Listen
graphic
View largeDownload slide

This article is focused on proposing a novel approach to determining flood mitigation practices based on coupled hydrodynamic and rainfall-tracking models. V-shaped and Baogaisi catchments were applied to assess the effects of the novel approach in mitigating peak discharge and runoff volume, respectively. Specifically, traditional afforestation scenarios were planned from downstream to upstream, while novel afforestation scenarios were designed based on the maximum and minimum contribution rate of sub-areas. Then two types of cases were simulated by the coupled hydrodynamic and rainfall-tracking model again to evaluate the mitigating effects of different afforestation practices. Results show that the coupled model is able to obtain accurate hydrodynamic and rainfall-tracing information simultaneously within each computing grid under the flooding caused by rainfall events. Moreover, in comparison with the reduction effects of traditional cases, results simulated under two catchments illustrated that the novel approach is able to determine spatial prioritization of afforestation management practices. In particular, there is a positive correlation between the contribution rate of afforestation area and the reduction effect of runoff volume caused by rainfall events. Thus, the research could provide a more scientific and reasonable guide in determining spatial prioritization flooding mitigation practices for planners and governments.

Listen

  • Develop a source-tracking analysis based on coupled hydrodynamic and rainfall-tracking models.

  • Obtain the runoff contribution rates of various sub-areas to flooding in impacted areas.

  • Offer a hydrodynamic and systematic insight in identifying spatial prioritization of afforestation management practices.

contribution rate, coupled hydrodynamic and rainfall-tracking model, mitigating flood regime, natural flood management (NFM), spatial prioritization
Listen

There is an increase in the frequency of floods in many places worldwide, especially in developing countries (Bradshaw et al. 2007; Rogger et al. 2017). Most efforts have been focused on the significant influence of climate change, which leads to more frequent and severe extreme rainfall events (Hall et al. 2014; Merz et al. 2014; Bevacqua et al. 2019; Roy et al. 2020), while only a few expensive efforts were asserted on the effects of land use change in altering floods (Rogger et al. 2017).

In recent years, afforestation in uplands is increasingly considered a significant component of natural flood management (NFM) to mitigate flood at the watershed scale (Dadson et al. 2017; Liu et al. 2020; Murphy et al. 2020). It is clear that forests improve the NFM potential through high water use, greater hydraulic roughness and canopy interception, and amelioration of soil structure (Wahren et al. 2012; Murphy et al. 2020). Consequently, the gross runoff volume and flow peak reduce with the increase in water losses and water storage capacity, and the time to runoff climax delays due to the slower velocity of surface flow and temporary runoff retention (Rogger et al. 2017; Jayapadma et al. 2022). Many studies have formed the common view that the establishment of afforestation on sparsely vegetated land can decrease the water yield and lead to lower and more delayed flood peaks (Brown et al. 2005; Brookhuis & Hein 2016; Rogger et al. 2017; Murphy et al. 2020). However, there are some reports stating the effects of reducing flooding by forests are weak, especially for the largest and most devastating floods (Aylward et al. 2005; Rogger et al. 2017). To identify and quantify the effectiveness of afforestation management practices in flood mitigation, paired catchment studies and scenario analysis based on hydrological models are used as major means of determining the magnitude of flooding changes resulting from forest cover (Brown et al. 2005; Adhami et al. 2019). For example, after reviewing paired catchment studies reported in the literature, Brown et al. (2005) concluded that forest cover causes a proportionally larger impact on low flows from the perspective of seasonal changes in water yield. Dittrich et al. (2019) suggested that afforestation as a sole NFM measure provides a positive net present value only in some cases. Furthermore, some scientists invested considerable efforts in identifying determinants of the cause–effect relationships between afforestation and flood control. By analyzing the decisive factors of the number of large floods reported since 1990, Ferreira & Ghimire (2012) suggested that the link between forest cover and reported flood frequency is not robust from the perspective of the country level and seems to be disturbed by sample selection and omitted variable bias. A study reported by Brookhuis & Hein (2016) demonstrated a nonlinear relationship between catchment's forest cover and the outcomes of the flood management practices. Wahren et al. (2012) drew the conclusion that the peak reduction for flood events varies from 3 to 70% and has a high relationship with the pre-event soil moisture. Tembata et al. (2020) first proposed that the type of forest is significant for flood mitigation based on a rigorous econometric analysis spanning numerous areas of interest.

However, there are few efforts toward the effects of spatial patterns of forests to mitigate floods. Rogger et al. (2017) believed that adopting the connectivity of flow paths and their spatial patterns as unifying themes can obtain major progress in determining causal mechanisms between land use change and floods. Besides, Brookhuis & Hein (2016) discovered that even small losses of forest cover can result in a significant increase in flood risks in the Trinidad case study. Murphy et al. (2020) highlighted that it is critical for land planners and policymakers to consider past and present management, and catchment characteristics, such as the type of soil and the location of catchment, in managing new NFM schemes to accomplish commensuration between public money and outcomes. Due to the complexity of the catchment hydrology process, it is important to use an effective methodology to study the impacts of land use change on storm runoff. Most researchers applied hydrological models in evaluating such impacts, including lumped, semi-distributed, and fully distributed hydrological models, as well as hydrology–hydrodynamics models (Semenova & Beven 2015). Siriwardena et al. (2006) investigated the relationship between natural forest cover reduction and runoff through a simple conceptual rainfall–runoff model. Notter et al. (2007) applied a semi-distributed model based on grid water balance to simulate the discharge change in a mesoscale catchment in Kenya. Fully distributed hydrological models with more complex processes, such as the European Hydrological System Model (MIKE-SHE) and soil and water assessment tool, have been widely used in assessing the impacts of land use on the hydrological process (Öztürk et al. 2013; Zhang et al. 2017). But above all, hydrological models perform poorly in simulating surface runoff processes accurately as the physical processes are not fully taken into account (Hou et al. 2018). To compensate for the deficiency of hydrological models in studying the effects of land use on runoff, Hou et al. (2018) applied a hydrodynamic-based numerical model to investigate quantitatively the impacts of land use on runoff under different rainfall scenarios. However, most studies used the scenario simulation method to investigate the impacts of land use on runoff, i.e., simulating the runoff results caused by different hypothetical land use change cases through numerical models. This approach may not obtain the optimal change pattern of land use, because all land use change cases cannot be simulated in research.

In this research, the major efforts are focused on identifying the relationship between the spatial location of afforestation and flood mitigation effectiveness. Contribution rate, defining the amount of runoff volume from a special sub-area to the flooding-impacted area, is quantified through the coupled hydrodynamic and rainfall-tracking model proposed by Wang et al. (2022). Then, the contribution rate is considered as an essential criterion to determine spatial prioritization of sub-areas conducting afforestation. Two study areas including a V-shaped catchment and a realistic catchment, named Baogaisi watershed, are applied to demonstrate the availability of the proposed method in implementing afforestation for mitigating flood regime at the catchment scale. Besides, the scenario analysis method generally used in traditional spatial priority practices of afforestation is compared with the proposed approach. The rest of this article is arranged as follows: Section 2 introduces a novel approach to determine spatial prioritization of afforestation management practices for mitigating flood regime at the catchment scale and study cases that are applied to demonstrate the effectiveness of the new method in Section 3; then, results and discussions are presented in detail in Sections 3 and 4, respectively; finally, brief conclusions are drawn in Section 5.

Listen
To determine the spatial prioritization of flooding mitigation management, a novel framework is designed based on the coupled hydrodynamic and rainfall-tracking model, which has been used in urban catchments to quantify how much individual areas contribute to the inundated area during a simulated event (Wang et al. 2022). In this article, the method is developed to advance the solution of the decision problem for identifying priority locations in afforestation and consists of four major steps (Figure 1), described as follows:
  • (i)

    Sub-area representation: Divide the study area into approximately equal individual areas.

  • (ii)

    Baseline scenario: Run the coupled model driven by equal rainfall across all cells to generate hydraulic characteristics including water depth and velocity, and value of rainfall mark variable (R).

  • (iii)

    Source identification: Extract contribution rates of all individual areas to impacted areas where afforestation management practices aim at mitigation.

  • (iv)

    Spatial prioritization determination: Determine the spatial prioritization of afforestation management practices by maximizing contribution rates of selected areas at the limitation of available land and public money.

Figure 1
Framework of determining spatial prioritization of afforestation management practices for mitigating flood regime.
View largeDownload slide

Framework of determining spatial prioritization of afforestation management practices for mitigating flood regime.

Figure 1
Framework of determining spatial prioritization of afforestation management practices for mitigating flood regime.
View largeDownload slide

Framework of determining spatial prioritization of afforestation management practices for mitigating flood regime.

Close modal

Coupled hydrodynamic and rainfall-tracking model

Listen

The coupled model developed by Wang et al. (2022) combines the 2D surface hydrodynamic model (Hou et al. 2013) with the rainfall-tracking model. Driven by data including digital elevation model (DEM), land use, sub-area division, and rainfall, detailed flooding dynamic characteristics (such as water depth, velocity, and so on) and contribution rates of individual areas to flooding areas are obtained. Moreover, the coupled model has been validated sufficiently against two idealized test cases including dissymmetric V-shaped and V-shaped catchments (Wang et al. 2022). Specifically, runoff contribution rates of various sub-areas to flooding were validated based on a dissymmetric V-shaped catchment, and the trajectory of runoff caused by rainfall on different sub-areas was validated by a V-shaped catchment.

2D Surface hydrodynamic model

Listen
The 2D surface hydrodynamic model couples hydrological with hydrodynamic processes, in which the hydrodynamic model is controlled by the 2D shallow water equations solved using the Godunov-type finite-volume method, which has been widely applied and validated in flooding simulations (Hou et al. 2021; Li et al. 2021; Wang et al. 2021). The governing equations are as follows:
(1)
(2)
(3)
where t is the time; x and y are the Cartesian coordinates; q denotes the vector of flow variables containing h, qx, and qy, which represent the water depth, unit-width discharges along the x- and y-directions, respectively; f and g denote the flux vector terms in the x- and y-directions, respectively; and S represents the source vector including net rainfall source R, slope source Sb and friction source Sf; is the net rainfall intensity; is the bed elevation; is the water density; and and are frictional stresses computed by the Manning equation:
(4)
where Cf is the roughness coefficient evaluated as follows:
(5)
where n is the Manning coefficient.

Rainfall-tracking model

Listen
The rainfall-tracking model is developed based on the hypothesis that water flows from all directions are mixed instantaneously and distributed uniformly in cells. Then the 2D surface hydrodynamic model provides robust and accurate flux exchange between cells. Thus, rainfall mark variable (R), which is initialized with zero or one in each cell (R equals to 1 represents the cell belonging to the marked area, and R = 0 is the contrary), is established as a marker to track the trajectory of runoff produced by the rainfall slumped down in individual areas. Calculation steps are as follows (Wang et al. 2022):
(6)
(7)
(8)
where i and j denote the computed and neighboring grids, respectively; t and Δt denote the time and time step, respectively; h and Δh denote the water depth and variation of the water depth in grid, respectively; R and ΔR denote the rainfall mark variable and variation of rainfall mark variable in grid, respectively; I denotes the net rainfall intensity.

2.2 Evaluation of spatial prioritization for flood mitigation management

Listen

The focus of this section is proposing a prioritized criterion for flooding mitigation management based on the hydrodynamic and source-tracking data. First, the location that the management practice intends to mitigate has to be defined, such as the outlet of watershed, reservoir in watershed, and so on. Then it is critical to determine the hydraulic element, like flow discharge, water volume, and so on, and the time to intervene. Further, the contribution rates of individual areas to the defined location at special time are quantified from the results simulated by the coupled model. Finally, spatial prioritizations of flooding mitigation management are arranged in accordance with contribution rates.

Study cases

Listen

In this section, two test cases, including a typical V-channel catchment and a realistic watershed, are applied to demonstrate the effectiveness of the developed approach in determining spatial prioritization of flooding mitigation management. Specially, the reduction effects are compared between novel schemes implemented in locations with maximum and minimum contribution rate and traditional schemes from upstream to downstream relative to the defined location. Besides, both of the two study areas have been validated by the previous authors based on the 2D surface hydrodynamic model (Hou et al. 2018; Liu et al. 2018).

V-shaped catchment

Listen
Overton & Brakensiek (1970) proposed the classic theoretical V-shaped catchment, as shown in Figure 2, and provided analytical hydrographs in side slope and channel. There are two symmetrical side slopes with a single slope of 0.05 along the vertical channel and a channel with a single slope of 0.02 in the middle of the catchment, and the minimum depth in the channel is 1 m, which in any case is sufficient to avoid backwater effects on the valley side (Overton & Brakensiek 1970). The validation of the condition has been tested by Hou et al. (2018) using the 2D surface hydrodynamic model. The aim of this section is to determine the spatial prioritization practicing afforestation to mitigate the peak discharge at the outlet of the channel. As shown in Figure 3, DEM and land use with a high-resolution grid of 5 m, sub-areas with 200 m × 200 m marked by mutually independent R described with continuous positive integer shown in Figure 3(a) and 3(b), and different return periods of rainfalls designed by Equation (9) (Hou et al. 2018) are used to run the coupled model. Besides, the Manning and infiltration coefficients are set according to Table 1.
(9)
where i represents rainfall intensity, mm/min; p is the return periods, year; and t represents rainfall duration, min.
Table 1

Coefficients for different land use (Hou et al. 2018)

Land useBare landChannelForest land
Infiltration rate (mm/h) 2.48 0 4.12 
Manning (s/m1/3) 0.044 0.02 0.2 
Land useBare landChannelForest land
Infiltration rate (mm/h) 2.48 0 4.12 
Manning (s/m1/3) 0.044 0.02 0.2 
View Large
Figure 2
Schematic of V-shaped catchment.
View largeDownload slide

Schematic of V-shaped catchment.

Figure 2
Schematic of V-shaped catchment.
View largeDownload slide

Schematic of V-shaped catchment.

Close modal
Figure 3
Input data of V-shaped catchment: (a) DEM, (b) land use, and (c) design rainfalls.
View largeDownload slide

Input data of V-shaped catchment: (a) DEM, (b) land use, and (c) design rainfalls.

Figure 3
Input data of V-shaped catchment: (a) DEM, (b) land use, and (c) design rainfalls.
View largeDownload slide

Input data of V-shaped catchment: (a) DEM, (b) land use, and (c) design rainfalls.

Close modal

Baogaisi catchment

Listen
The Baogaisi catchment (Figure 4), which is located in Liuyang City, Hunan Province, China, is applied to further demonstrate the effectiveness of the research in afforestation management practices. Moreover, the catchment area of the Baogaisi watershed is 22.1 km2 and the river length is 9.1 km, and besides, a reservoir is built in the downstream of the river; land use types include 94.6% of forest land, 1.8% of grassland, and 3.6% of others. The soil texture types include 46.2% of sandy loam and 53.8% of clay loam (Kang et al. 2020). The basin is located in a subtropical humid climate zone, with hot and humid summers and dry winters, and the average annual precipitation is about 1,569 mm. The validation of the rainfall–runoff event observed at Qingshui hydrologic station on 9 May 2012 year has been tested by Liu et al. (2018) using the 2D surface hydrodynamic model. The section is focused on the determination of spatial prioritization practicing afforestation to reduce the runoff volume at the reservoir. As shown in Figure 5, DEM and land use with a high-resolution grid of 10 m, and sub-areas with approximately 1 km × 1 km marked by mutually independent R described with continuous positive integer shown in Figure 5(a) and 5(b), to analyze the effects of the proposed approach in determining spatial prioritization of afforestation for Baogaisi catchment, land use type is assumed to be bare land. Different return periods of rainfalls designed by Equation (9) (Hou et al. 2018) are used to run the coupled hydrodynamic and rainfall-tracking model. Besides, the Manning and infiltration coefficients are set according to Table 1.
Figure 4
Location of Baogaisi catchment.
View largeDownload slide

Location of Baogaisi catchment.

Figure 4
Location of Baogaisi catchment.
View largeDownload slide

Location of Baogaisi catchment.

Close modal
Figure 5
Input data of Baogaisi catchment: (a) DEM, (b) land use, and (c) design rainfalls.
View largeDownload slide

Input data of Baogaisi catchment: (a) DEM, (b) land use, and (c) design rainfalls.

Figure 5
Input data of Baogaisi catchment: (a) DEM, (b) land use, and (c) design rainfalls.
View largeDownload slide

Input data of Baogaisi catchment: (a) DEM, (b) land use, and (c) design rainfalls.

Close modal
Listen

V-shaped catchment

Listen
The flow discharge at the outlet of the channel under different return periods is plotted in Figure 6. The time to reach flood peak is 75 min except for the condition of 2 years with 80 min, and the peak discharges increase as the return periods increase and are equal to 20.8, 36.1, 51.8, and 58.3 m3/s. Further, contribution rates of all sub-areas from numbers 1–21 to the flood peak at the outlet are extracted and illustrated in descending order in Figure 7. The sub-area 3 contributes the maximum of 10%, while the minimum is only 1.5%. Sub-areas that contribute the top five to peak discharge at the outlet are regions 3, 4, 2, 8, and 9, under all return periods of rainfalls, while the last five are regions 16, 11, 6, 17, and 21, respectively.
Figure 6
Flow discharges at outlet of channel under different return periods of rainfalls.
View largeDownload slide

Flow discharges at outlet of channel under different return periods of rainfalls.

Figure 6
Flow discharges at outlet of channel under different return periods of rainfalls.
View largeDownload slide

Flow discharges at outlet of channel under different return periods of rainfalls.

Close modal
Figure 7
Contribute rates of sub-areas to peak flow at outlet under different return periods of rainfalls.
View largeDownload slide

Contribute rates of sub-areas to peak flow at outlet under different return periods of rainfalls.

Figure 7
Contribute rates of sub-areas to peak flow at outlet under different return periods of rainfalls.
View largeDownload slide

Contribute rates of sub-areas to peak flow at outlet under different return periods of rainfalls.

Close modal
With reference to the land use scenarios simulated by Hou et al. (2018), as shown in Figure 8, cases 3–6 (Figure 8(c)–8(f)) are designed from downstream to upstream, and cases 1–2 are designed based on the proposed novel approach. Specifically, case 1 is planned according to the maximum gross contribution rate of forest land, while case 2 is planned based on minimum the gross contribution rate of the forest land. Moreover, gross contribution rates of planned sub-areas for different cases are shown in Table 2, and selected sub-areas in case 1 reach the maximum contribution rate of 42%, while case 2 contributes the minimum runoff volume with merely 9%, and gross contribution rates of cases 3–6 are 33, 27, 21, and 13%, which are between 9 and 42%. Six different planning scenarios for practicing afforestation management are simulated under the same boundaries. Modeling hydrographs resulting from different afforestation cases are plotted in Figure 9. It is noted that there are obvious differences in flow processes and peak flow between different cases. The peak discharge of case 1 maintains the minimum and that of case 3 maintains the maximum under all return periods, and the second of the reduction effects is case 5. For the lagging time of peak discharge, cases 1, 3, 4, and 5 have the same impact, and cases 2 and 6 have the same effect, and the former is superior to the latter. Comparisons of peak flow, such as reducing rates of peak discharge and lagging times of flow peak, between afforestation scenarios and pre-afforestation conditions are quantified in Table 3 to describe the causal relationship between gross contribution rates of afforested areas and effects of flooding mitigation. For the reduction rates of peak discharge, under all rainfall events, case 1, in which the contribution rate is the maximum, has the best mitigation effect. While the reduction effect of other cases performs inconsistently with the contribution rate, for example, case 2 with a minimum contribution rate decreases more runoff than case 3, and case 5 with a 21% contribution rate reduces the second most peak discharge. For the lagging time of flow peak, the differences of reduction for different cases are various when the return period is less than 50 years, and case 1 maintains the maximum effect. While for other cases, there is no consistent pattern for the delay time with contribution rate. As shown clearly in Figure 10, not only the reducing rate of peak discharge but also the lagging time of flow peak, case 1 plotted in red performs the best flooding mitigation management, except for the lagging time at 100 years. For all cases, the effects of flood mitigation decrease as return periods increase except for case 2. For the same return period, however, there is no robust positive correlation between gross contribution rates and effects of flood mitigation, except for case 1 that contributes the maximum to the flood peak.
Table 2

Gross contribution rates of planned sub-areas in V-shaped catchment

Case 1Case 2Case 3Case 4Case 5Case 6
2 years 44 9 35 27 21 11 
10 years 42 9 34 28 21 12 
50 years 41 9 32 27 22 14 
100 years 41 9 31 27 22 15 
Mean value 42 9 33 27 21 13 
Case 1Case 2Case 3Case 4Case 5Case 6
2 years 44 9 35 27 21 11 
10 years 42 9 34 28 21 12 
50 years 41 9 32 27 22 14 
100 years 41 9 31 27 22 15 
Mean value 42 9 33 27 21 13 
View Large
Table 3

Comparison of flood mitigating effects between different afforestation cases

Case 1Case 2Case 3Case 4Case 5Case 6
Reducing rates of peak discharge under different cases (%) 
 2 years 54 19 13 28 46 22 
 10 years 43 20 8 17 33 20 
 50 years 35 9 10 14 26 19 
 100 years 31 17 11 9 23 18 
Lagging times of flow peak under different cases (min) 
 2 years 25 0 10 10 0 0 
 10 years 20 5 10 10 10 0 
 50 years 10 0 10 5 10 0 
 100 years 5 0 10 5 5 0 
Case 1Case 2Case 3Case 4Case 5Case 6
Reducing rates of peak discharge under different cases (%) 
 2 years 54 19 13 28 46 22 
 10 years 43 20 8 17 33 20 
 50 years 35 9 10 14 26 19 
 100 years 31 17 11 9 23 18 
Lagging times of flow peak under different cases (min) 
 2 years 25 0 10 10 0 0 
 10 years 20 5 10 10 10 0 
 50 years 10 0 10 5 10 0 
 100 years 5 0 10 5 5 0 
View Large
Figure 8
Different cases of afforestation in V-shaped catchment: (a) case 1, (b) case 2, (c) case 3, (d) case 4, (e) case 5, (f) case 6.
View largeDownload slide

Different cases of afforestation in V-shaped catchment: (a) case 1, (b) case 2, (c) case 3, (d) case 4, (e) case 5, (f) case 6.

Figure 8
Different cases of afforestation in V-shaped catchment: (a) case 1, (b) case 2, (c) case 3, (d) case 4, (e) case 5, (f) case 6.
View largeDownload slide

Different cases of afforestation in V-shaped catchment: (a) case 1, (b) case 2, (c) case 3, (d) case 4, (e) case 5, (f) case 6.

Close modal
Figure 9
Simulated hydrographs resulting from different afforestation cases.
View largeDownload slide

Simulated hydrographs resulting from different afforestation cases.

Figure 9
Simulated hydrographs resulting from different afforestation cases.
View largeDownload slide

Simulated hydrographs resulting from different afforestation cases.

Close modal
Figure 10
Comparison of flood mitigating effects between different afforestation cases: (a) reducing rates of peak dishcarge and (b) lagging times of flow peak.
View largeDownload slide

Comparison of flood mitigating effects between different afforestation cases: (a) reducing rates of peak dishcarge and (b) lagging times of flow peak.

Figure 10
Comparison of flood mitigating effects between different afforestation cases: (a) reducing rates of peak dishcarge and (b) lagging times of flow peak.
View largeDownload slide

Comparison of flood mitigating effects between different afforestation cases: (a) reducing rates of peak dishcarge and (b) lagging times of flow peak.

Close modal

Baogaisi catchment

Listen
Figure 11 plots the water depth in the reservoir caused by various return periods of rainfall events with a simulated duration of 10 h. The runoff volume into the reservoir increases as the return period increases and reaches 1.07, 4.15, 6.22, and 6.89 m. Furthermore, contribution rates of all sub-areas from number 1 to 31 to runoff volume are extracted and illustrated in descending order in Figure 12. Considering the special distribution of contribution rates under a 2-year return period caused by DEM which reserves a hump before the reservoir, thus the following results are analyzed excluding the 2-year return period. The sub-area 11 contributes the maximum with 6%, while the minimum is close to 0%. While under a 2-year return period, the water flow into the reservoir mainly comes from the neighboring sub-areas, including regions 11, 6, 5, 10, 2, and 11. As rainfall volume increases, the runoff from upstream is able to cross the bump into reservoir, such as sub-area 14.
Figure 11
Water depth at reservoir under different return periods of rainfalls.
View largeDownload slide

Water depth at reservoir under different return periods of rainfalls.

Figure 11
Water depth at reservoir under different return periods of rainfalls.
View largeDownload slide

Water depth at reservoir under different return periods of rainfalls.

Close modal
Figure 12
Contribute rates of sub-areas to runoff volume at reservoir under different return periods of rainfalls.
View largeDownload slide

Contribute rates of sub-areas to runoff volume at reservoir under different return periods of rainfalls.

Figure 12
Contribute rates of sub-areas to runoff volume at reservoir under different return periods of rainfalls.
View largeDownload slide

Contribute rates of sub-areas to runoff volume at reservoir under different return periods of rainfalls.

Close modal
Figure 13 describes the arrangement of afforestation for different cases designed based on the maximum and minimum contribution rates (cases 1 and 2), cases 3–5 are planned from downstream to upstream (cases 3–5), and gross contribution rates of planned sub-areas for cases 1–5 are 23, 3, 7, 13, and 19%, respectively (Table 4). Rainfall–runoff processes of different afforestation scenarios are modelled under the same input as the pre-afforestation scenario except for the land use change. Simulated hydrographs resulting from different afforestation cases are shown in Figure 14. Results show that cases 1 and 5 have similar water depths, and other cases such as cases 2–4 have similar water depths; the differences among the five cases decrease as the return period increases, and when the 50-year return period is exceeded, results indicate that afforestation has few mitigation effects for catchment. The mitigated effects of different afforestation cases on runoff volume are quantified and plotted in Table 5 and Figure 15. When the return period is less than 50 years, it is clear that the effects of afforestation and spatial prioritization of afforestation to reduce runoff volume decrease as the return period increases. However, under large rainfall events such as 50-year and 100-year, these effects are insignificant and irrelevant to the contribution rates of planned areas. For instance, case 1 reduces the water depth with 0.29, 0.22, and 0.06 under 2-, 10-, and 50-year rainstorm scenarios, while the mitigation effect for 100 years is the same as for 50 years. Under the small and moderate rainfall events such as 2 and 10 years, mitigating effects of afforestation are evident, and the effects are positively related to contribution rates of planned areas. The water depth in the reservoir is reduced by 0.08–0.29 and 0.06–0.22 m, respectively. For example, case 1 achieves the maximum mitigation benefit with 0.29 m under the 2-year return period, while case 2 reduces the water depth with only 0.08 m, and mitigating effects of cases 3–5 reach 0.11, 0.09, and 0.25, respectively. It is clear that mitigating effects of other cases are between case 1 (maximum contribution rate) and case 2 (minimum contribution rate).
Table 4

Gross contribution rates of planned sub-areas in Baogaisi catchment

Case 1Case 2Case 3Case 4Case 5
10 years 23 3 6 14 18 
50 years 23 3 7 13 19 
100 years 24 3 7 13 19 
Mean value 23 3 7 13 19 
Case 1Case 2Case 3Case 4Case 5
10 years 23 3 6 14 18 
50 years 23 3 7 13 19 
100 years 24 3 7 13 19 
Mean value 23 3 7 13 19 
View Large
Table 5

Comparison of flood mitigating effects between different afforestation cases (unit: m)

Case 1Case 2Case 3Case 4Case 5
2 years 0.29 0.08 0.11 0.09 0.25 
10 years 0.22 0.06 0.09 0.07 0.19 
50 years 0.06 0.04 0.06 0.04 0.05 
100 years 0.06 0.04 0.05 0.04 0.05 
Case 1Case 2Case 3Case 4Case 5
2 years 0.29 0.08 0.11 0.09 0.25 
10 years 0.22 0.06 0.09 0.07 0.19 
50 years 0.06 0.04 0.06 0.04 0.05 
100 years 0.06 0.04 0.05 0.04 0.05 
View Large
Figure 13
Different cases of afforestation in Baogaisi catchment: (a) case 1, (b) case 2, (c) case 3, (d) case 4, and (e) case 5.
View largeDownload slide

Different cases of afforestation in Baogaisi catchment: (a) case 1, (b) case 2, (c) case 3, (d) case 4, and (e) case 5.

Figure 13
Different cases of afforestation in Baogaisi catchment: (a) case 1, (b) case 2, (c) case 3, (d) case 4, and (e) case 5.
View largeDownload slide

Different cases of afforestation in Baogaisi catchment: (a) case 1, (b) case 2, (c) case 3, (d) case 4, and (e) case 5.

Close modal
Figure 14
Simulated hydrographs resulting from different afforestation cases.
View largeDownload slide

Simulated hydrographs resulting from different afforestation cases.

Figure 14
Simulated hydrographs resulting from different afforestation cases.
View largeDownload slide

Simulated hydrographs resulting from different afforestation cases.

Close modal
Figure 15
Comparison of flood mitigating effects between different afforestation cases.
View largeDownload slide

Comparison of flood mitigating effects between different afforestation cases.

Figure 15
Comparison of flood mitigating effects between different afforestation cases.
View largeDownload slide

Comparison of flood mitigating effects between different afforestation cases.

Close modal
Listen

To practice flood management from a catchment perspective and determine spatial prioritization of afforestation management practices for mitigating flood regimes, insights should be into the flood dynamics at the catchment scale (Vercruysse et al. 2019). A coupled hydrodynamic and rainfall-tracking model is applied to reproduce complex hydrodynamic processes of floods and trace outflow pathways. Contribution rates of individual areas to impacted areas are obtained to guide spatial prioritization of afforestation management practices. Traditional mitigating flood regimes that generally implement afforestation from upstream to downstream (Hou et al. 2018) are compared with the method proposed in this article to determine spatial prior locations based on contribution rates to floods.

According to the aforementioned results, the coupled model is able to obtain the hydrodynamic and rainfall-tracing information simultaneously. The hydrodynamic information can quantify the water depth and velocity within each computing grid, and rainfall-tracing information can quantify the contribution rate for different source areas to water depth within each computing grid. For the hydrodynamic results (Figures 5 and 10), both the discharge of the V-shaped catchment and water depth of the Baogaisi catchment increase with the increasing return period. For the process of water depth in the Baogaisi catchment, there are two distinct stages of increase, and the second step increases more rapidly than the first. The reason is that there is a topographical barrier in front of the reservoir, and thus, more runoff from the upstream area enters the reservoir when the water depth exceeds the elevation of the barrier. For rainfall-tracing results (Figures 6 and 11), the spatial distribution of contribution rates is relatively constant under different return periods of rainfall events, which means that the spatial prioritization is primarily dependent on hydraulic and hydrological characteristics of the underlying surface. The characteristics of rainfall events can cause some effects to the spatial prioritization. For flood mitigation results (Figures 9 and 14), not only peak discharge but also runoff volume, effects of flood mitigation for case 1 designed as the afforestation regime with maximum contribution rate both in two catchments are the best under all rainfall events, and this suggests that it is effective for the proposed approach to determine spatial prioritization of afforestation management practices for mitigating flood regime at the catchment scale. Besides, considering the case study of the Baogaisi catchment, the relationship of flood mitigation effect on runoff volume in the reservoir with contribution rates of cases is clearly illustrated as positively correlated in Figure 14. However, there is no obvious correlation between flood mitigation effects and the contribution rate of cases for the case study of the V-shaped catchment, except for case 1. The possible reason may be that the coupled hydrodynamic and rainfall-tracking model only considers the contribution of sub-areas from the perspective of runoff generation without consideration of the runoff confluence process, and the other reason may be that the coupled model is more suitable in the analysis of runoff volume. Finally, these results suggest that the afforestation is an effective mitigating flood regime and the mitigation effects have a strong relation with spatial locations under small to moderate floods, but there are few influences under large rainfall events. To further illustrate the study, the results of this article are compared with similar research studies. The study by Hou et al. (2018) revealed that land use could considerably influence the rainfall-flood process and varies with the catchment terrain, land use scenario, and the rainfall events; and Brookhuis & Hein (2016) demonstrated a nonlinear relationship between the catchment's forest cover and the generation of the flood control service. These results can be proven well by different sub-areas having different contribution rates to flood.

Listen

In this article, a novel approach to determine spatial prioritization of afforestation management practices for mitigating flood regimes at the catchment scale is proposed based on the coupled hydrodynamic and rainfall-tracking model. With the comparison of the traditional method, finding the novel approach based on contribution rate could determine the optimal sub-area to implement afforestation.

  • The coupled model is able to obtain accurate hydrodynamic and rainfall-tracing information simultaneously within each computing grid under the flooding caused by rainfall events.

  • Compared to the results of V-shaped Baogaisi catchment, there is a positive correlation between the contribution rate of afforestation area and the reduction effect of runoff volume, but no correlation with the reduction effect of peak discharge.

  • The afforestation case with the maximum contribution rate has the greatest reduction effect in peak discharge and runoff volume, which illustrates that the novel approach is able to determine spatial prioritization of afforestation management practices.

Further, the proposed method performs well in determining spatial prioritization of afforestation for mitigating flood volume while remaining uncertain for mitigating flood peak. In the future, we will be focused on improving the coupled model and research to enable it to be more effective in identifying spatial prioritization of afforestation management practices from the perspective of mitigating flood peaks at the catchment scale.

Listen

Conceptualization and methodology: J. Hou, X. Wang; writing – original draft preparation: X. Wang; material preparation and collection and analysis: X. Wang, X. Pan, and G. Chen; supervision: X. Gao; funding acquisition: J. Hou.

Listen

This work was partly supported by the National Natural Science Foundations of China (Nos. 52079106 and 52009104), the Sino-German Mobility Program (No. M-0427), the Shaanxi Province Innovation Talent Promotion Plan Project Technology Innovation Team (No. 2020TD-023), and the Natural Science Foundations of Shaanxi Province (No. 2021SF-484).

Listen

All relevant data are included in the paper or its Supplementary Information.

Listen

The authors declare there is no conflict.

Adhami
M.
,
Sadeghi
S. H.
,
Duttmann
R.
 &
Sheikhmohammady
M.
2019
Changes in watershed hydrological behavior due to land use comanagement scenarios
.
Journal of Hydrology
577
.
doi: 10.1016/j.jhydrol.2019.124001
.
Google Scholar
 
Aylward
F. A. O. B.
,
Brooks
K. N.
,
Bryon
N.
,
Calder
I. R.
,
Cambell
B.
,
Dugan
P. C.
 &
Upadhya
K.
2005
Forests and Floods Drowning in Fiction or Thriving on Facts?
RAP Publication 2005/03
.
Bevacqua
E.
,
Maraun
D.
,
Vousdoukas
M. I.
,
Voukouvalas
E.
,
Vrac
M.
,
Mentaschi
L.
 &
Widmann
M.
2019
Higher probability of compound flooding from precipitation and storm surge in Europe under anthropogenic climate change
.
Science Advances
5
(
9
),
eaaw5531
.
Google Scholar
Crossref
Search ADS
PubMed
 
Bradshaw
C. J. A.
,
Sodhi
N. S.
,
Peh
K. S. H.
 &
Brook
B. W.
2007
Global evidence that deforestation amplifies flood risk and severity in the developing world
.
Global Change Biology
13
(
11
),
2379
–
2395
.
doi: 10.1111/j.1365-2486.2007.01446.x
.
Google Scholar
Crossref
Search ADS
 
Brookhuis
B. J.
 &
Hein
L. G.
2016
The value of the flood control service of tropical forests: A case study for Trinidad
.
Forest Policy and Economics
62
,
118
–
124
.
doi: 10.1016/j.forpol.2015.10.002
.
Google Scholar
Crossref
Search ADS
 
Brown
A. E.
,
Zhang
L.
,
McMahon
T. A.
,
Western
A. W.
 &
Vertessy
R. A.
2005
A review of paired catchment studies for determining changes in water yield resulting from alterations in vegetation
.
Journal of Hydrology
310
(
1–4
),
28
–
61
.
doi: 10.1016/j.jhydrol.2004.12.010
.
Google Scholar
 
Dadson
S. J.
,
Hall
J. W.
,
Murgatroyd
A.
,
Acreman
M.
,
Bates
P.
,
Beven
K.
,
Heathwaite
L.
,
Holden
J.
,
Holman
I. P.
,
Lane
S. N.
,
O'Connell
E.
,
Penning-Rowsell
E.
,
Reynard
N.
,
Sear
D.
,
Thorne
C.
 &
Wilby
R.
2017
A restatement of the natural science evidence concerning catchment-based ‘natural’ flood management in the UK
.
Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences
473
(
2199
),
20160706
.
doi: 10.1098/rspa.2016.0706
.
Google Scholar
Crossref
Search ADS
 
Dittrich
R.
,
Ball
T.
,
Wreford
A.
,
Moran
D.
 &
Spray
C. J.
2019
A cost-benefit analysis of afforestation as a climate change adaptation measure to reduce flood risk
.
Journal of Flood Risk Management
12
(
4
),
e12482
.
Google Scholar
Crossref
Search ADS
 
Ferreira
S.
 &
Ghimire
R.
2012
Forest cover, socioeconomics, and reported flood frequency in developing countries
.
Water Resources Research
48
(
8
),
0
–
0
.
doi: 10.1029/2011wr011701
.
Google Scholar
Crossref
Search ADS
 
Hall
J.
,
Arheimer
B.
,
Borga
M.
,
Brázdil
R.
,
Claps
P.
,
Kiss
A.
,
Kjeldsen
T. R.
,
Kriaučiūnienė
J.
,
Kundzewicz
Z. W.
,
Lang
M.
,
Llasat
M. C.
,
Macdonald
N.
,
McIntyre
N.
,
Mediero
L.
,
Merz
B.
,
Merz
R.
,
Molnar
P.
,
Montanari
A.
,
Neuhold
C.
,
Parajka
J.
,
Perdigão
R. A. P.
,
Plavcová
L.
,
Rogger
M.
,
Salinas
J. L.
,
Sauquet
E.
,
Schär
C.
,
Szolgay
J.
,
Viglione
A.
 &
Blöschl
G.
2014
Understanding flood regime changes in Europe: A state-of-the-art assessment
.
Hydrology and Earth System Sciences
18
(
7
),
2735
–
2772
.
doi: 10.5194/hess-18-2735-2014
.
Google Scholar
Crossref
Search ADS
 
Hou
J.
,
Liang
Q.
,
Simons
F.
 &
Hinkelmann
R.
2013
A 2D well-balanced shallow flow model for unstructured grids with novel slope source term treatment
.
Advances in Water Resources
52
,
107
–
131
.
doi: 10.1016/j.advwatres.2012.08.003
.
Google Scholar
Crossref
Search ADS
 
Hou
J. M.
,
Guo
K. H.
,
Liu
F. F.
,
Han
H.
,
Liang
Q. H.
,
Tong
Y.
 &
Li
P.
2018
Assessing slope forest effect on flood process caused by a short-Duration storm in a small catchment
.
Water
10
(
9
).
doi: 10.3390/w10091256
.
Google Scholar
 
Hou
J. M.
,
Ma
Y. Y.
,
Wang
T.
,
Li
B. Y.
,
Li
X.
,
Wang
F.
,
Jin
S. L.
 &
Ma
H. L.
2021
A river channel terrain reconstruction method for flood simulations based on coarse DEMs
.
Environmental Modelling & Software
140
.
doi: 10.1016/j.envsoft.2021.105035
.
Google Scholar
 
Jayapadma
J.
,
Wickramaarachchi
T. N.
,
Silva
G.
,
Ishidaira
H.
 &
Magome
J.
2022
Coupled hydrodynamic modelling approach to assess land use change induced flood characteristics
.
Environmental Monitoring and Assessment
194
(
5
),
354
.
doi: 10.1007/s10661-022-09986-7
.
Google Scholar
Crossref
Search ADS
PubMed
 
Kang
J.
,
Wang
J.
,
Ni
H.
 &
Kang
P.
2020
Study on two-dimensional numerical simulation of rainstorm and torrent in small watershed based on lidar data
.
Journal of Hydrologic Engineering
25
(
9
).
doi: 10.1061/(asce)he.1943-5584.0001998
.
Google Scholar
 
Li
B. Y.
,
Hou
J. M.
,
Li
D. L.
,
Yang
D.
,
Han
H.
,
Bi
X.
,
Wang
X. H.
,
Hinkelmann
R.
 &
Xia
J. Q.
2021
Application of LiDAR UAV for high-resolution flood modelling
.
Water Resources Management
35
(
5
),
1433
–
1447
.
doi: 10.1007/s11269-021-02783-w
.
Google Scholar
Crossref
Search ADS
 
Liu
F.
,
Hou
J.
,
Guo
K.
,
Li
D.
,
Xu
S.
 &
Zhang
X.
2018
High-performance numerical model for rainfall catchment process based on hydrodynamic method
.
Chinese Journal of Hydrodynamics
33
(
6
),
778
–
785
.
doi: 10.16076/j.cnki.cjhd.2018.06.015
.
Google Scholar
 
Liu
D.
,
Fan
Z.
,
Fu
Q.
,
Li
M.
,
Faiz
M. A.
,
Ali
S.
,
Li
T.
,
Zhang
L.
 &
Khan
M. I.
2020
Random forest regression evaluation model of regional flood disaster resilience based on the whale optimization algorithm
.
Journal of Cleaner Production
250
.
doi: 10.1016/j.jclepro.2019.119468
.
Google Scholar
 
Merz
B.
,
Aerts
J.
,
Arnbjerg-Nielsen
K.
,
Baldi
M.
,
Becker
A.
,
Bichet
A.
,
Blöschl
G.
,
Bouwer
L. M.
,
Brauer
A.
,
Cioffi
F.
,
Delgado
J. M.
,
Gocht
M.
,
Guzzetti
F.
,
Harrigan
S.
,
Hirschboeck
K.
,
Kilsby
C.
,
Kron
W.
,
Kwon
H. H.
,
Lall
U.
,
Merz
R.
,
Nissen
K.
,
Salvatti
P.
,
Swierczynski
T.
,
Ulbrich
U.
,
Viglione
A.
,
Ward
P. J.
,
Weiler
M.
,
Wilhelm
B.
 &
Nied
M.
2014
Floods and climate: Emerging perspectives for flood risk assessment and management
.
Natural Hazards and Earth System Sciences
14
(
7
),
1921
–
1942
.
doi: 10.5194/nhess-14-1921-2014
.
Google Scholar
Crossref
Search ADS
 
Murphy
T. R.
,
Hanley
M. E.
,
Ellis
J. S.
 &
Lunt
P. H.
2020
Native woodland establishment improves soil hydrological functioning in UK upland pastoral catchments
.
Land Degradation & Development
32
(
2
),
1034
–
1045
.
doi: 10.1002/ldr.3762
.
Google Scholar
Crossref
Search ADS
 
Notter
B.
,
Macmillan
L.
,
Viviroli
D.
,
Weingartner
R.
 &
Liniger
H.-P.
2007
Impacts of environmental change on water resources in the Mt. Kenya region
.
Journal of Hydrology
343
(
3–4
),
266
–
278
.
Google Scholar
 
Overton
D. E.
 &
Brakensiek
D. L.
1970
Kinematic model of surface runoff response
. In:
Proceedings of the Wellington Symposium
.
Unesco/IAHS
,
Paris, France
, pp.
100
–
112
.
Google Scholar
 
Öztürk
M.
,
Copty
N. K.
 &
Saysel
A. K.
2013
Modeling the impact of land use change on the hydrology of a rural watershed
.
Journal of Hydrology
497
,
97
–
109
.
doi: 10.1016/j.jhydrol.2013.05.022
.
Google Scholar
Crossref
Search ADS
 
Rogger
M.
,
Agnoletti
M.
,
Alaoui
A.
,
Bathurst
J. C.
,
Bodner
G.
,
Borga
M.
,
Chaplot
V.
,
Gallart
F.
,
Glatzel
G.
,
Hall
J.
,
Holden
J.
,
Holko
L.
,
Horn
R.
,
Kiss
A.
,
Kohnova
S.
,
Leitinger
G.
,
Lennartz
B.
,
Parajka
J.
,
Perdigao
R.
,
Peth
S.
,
Plavcova
L.
,
Quinton
J. N.
,
Robinson
M.
,
Salinas
J. L.
,
Santoro
A.
,
Szolgay
J.
,
Tron
S.
,
van den Akker
J. J. H.
,
Viglione
A.
 &
Bloschl
G.
2017
Land use change impacts on floods at the catchment scale: Challenges and opportunities for future research
.
Water Resources Research
53
(
7
),
5209
–
5219
.
doi: 10.1002/2017WR020723
.
Google Scholar
Crossref
Search ADS
PubMed
 
Roy
P.
,
Pal
S. C.
,
Chakrabortty
R.
,
Chowdhuri
I.
,
Malik
S.
 &
Das
B.
2020
Threats of climate and land use change on future flood susceptibility
.
Journal of Cleaner Production
272
.
doi: 10.1016/j.jclepro.2020.122757
.
Google Scholar
 
Semenova
O.
 &
Beven
K.
2015
Barriers to progress in distributed hydrological modelling
.
Hydrological Processes
29
(
8
),
2074
–
2078
.
doi: 10.1002/hyp.10434
.
Google Scholar
Crossref
Search ADS
 
Siriwardena
L.
,
Finlayson
B. L.
 &
Mcmahon
T. A.
2006
The impact of land use change on catchment hydrology in large catchments: The comet river, central Queensland, Australia
.
Journal of Hydrology
326
(
1–4
),
199
–
214
.
Google Scholar
 
Tembata
K.
,
Yamamoto
Y.
,
Yamamoto
M.
 &
Matsumoto
K.
2020
Don't rely too much on trees: Evidence from flood mitigation in China
.
Science of the Total Environment
732
,
138410
.
doi: 10.1016/j.scitotenv.2020.138410
.
Google Scholar
Crossref
Search ADS
PubMed
 
Vercruysse
K.
,
Dawson
D. A.
,
Glenis
V.
,
Bertsch
R.
,
Wright
N.
 &
Kilsby
C.
2019
Developing spatial prioritization criteria for integrated urban flood management based on a source-to-impact flood analysis
.
Journal of Hydrology
578
.
doi: 10.1016/j.jhydrol.2019.124038
.
Google Scholar
 
Wahren
A.
,
Schwärzel
K.
 &
Feger
K. H.
2012
Potentials and limitations of natural flood retention by forested land in headwater catchments: Evidence from experimental and model studies
.
Journal of Flood Risk Management
5
(
4
),
321
–
335
.
doi: 10.1111/j.1753-318X.2012.01152.x
.
Google Scholar
Crossref
Search ADS
 
Wang
N.
,
Hou
J. M.
,
Du
Y. G.
,
Jing
H. X.
,
Wang
T.
,
Xia
J. Q.
,
Gong
J. H.
 &
Huang
M. S.
2021
A dynamic, convenient and accurate method for assessing the flood risk of people and vehicle
.
Science of the Total Environment
797
.
doi: 10.1016/j.scitotenv.2021.149036
.
Google Scholar
 
Wang
X.
,
Hou
J.
,
Li
D.
,
Li
B.
,
Gao
X.
 &
Liu
Y.
2022
A coupled hydrodynamic and rainfall-tracking model for source-to-impact analysis in urban inundation
.
Water Resources Management
36 (14), 5581–5598. doi: 10.1007/s11269-022-03319-6
.
Google Scholar
 
Zhang
L.
,
Karthikeyan
R.
,
Bai
Z.
 &
Srinivasan
R.
2017
Analysis of streamflow responses to climate variability and land use change in the loess plateau region of China
.
Catena
154, 1–11. doi: 10.1016/j.catena.2017.02.012
.
Google Scholar
 
© 2024 The Authors
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).