test data obtained in step three. This analysis aimed to identify the peak pollution periods in the freeze-thaw watershed, as well as the pollution levels and key pollution areas in different sub-basins within the freeze-thaw watershed. The specific steps are as follows:
[0029] Time dimension: Identifying pollution peak periods. The test results of samples from different key periods (snowmelt period and rainfall period) were categorized, and box plots of total nitrogen and total phosphorus concentrations (unit: mg / L) at each site were drawn for each key period. Pollution peak periods were determined by characteristics such as concentration distribution and average values. Periods with more concentrated concentration distribution and higher values, or higher average concentrations, indicate a greater loss load and are considered peak loss periods.
[0030] Spatial Dimension: By plotting the variation curves of nitrogen and phosphorus concentrations along the river from upstream to downstream, and combining the concentration levels at each sampling point with their relative trends to upstream points, the relative contribution of each sampling point's corresponding main stream sub-basin and its tributary sub-basins to the overall pollution can be identified. This allows for the determination of the pollution level in different sub-basins and the delineation of key pollution areas. The concentration levels at sampling points are determined using the quartile method, including low (concentration < first quartile), median (first quartile < concentration < third quartile), and high (concentration > third quartile). The first and third quartiles are the 25th and 75th percentile concentrations, respectively, after arranging all points in ascending order. Pollution levels are categorized into high, medium, and low, with "high-pollution" areas considered priority areas for remediation. The specific determination logic is as follows:
[0031] ① The concentration at the sampling point is low, and the concentration at the upstream sampling point is also low → This indicates that the pollution load of the sub-basin where the sampling point is located and its upstream tributary sub-basin are both relatively small, and the overall pollution level is low.
[0032] ② The concentration at the sampling point is low, while the concentration at the upstream sampling point is at a medium or high level → This indicates that the pollution load of the sub-basin where the sampling point is located is relatively small, and at the same time, there is a dilution effect in the upstream tributary sub-basin, and the overall pollution level is low.
[0033] ③ The concentration at the monitoring point is at a medium level, and the concentration at the upstream monitoring point is at a low level → This indicates that there is an increase in pollution load in the sub-basin where the monitoring point is located and its upstream tributary sub-basins, and the overall pollution level is moderate.
[0034] ④ The concentration at the sampling point is at the median level, and the concentration at the upstream sampling point is at the median level → This indicates that both the sub-basin where the sampling point is located and the upstream tributary sub-basin have a certain pollution load, and the overall pollution level is moderate.
[0035] ⑤ The concentration at the sampling point is at a medium level, while the concentration at the upstream sampling point is at a high level → This indicates that the pollution in the watershed where the sampling point is located has been relatively alleviated. This may be due to the dilution effect of the upstream tributaries or the reduction of pollution in the main stream. Therefore, the pollution level of the sub-watershed where the sampling point is located is moderate, and the pollution level of the upstream tributary sub-watershed is low.
[0036] ⑥ The concentration at the monitoring point is high, while the concentration at the upstream monitoring point is low or medium → This indicates that the pollution load in the sub-basin where the monitoring point is located has increased significantly, and there may also be new inputs in the upstream tributary sub-basins. The pollution level in the main stream is high, and the pollution level in the tributary is medium.
[0037] ⑦ The concentration at the sampling point is high, and the concentration at the upstream sampling point is also high → The pollution contribution from the sub-basin and its tributary sub-basins at this sampling point is significant, and the pollution level is high.
[0038] ⑧ Considering the pollution load accumulation effect of tributaries, the pollution level of the upstream sub-basin decreases by one level each time it passes through a confluence, using the confluence as a node. Taking scenario ⑦ as an example, when the concentration at point A and the concentration at the upstream point B are both high, the pollution level of the sub-basin at point A is high, and the pollution level of sub-basin C, where the tributary directly flowing into this sub-basin is located, is also high. If sub-basins D and E flow into C as upstream tributaries of C, then the pollution levels of sub-basins D and E are considered moderate.
[0039] The advantages and beneficial effects of this invention are as follows:
[0040] (1) The method is highly innovative: This invention integrates multiple information sources such as remote sensing identification, model division and seasonal sampling to observe the loss of nutrients in the freeze-thaw watershed in a systematic way.
[0041] (2) Strong adaptability: This method is particularly suitable for high-altitude, cold, and data-scarce regions, solving the shortcomings of traditional monitoring methods in terms of strong equipment dependence and weak spatial representativeness, and has good portability and regional adaptability.
[0042] (3) Strong representativeness of results: This invention conducts representative sampling based on the hydrology, meteorology and watershed characteristics of the freeze-thaw basin, and obtains the most representative information while minimizing the amount of field work.
[0043] (4) Outstanding technical and economic advantages: Compared with long-term automated equipment monitoring, this method is low in cost and high in efficiency, and is especially suitable for promotion and use in areas with weak technical capabilities but prominent water environment problems. [Attached Image Description]
[0044] Figure 1 This is a technical flowchart of the nutrient loss monitoring method described in this invention.
[0045] Figure 2 This is a schematic diagram of snowmelt and precipitation runoff characteristics based on remote sensing.
[0046] Figure 3 This is a schematic diagram of sub-basins and river networks based on SWAT.
[0047] Figure 4 This is a schematic diagram showing the sampling time and locations in the field.
[0048] Figure 5 This is a map showing the distribution of major pollutant concentrations in surface water at different times.
[0049] Figure 6 This is an assessment map of key areas for the loss of major nutrients in the watershed. [Specific Implementation Methods]
[0050] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. The embodiments described below are only for explaining the present invention and do not limit the scope of application of the present invention.
[0051] This case study selects the Orkhon basin, located in north-central Mongolia, as the case analysis site. This region is a typical mid-to-high latitude freeze-thaw basin. Due to its large basin area, the impact of extreme climate, and economic and technological limitations, the Orkhon basin lacks a scientific and systematic monitoring of nutrient loss.
[0052] This invention provides a comprehensive monitoring method for nutrient loss in freeze-thaw watersheds, such as... Figure 1 The process includes the following steps:
[0053] Step 1: Analysis of Snowmelt and Rainfall Runoff Characteristics and Identification of Key Periods in the Orkhon Watershed Based on Remote Sensing Data
[0054] Nutrient loss is primarily driven by surface runoff from rainfall or snowmelt. Analysis of regional snowmelt and rainfall runoff characteristics helps reveal the occurrence and evolution patterns of typical hydrological events, thereby identifying key periods for nutrient migration. The specific steps are as follows:
[0055] 1) Determining the Key Periods of Snowmelt Runoff: MODIS' MOD10A1 product provides Normalized Differential Snow Index (NDSI) raster data, with values ranging from 0 to 1 and a spatial resolution of 500m. This case study uses the Orkhon watershed vector boundary and Google Earth Engine (GEE) platform to acquire NDSI data for the Orkhon watershed, and then processes it using ArcGIS's reclassification tool: when NDSI > 0.1, a value of 1 is assigned (indicating snow cover); when NDSI ≤ 0.1, a value of 0 is assigned (indicating no snow cover), thus generating a binary raster map of snow cover in the study area. Simultaneously, diurnal snowmelt raster data (spatial resolution of 0.1°) for the Orkhon watershed from 1980 to 2023 is downloaded based on the ERA5-Land dataset. To ensure spatial consistency, the aforementioned binary snow cover map is first resampled to the same 0.1° resolution as the ERA5-Land snowmelt data using ArcGIS's resampling tool. Subsequently, using the raster calculator tool in ArcGIS, and with the resampled snow cover map as a mask, the snowmelt data was filtered, retaining only valid snowmelt values within snow-covered areas. Further, the Zonal Statistics as Table tool in ArcGIS was used to statistically analyze the daily masked snowmelt raster data, extracting the total daily snowmelt amount in the Orkhon watershed. Based on this, the daily totals were summarized monthly to obtain the total snowmelt amount for each month, and the multi-year average monthly snowmelt distribution was calculated. Figure 2 As shown in Figure (a), S, O, N, D, J, F, M, A, M, J, J, A represent October, November, December, January, February, March, April, May, June, July, August, and September, respectively. Snowmelt runoff in the Orkhon watershed is mainly concentrated in March to May, and May is selected as the critical period for snowmelt runoff in this case study.
[0056] 2) Determining Key Periods for Rainfall-Runoff: Based on the Orkhon watershed vector boundary, daily-scale temperature and precipitation raster data (resolution 0.1°) from the ERA5-Land dataset were acquired using the GEE platform. The daily temperature and precipitation data were statistically analyzed using the Zonal Statistics as Table tool in ArcGIS, extracting the total daily precipitation and average daily temperature for the study area. Based on this, the daily data were aggregated monthly to calculate the total monthly precipitation and average monthly temperature, and further, the time series of multi-year average monthly precipitation and average monthly temperature were obtained. Figure 2 As shown in Figure (b), temperatures generally remain above 0°C after April, indicating that precipitation after April is primarily in the form of rainfall. Rainfall is concentrated mainly between June and August, and this case study selects August as the critical period for rainfall runoff.
[0057] Step 2: Sub-basin delineation and sampling point layout of the Orkhon watershed based on the SWAT model
[0058] 90×90m digital elevation data for the Orkhon watershed was obtained from geospatial data cloud, and the Orkhon watershed river network database was acquired from HydroRIVERS. The SWAT model was then used to divide the Orkhon watershed into sub-basins and river networks, as follows: Figure 3 As shown, a total of 27 sub-basins were divided. Based on the division results, the specific sampling points were determined according to four principles for sampling point layout. A total of 10 surface water sampling points were set up in May (P1-P10), and 6 surface water sampling points were set up in August (P11-P16), as follows. Figure 4 As shown.
[0059] Step 3: Conduct field sampling and laboratory analysis in the Orkhon watershed.
[0060] Steps one and two jointly determined the sampling time and sampling points, ensuring comprehensive coverage of important runoff periods and key confluence points in the Orkhon watershed. During each sampling process, surface water samples were collected and preserved in strict accordance with the Technical Specifications for Surface Water Environmental Quality Monitoring (HJ 91.2—2022) for subsequent laboratory analysis. Simultaneously, the total nitrogen and total phosphorus contents in the surface water samples were determined using alkaline potassium persulfate digestion ultraviolet spectrophotometry and ammonium molybdate spectrophotometry, respectively.
[0061] Step 4: Monitoring and Identification of Pollutant Loss Intensity in the Orkhon Watershed
[0062] The experimental analysis results from step three were compared and analyzed according to season and river network distribution to identify box plots and longitudinal variation diagrams of nutrient concentrations at different times, such as... Figure 5 As shown.
[0063] 1) In terms of time: Figure 5 Figure (a) shows the total nitrogen (TN) and different forms of nitrogen (ammonia nitrogen (NH4+)) in water samples from May and August. + 4) Nitrogen (NO) - 3) The concentration distribution of TN: From the distribution of TN, the TN fluctuation range in May and August is relatively large, the peak values are close, and the average value in August is slightly higher than that in May. Figure 5 Figure (c) shows the concentration distribution of total phosphorus (TP) and different forms of phosphorus (dissolved phosphorus (DP), particulate phosphorus (PP), etc.) in water samples from May and August. The TP distribution shows that both the peak and average TP values in May were lower than those in August. This indicates that the TN loss intensity was similar in May and August, the two key periods, while the TP loss intensity was higher in August than in May.
[0064] 2) In terms of spatial dimension: Figure 5Figure (b) shows the variation of total nitrogen (TN) in the water body during May and August. P1, P2, P3, and P15 were at low levels with TN concentrations of 0.41, 0.47, 0.49, and 0.80 mg / L, respectively; P4, P7, P8, P9, P10, P12, P14, and P16 were at medium levels with TN concentrations between 1.08 and 2.06 mg / L; and P5, P6, P11, and P13 were at high levels with TN concentrations of 2.56, 3.60, 2.54, and 2.69 mg / L, respectively. Figure 5 Figure (d) shows the longitudinal variation of total nitrogen (TP) in the water body in May and August. P1, P2, P3, and P7 are at low levels, with TP concentrations of 0.014, 0.023, 0.019, and 0.036 mg / L, respectively; P4, P5, P8, P9, P10, P12, P14, and P15 are at medium levels, with TP concentrations between 0.045 and 0.132 mg / L; P6, P11, P13, and P16 are at high levels, with TP concentrations of 0.19, 0.15, 0.42, and 0.35 mg / L, respectively. Based on the spatial dimension and the logic for determining key areas, key watersheds for total nitrogen (TN) and total TP were identified, such as... Figure 6 As shown. Figure 6 Figure (a) shows the key areas of TN, namely sub-basins 3, 10, 12, 17, 22 and 23; Figure 6 Figure (b) shows the key areas of TP, namely sub-basins 3, 10 and 23.
Claims
1. A method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds, characterized in that, Includes the following steps: Step 1: Identification of Critical Periods of Nutrient Loss Based on Remote Sensing Data Nutrient loss is driven by surface runoff from rainfall or snowmelt, and it is necessary to determine the critical periods for snowmelt runoff and rainfall runoff. Step 2: Sub-basin delineation and sampling point layout based on the SWAT model First, digital elevation data (DEM) of the watershed area was acquired based on a geospatial data cloud platform, and high-precision river network information was extracted using the HydroRIVERS database. Then, in the ArcGIS environment, a SWAT model of the study area was constructed using the ArcSWAT module. Specifically, using the sub-watershed division function in the ArcSWAT module, based on the DEM and with the HydroRIVERS river network as a river mask, hydrological analysis was used to complete the sub-watershed division and river network structure generation. Based on the sub-watershed and river network distribution, the sampling point locations were determined. Step 3: Field sample collection and laboratory analysis Based on the clear time window and spatial layout of surface water sampling points, standardized surface water sampling work was carried out in accordance with the "Technical Specification for Surface Water Environmental Quality Monitoring" (HJ 91.2—2022). During the sampling process, pre-washed glass bottles were used to collect surface water samples. The sampling bottles were rinsed three times with the water sample to be tested beforehand. After sampling, the samples were immediately labeled, stored in the dark and refrigerated (4℃), and transported to the laboratory for analysis and testing within 24 hours after sampling. The measured parameters include: total nitrogen and total phosphorus; Step 4: Monitoring and Identification of Pollutant Loss Intensity in Freeze-Thaw Watersheds Based on the laboratory test data obtained in step three, a spatiotemporal comparative analysis is conducted to identify the peak pollution period in the freeze-thaw basin, as well as the pollution levels and key pollution areas in different sub-basins within the freeze-thaw basin.
2. The method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1, characterized in that: Step one further includes: determining the key periods for snowmelt runoff: MODIS MOD10A1 provides Normalized Differential Snow Index (NDSI) raster data with a numerical range of 0–1 and a spatial resolution of 500m; based on the vector boundary of the study area and Google... The EarthEngine platform acquired NDSI data for the region and processed it using ArcGIS's reclassification tool: when NDSI > 0.1, a value of 1 was assigned, indicating snow cover; when NDSI ≤ 0.1, a value of 0 was assigned, indicating no snow cover, thus generating a binary raster map of snow cover for the study area. Simultaneously, diurnal snowmelt raster data for the study area from 1980 to 2023 were downloaded based on the ERA5-Land dataset. To ensure spatial consistency, the binary snow cover map was first resampled to the same 0.1° resolution as the ERA5-Land snowmelt data using ArcGIS's resampling tool. Subsequently, using ArcGIS's raster calculator tool, the resampled snow cover map was used as a mask to perform multiplication calculations with the snowmelt data, retaining only the valid snowmelt values within the snow-covered area. Furthermore, ArcGIS's Zonal Statistics as... The Table tool statistically analyzes the effective daily snowmelt raster data to extract the total daily snowmelt amount in the study area. Based on this, the daily total is summarized monthly to obtain the total snowmelt amount for each month, and the multi-year average monthly snowmelt distribution is calculated. According to the temporal characteristics of the multi-year average monthly snowmelt, the months with significant snowmelt concentration are determined, thereby defining the critical period of snowmelt runoff in the study area.
3. A method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1 or 2, characterized in that: Step one further includes: determining the key periods for rainfall runoff: based on the vector boundary of the study area, daily-scale temperature and precipitation raster data from the ERA5-Land dataset are obtained using the Google Earth Engine platform; the daily temperature and precipitation data are statistically analyzed using the Zonal Statistics as Table tool in ArcGIS, and the total daily precipitation and average daily temperature of the study area are extracted respectively; based on this, the daily data are summarized by month to calculate the total monthly precipitation and average monthly temperature, and the time series of multi-year average monthly precipitation and average monthly temperature are further calculated; based on the temporal characteristics of multi-year average monthly precipitation and average monthly temperature, the months in which the temperature is above 0℃ and precipitation occurs in a concentrated manner are identified, and the key periods for rainfall runoff in the study area are defined.
4. The method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1, characterized in that: Step two further includes: systematically deploying sampling points from upstream to downstream along the main stream: sampling points are deployed sequentially from upstream to downstream according to the main river flow path to ensure coverage of all sub-basin outlets within the basin, which helps to identify the changing trends of longitudinal water quality processes in the basin and realize the spatial continuity expression of the basin as a whole.
5. A method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1 or 4, characterized in that: Step two further includes: adjusting the sampling density based on land use intensity: increasing the sampling point density in areas with high levels of human disturbance to capture the impact of non-point source pollution input caused by human activities; and reducing the number of sampling points in areas with relatively low disturbance such as original grasslands and woodlands to optimize resource allocation.
6. The method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1, characterized in that: Step two further includes: prioritizing the confluence of tributaries and the main stream: prioritizing the placement of sampling points at the locations where major tributaries flow into the main stream can clarify the pollutant flux contribution of each sub-basin to the main river body, thereby revealing the structural characteristics of spatial heterogeneous input.
7. The method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1, characterized in that: Step three further includes: a method for determining total nitrogen: alkaline potassium persulfate digestion ultraviolet spectrophotometry; Specific steps: ① Digestion: Take 10 mL of water sample into a 25 mL stoppered graduated tube, add 5 mL of alkaline potassium persulfate solution to the water sample, weigh 40.0 g of potassium persulfate and dissolve it in 600 mL of water; separately weigh 15.0 g of sodium hydroxide and dissolve it in 300 mL of water; after the sodium hydroxide solution cools to room temperature, mix the two solutions and make up to 1000 mL; seal and place in a high-temperature and high-pressure digestion apparatus, digest at 120 °C for 30 min; remove after natural cooling; ② Color development: Then add 1.0 ml of hydrochloric acid solution, dilute with water to the 25 ml mark, stopper and mix well; ③ Using a 10mm quartz cuvette, measure the absorbance at wavelengths of 220nm and 275nm on a UV spectrophotometer with water as a reference; calculate the nitrogen concentration of the sample based on the linear relationship between absorbance and nitrogen concentration in the nitrogen standard curve. ④ Method for constructing the nitrogen standard curve: Take six 25ml stoppered graduated tubes and add 0.00, 0.20, 0.50, 1.00, 3.00, and 7.00ml of potassium nitrate standard working solution, respectively. Preparation of the potassium nitrate standard working solution: Weigh 0.7218g of potassium nitrate, dissolve it in an appropriate amount of water, transfer it to a 1000ml volumetric flask, dilute with water to the mark, and mix well. Measure 10.00ml of the above solution into a 100ml volumetric flask, dilute with water to the mark, and mix well. This is the potassium nitrate standard working solution with a concentration of 10.0mg / L. Dilute with water to 10ml, and then proceed with the determination steps ①, ②, and ③. (Using A...) r A standard curve is plotted with the ordinate as the vertical axis and the corresponding nitrogen content as the horizontal axis; the calculation formula is as follows: A b =A b220 -2A b275 A s =A s220 -2A s275 A r =A s -A b In the formula, A b Corrected absorbance for blank solution; A b220 and A b275 The absorbance of the blank solution at wavelengths of 220 nm and 275 nm are respectively; A s Corrected absorbance for standard solutions; A s220 and A s275 These are the absorbances of the standard-transferred solution at wavelengths of 220 nm and 275 nm, respectively; A r The difference between the absorbance of the standard solution and the absorbance of the blank solution is used to correct the absorbance.
8. The method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1, characterized in that: Step three further includes: a method for determining total phosphorus: ammonium molybdate spectrophotometry; Specific steps: ① Digestion: Take 25ml of sample into a 50ml graduated tube with a stopper, add 4ml of potassium persulfate (50g / L), seal the graduated tube tightly, tie the glass stopper with a small piece of cloth and thread, place it in a large beaker and heat it in a high-temperature and high-pressure digester. When the pressure reaches 1.1kg / cm2 and the corresponding temperature is 120℃, maintain this temperature for 30 minutes and then stop heating. After the pressure gauge reading drops to zero, remove the sample and let it cool. Then dilute it with water to the mark. ② Color development: Add 1 mL of 100 g / L ascorbic acid solution to the diluted digestion solution and mix well. After 30 seconds, add 2 mL of molybdate solution and mix thoroughly. Molybdate solution preparation method: Dissolve 13 g of ammonium molybdate in 100 mL of water; dissolve 0.35 g of potassium antimony tartrate in 100 mL of water; slowly add the ammonium molybdate solution to 300 mL of sulfuric acid while stirring continuously, then add the potassium antimony tartrate solution and mix thoroughly. ③Absorbance measurement: After being placed at room temperature for 15 minutes, the absorbance was measured at a wavelength of 700 nm using a 30 mm cuvette and water as a reference. After subtracting the absorbance of the blank test, the phosphorus concentration of the sample was calculated based on the linear relationship between absorbance and phosphorus concentration in the phosphorus standard curve. ④ Method for constructing the phosphorus standard curve: Take seven 50ml stoppered graduated tubes and add 0.0, 0.50, 1.00, 3.00, 5.00, 10.0, and 15.0mL of phosphate standard solution, respectively. Preparation of the phosphate standard solution: Weigh 0.2197±0.001g of potassium dihydrogen phosphate that has been dried at 110℃ for 2 hours and cooled in a desiccator. Dissolve it in water and transfer it to a 1000mL volumetric flask. Add 800mL of water and 5mL of sulfuric acid, dilute to the mark with water, and mix well. After mixing, take 10.0mL of the solution and transfer it to a 250mL volumetric flask. Dilute to the mark with water and mix well. This is the phosphate standard solution with a phosphorus content of 2.0μg / ml. Add water to 25mL. Then process according to steps ①, ②, and ③ of the determination. Use water as a reference and measure the absorbance. After subtracting the absorbance of the blank test, construct a standard curve with the corresponding phosphorus content.
9. The method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1, characterized in that: Step four further includes: Time dimension: identifying pollution peak periods; classifying the test results of samples from different key periods, namely snowmelt period and rainfall period, drawing box plots of total nitrogen and total phosphorus concentrations at each site during different key periods, and determining pollution peak periods by the characteristics of concentration distribution and average values; periods with more concentrated concentration distribution and higher values, or higher average concentration values, indicate a greater loss load and are considered to be peak loss periods.
10. A method for comprehensive monitoring of nutrient loss in freeze-thaw watersheds according to claim 1 or 9, characterized in that: Step four further includes: Spatial dimension: by plotting the change curves of nitrogen and phosphorus concentrations along the river from upstream to downstream, and combining the concentration levels of each sampling point and their relative change trends with upstream points, the relative contribution of the main stream sub-basins and their tributary sub-basins corresponding to each sampling point to the overall pollution is identified, thereby determining the pollution degree of different sub-basins and delineating key pollution areas; the concentration levels of the sampling points are determined by the quartile method.