Radial flow change attribution decomposition quantification method based on hydrological simulation and deviation correction

Abstract

The invention discloses a runoff change attribution decomposition quantification method based on hydrological simulation and deviation correction. The method comprises the following steps: dividing a natural period and a human activity influence period; calibrating model parameters by using hydrometeorological data in a natural period; calculating an average simulation deviation of monthly runoff based on a runoff flow process actually measured and simulated in a natural period; using the meteorological data of the human activity influence period to drive the watershed hydrological model, and simulating the natural runoff process of the human activity influence period; on the basis of monthly runoff simulation deviation, the simulated natural runoff volume in the natural period and the human activity influence period is corrected; and analyzing the influence of climate change and human activity on runoff change by utilizing the measured values of the reference period and the human activity influence period and the deviated and corrected natural runoff volume. The method not only has the flexibility of a runoff change attribution identification method based on hydrological simulation, but also can effectively avoid the influence of simulation errors on runoff change attribution identification to a certain extent.

Description

Technical Field

[0001] The invention belongs to the technical field of hydrological simulation, and in particular relates to a runoff change attribution decomposition and quantification method based on hydrological simulation and deviation correction. Background Art

[0002] Scientifically and accurately quantifying and evaluating the causes of river runoff changes has very important scientific significance and practical value for regional water resources evaluation and scientifically supporting the sustainable development of the economy and society with water resources as a rigid constraint. The impact of climate change and human activities on river runoff has always been a hot topic of concern for river basin managers and a key scientific issue for hydrologists. Research in this area has gradually developed from the initial qualitative trend analysis to quantitative analysis of runoff change attribution based on hydrological simulation.

[0003] At present, the quantitative analysis methods of runoff change attribution based on hydrological simulation are based on the important premise that the basin hydrological model can perfectly simulate the runoff in the natural period, that is, the simulated runoff in the natural period is completely consistent with the measured runoff. In the actual runoff simulation, although the simulation error of the multi-year average is small, due to the complexity of the hydrological process, the error of the measured runoff monitoring, and the limitation of the mathematical generalization of the hydrological process by the basin hydrological model, the basin hydrological model cannot completely reproduce the measured natural runoff process. Even on the interdecadal scale, there will be certain simulation errors, and the simulation errors on the monthly scale are even greater (more than 10%). The existence of the simulation error of natural runoff is bound to have a certain impact on the quantitative identification of runoff change attribution, and the error of runoff change attribution identification is even greater on the monthly scale. Summary of the invention

[0004] Aiming at the deficiency of large simulation error in the existing methods, the present invention proposes a runoff change attribution decomposition and quantification method based on hydrological simulation and deviation correction.

[0005] To achieve the above object, the present invention adopts the following technical solution: A quantitative method for attribution decomposition of runoff change based on hydrological simulation and bias correction includes the following steps: (1) Collect long-term hydrological and meteorological data and human activity data in the basin, calculate the annual runoff coefficient sequence of the basin, diagnose the mutation points of the annual runoff coefficient sequence, and divide the sequence into the natural period and the period affected by human activities.

[0006] (2) Select a basin hydrological model suitable for studying the runoff characteristics of the basin, use the hydrological and meteorological data of the natural period, take the NSE coefficient and relative error RE as the target, calibrate the parameters of the basin hydrological model, and simulate the runoff process in the natural period.

[0007] (3) Based on the measured and simulated runoff processes during the natural period, the average simulation deviation of runoff from January to December was analyzed.

[0008] (4) Using meteorological data during the period of human activity as model input, driving the calibrated basin hydrological model, and simulating the natural runoff process during the period of human activity; (5) Based on the average simulation deviation of runoff from January to December in the natural period, the simulated runoff in the natural period is revised and corrected to obtain the runoff process based on deviation correction; (6) Taking the natural period as the reference period, the impact of climate change and human activities on runoff changes and its annual distribution is analyzed based on the measured runoff in the reference period and the period affected by human activities and the natural runoff after deviation correction.

[0009] Furthermore, the hydrological and meteorological data in step (1) include runoff, precipitation, water surface evaporation and temperature; the human activity data include water conservancy projects, industrial and agricultural water withdrawal and water consumption data.

[0010] The annual runoff coefficient sequence of the basin is calculated, and mathematical statistics methods, such as Mann-Kendall mutation test and ordered cluster analysis, are used to diagnose the mutation points of the annual runoff coefficient sequence. Combined with the human activities in the basin, the sequence is divided into the natural period and the period affected by human activities.

[0011] Furthermore, the watershed hydrological model in step (2) is a RCCC-WBM water balance model. The parameters of the watershed hydrological model are calibrated using a genetic algorithm, which belongs to the prior art.

[0012] If the Nash-Sutcliffe efficiency coefficient NSE is greater than 60% and the relative error RE is less than 5%, the model is considered qualified.

[0013] Furthermore, the method for calculating the deviation coefficient of the monthly runoff simulation in step (3) is:

[0014] In the formula, For the i The deviation coefficient of the monthly runoff simulation, i The value ranges from 1 to 12; and The natural period i The average of the measured and simulated runoff in a month is calculated as follows:

[0015] In the formula, and Respectively j Year i Measured and simulated runoff volume for the month; jThe value range is 1~M, where M is the number of years in the natural period.

[0016] Furthermore, the calculation method of runoff based on deviation correction in the natural period in step (5) is: ; In the formula, For the natural period j Year i Simulated runoff volume for the month; is the natural runoff in month i of year j after correction based on deviation in the natural period; The calculation method of runoff based on deviation correction during the period of human activity in step (5) is: ; In the formula, The period of human activities j’ Year i Simulated runoff volume for the month; The first period of human activity impact based on deviation correction j’ The natural runoff in month i of the year is j’ The value of is 1~N, where N is the year of the runoff sequence.

[0017] Furthermore, step (6) includes: (63) Calculate the mean annual runoff measured during the baseline period and the period of human activity influence and the mean annual natural runoff corrected based on the deviation; (64) Calculate the changes in annual runoff during the period of human activity, as well as the impact and contribution of climate change and human activities on annual runoff changes; (65) Calculate the change in monthly runoff compared with the baseline period during the period of human activity, as well as the impact of climate change and human activities on the change in monthly runoff.

[0018] Furthermore, before step (63), the following steps are further included: (61) Calculation of the mean of measured runoff from January to December of the reference period and the mean of natural runoff corrected based on deviations; (62) Calculation of the mean measured runoff from January to December during the period of human activity and the mean natural runoff corrected based on deviations.

[0019] The calculation content of step (6) is specifically as follows: (61) Calculation of the mean of measured runoff from January to December of the reference period and the mean of natural runoff corrected based on deviation:

[0020] Where: They are the average of the measured runoff in the benchmark period from January to December (mm) and the average of the natural runoff based on deviation correction (mm). Through deviation correction, from the average of many years, and The two are completely equal; (62) Calculation of the mean of measured runoff from January to December during the period of human activity impact and the mean of natural runoff based on deviation correction:

[0021] Where: They are respectively the multi-year mean measured runoff (mm) during the period of human activities from January to December and the mean natural runoff (mm) based on deviation correction.

[0022] (63) Calculate the mean annual runoff measured during the base period and the period of human activity influence and the mean annual natural runoff corrected based on deviations. Since the natural runoff corrected based on deviations during the base period is revised based on the measured runoff during that period, the two are exactly the same. Therefore, only the measured annual runoff during that period is calculated.

[0023]

[0024] Where: is the mean annual runoff measured during the reference period (mm), and They are respectively the mean annual runoff measured during the period of human activity impact (mm) and the mean annual natural runoff after deviation correction (mm).

[0025] (64) Calculate the changes in annual runoff during the period of human activity, as well as the impact and contribution of climate change and human activities on annual runoff changes:

[0026] In the formula, is the change in annual runoff during the period of human activity compared with the baseline period (mm), and are the climate change during the period of human activities and the impact of human activities on annual runoff changes (mm), and are the contribution rates (%) of climate change and human activities to annual runoff changes respectively.

[0027] (65) The impact of climate change and human activities on monthly runoff changes was analyzed on a monthly scale using a similar analytical approach to that used to analyze annual runoff changes. The change in monthly runoff compared to the baseline period during the period of human activity impact and the impact of climate change and human activities on monthly runoff changes were calculated as follows:

[0028] In the formula, is the change in runoff in month i during the period of human activity impact compared with the baseline period (mm); and are the climate change during the period of human activities and the impact of human activities on the runoff change in the i-th month (mm).

[0029] The beneficial effects of the present invention are: The method of the present invention not only has the flexibility of the attribution identification method based on hydrological simulation runoff change, but also can effectively avoid the influence of simulation errors on the attribution identification of runoff change to a certain extent, especially the influence on the attribution identification of runoff annual distribution change. This method can be more flexibly and effectively applied to the specific work of watershed water resources evaluation and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The measured and simulated monthly flow process lines of the Weihe River Huaxian Station during the natural period (1955-1969); Figure 2 This is the monthly distribution process of the measured and simulated runoff at Huaxian Station and the simulation deviation diagram; Figure 3 This is the change of runoff volume at Huaxian Station of Weihe River during the period of human activity impact (1970-2015) compared with the base period (1955-1969) and the annual distribution map of the impact of climate change and human activities; Figure 4 This is the process diagram of the measured runoff, simulation and deviation-corrected natural runoff at the Huaxian Station of the Weihe River from 1955 to 2015; Figure 5 This is a process diagram of the multi-year average measured monthly runoff at the Huaxian Station of the Weihe River during the benchmark period (1955-1969) and the multi-year average measured, simulated and bias-corrected monthly runoff during the period of human activity impact (1970-2015). DETAILED DESCRIPTION

[0031] The Wei River is a first-level tributary of the Yellow River. The basin area above Huaxian Station is 106,498 km 2 Taking the Weihe River Basin as an example, the specific analysis steps of the method of the present invention are as follows: (1) Collect hydrological and meteorological data in the Weihe River Basin from 1955 to 2015, including runoff, precipitation, evaporation and temperature data; and human activity data in the basin, including water conservancy projects, land use data and water consumption data.

[0032] The measured annual runoff process of Huaxian Station was analyzed, and the annual runoff coefficient sequence was calculated according to the annual runoff of Huaxian Station and the average annual precipitation in the basin. The ordinal cluster analysis method was used to diagnose that the mutation point of the annual runoff coefficient sequence was 1970. Combined with the human activities in the basin, it was determined that the natural period and the period affected by human activities in the basin were 1955-1969 and 1970-2015, respectively.

[0033] (2) The RCCC-WBM water balance model was selected to simulate the hydrological process. The RCCC-WBM water balance model is a large-scale hydrological model developed and gradually improved by the team of the Climate Change Research Center of the Ministry of Water Resources. The model has the advantages of simple structure, few parameters, easy to understand, and low data requirements. Using the hydrological and meteorological data of the natural period, the model parameters were calibrated with the NSE coefficient and relative error RE as the target to simulate the runoff process in the natural period, such as Figure 1 The Nash-Sutcliffe efficiency coefficient NSE is 0.72, and the multi-year average simulation relative error RE is -2.2%. The model is qualified and can be used to simulate the natural runoff process of the basin.

[0034] (3) Calculate the mean of measured and simulated runoff from January to December in the natural period using formula (2) and formula (3), where M is 15; analyze the simulated runoff deviation from January to December, such as Figure 2 shown.

[0035] (4) Fix the model parameters, use the meteorological data from the period of human activity influence to drive the calibrated basin hydrological model, and simulate the natural runoff process during the period of human activity influence (1970-2015).

[0036] (5) Based on the runoff simulation deviation coefficient from January to December calculated by formula (1), the simulated runoff during the natural period (1955-1969) and the period affected by human activities was corrected to obtain the natural runoff process based on deviation correction. Figure 4 This is a diagram of the natural runoff process of the Huaxian Station of the Weihe River from 1955 to 2015, including the measured runoff, simulation and deviation correction.

[0037] (6) Taking the natural period as the reference period, based on the measured runoff in the reference period and the period of human activity influence and the natural runoff after deviation correction, analyze the impact of climate change and human activities on the runoff changes and its annual distribution during the period of human activity influence, such as Figure 3 The figure shows the change of runoff during the human activity impact period at the Huaxian Station of the Weihe River compared with the base period and the annual distribution of the impact of climate change and human activities. Specifically, it includes: Calculate the mean measured runoff from January to December during the baseline period and the period of human activity impact and the mean natural runoff corrected based on deviation.

[0038] The mean annual runoff actually measured during the base period and the period affected by human activities and the mean annual natural runoff corrected based on deviations are calculated. Since the natural runoff corrected based on deviations during the base period is revised based on the actually measured runoff during that period, the two are exactly the same. Therefore, only the actually measured annual runoff during that period is calculated.

[0039] The changes in the measured annual runoff during the period of human activity, as well as the impact and contribution of climate change and human activities on the change in annual runoff are calculated, as shown in Table 1: Table 1 Impact of climate change and human activities on annual runoff changes

[0040] The analysis method similar to that of annual runoff change is adopted to analyze the impact of climate change and human activities on monthly runoff change on a monthly scale, the change of monthly runoff during the period of human activity compared with the base period, and the impact of climate change and human activities on monthly runoff change, such as Figure 5 shown.

[0041] in conclusion: (1) The measured runoff at Huaxian Station in the Weihe River Basin showed a decreasing trend from 1955 to 2015. The measured runoff series changed suddenly in 1970, which divided the study period into the natural period (1955-1969) and the period affected by human activities (1970-2015).

[0042] (2) Taking the natural period (1955-1969) as the baseline period, the measured runoff during the period of human activities decreased by 34.45 mm. The impact of human activities and climate change caused the annual runoff to decrease by 26.98 mm and 7.47 mm, respectively. The contribution rates of human activities and climate change to the runoff reduction were 78.3% and 27.7%, respectively.

[0043] (3) Climate change increased runoff by 0.05 mm and 0.45 mm in February and July, respectively, during the period of human activity, and reduced runoff in other months. The runoff in October was the most affected, decreasing by about 2.53 mm. Human activities reduced runoff in all months, with the largest reduction in the flood season. The runoff in the flood season from May to September decreased by 17.14 mm, accounting for about 64% of the total reduction due to human activities.

[0044] (4) Figure 2 Although the average measured and simulated monthly flow in the natural period generally match each other, the relative error of some months exceeds 50% after error correction. In this case, if the deviation correction is not performed, the error in the annual distribution of runoff attribution will be even greater. Figure 5It can also be seen that the difference between the "1970-2015 simulation" and the "1970-2015 actual measurement" is quite large, which fully illustrates the necessity of bias correction.

[0045] (5) The attribution identification of runoff change based on bias correction has a certain effect on the attribution identification of annual runoff change, but has a very good effect on the attribution identification of monthly runoff change.

[0046] As above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes may be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A quantitative method for attribution decomposition of runoff change based on hydrological simulation and deviation correction, characterized in that: The following steps are involved: (1) Collect long-sequence hydrological and meteorological data and human activity data in the basin, calculate the annual runoff coefficient sequence of the basin, diagnose the mutation points of the annual runoff coefficient sequence, and divide the sequence into the natural period and the period affected by human activities; (2) Select a basin hydrological model suitable for studying the runoff characteristics of the basin, use the hydrological and meteorological data of the natural period, take the NSE coefficient and relative error RE as the target, calibrate the parameters of the basin hydrological model, and simulate the runoff process in the natural period; (3) Based on the measured and simulated runoff processes during the natural period, the average simulation deviation of runoff from January to December was analyzed; (4) Using meteorological data during the period of human activity as model input, the basin hydrological model calibrated in step (2) is driven to simulate the natural runoff process during the period of human activity; (5) Based on the average simulation deviation of runoff from January to December in the natural period, the simulated runoff in the natural period and the period affected by human activities was revised and corrected to obtain the runoff process based on deviation correction; (6) Taking the natural period as the reference period, the impact of climate change and human activities on runoff changes and its annual distribution is analyzed based on the measured runoff in the reference period and the period affected by human activities and the natural runoff after deviation correction.

2. The method for attribution decomposition and quantification of runoff change based on hydrological simulation and deviation correction according to claim 1 is characterized in that: The hydrological and meteorological data in step (1) include runoff, precipitation, water surface evaporation and temperature; the human activity data include water conservancy projects, land use data and water consumption data.

3. The method for quantifying runoff change attribution decomposition based on hydrological simulation and deviation correction according to claim 1 is characterized in that: The watershed hydrological model in step (2) is the RCCC-WBM water balance model.

4. The method for attribution decomposition and quantification of runoff change based on hydrological simulation and deviation correction according to claim 1 is characterized in that: The method for calculating the deviation coefficient of monthly runoff simulation in step (3) is: ; In the formula, For the i The coefficient of deviation of the monthly runoff simulation, i The value ranges from 1 to 12; and The natural period i The calculation formula for the mean of the measured and simulated runoff volume for a month is as follows: ； In the formula, and Respectively j Year i Measured and simulated runoff volume for the month; j The value range is 1~M, where M is the number of years in the natural period.

5. The method for attribution decomposition and quantification of runoff change based on hydrological simulation and deviation correction according to claim 4 is characterized in that: The calculation method of runoff volume based on deviation correction in the natural period in step (5) is: ; In the formula, For the natural period j Year i Simulated runoff volume for the month; is the natural runoff in month i of year j after correction based on deviation in the natural period; The calculation method of runoff based on deviation correction during the period of human activity in step (5) is: ; In the formula, The period of human activities j’ Year i Simulated runoff volume for the month; The first period of human activity impact based on deviation correction j’ The natural runoff in month i of the year is j’ The value of is 1~N, where N is the year of the runoff sequence.

6. The method for quantifying runoff change attribution decomposition based on hydrological simulation and deviation correction according to claim 5 is characterized in that: Step (6) includes: (63) Calculate the mean annual runoff measured during the baseline period and the period of human activity influence and the mean annual natural runoff corrected based on the deviation; (64) Calculate the changes in annual runoff during the period of human activity, as well as the impact and contribution of climate change and human activities on annual runoff changes; (65) Calculate the change in monthly runoff compared with the baseline period during the period of human activity, as well as the impact of climate change and human activities on the change in monthly runoff.

7. The method for quantifying runoff change attribution decomposition based on hydrological simulation and deviation correction according to claim 6 is characterized in that: Before step (63), the following steps are also included: (61) Calculation of the mean of measured runoff from January to December of the reference period and the mean of natural runoff corrected based on deviations; (62) Calculation of the mean measured runoff from January to December during the period of human activity and the mean natural runoff corrected based on deviations.

Images