Laboratory comparison of vegetation and gravel on parameters affecting urban floods under different rainfall and runoff conditions

Initial rainfall losses depend on various factors such as soil permeability, rainfall intensity, previous soil moisture conditions, land use, slope, type, and density of vegetation and soil texture (Liu et al. 2021). Generally, specifying the amount of rainfall and its possible losses in different pavements and considering the aforementioned effective factors help us to reduce the volume of floods and the costs of flood damage and erosion (Bichai & Ashbolt 2017). Furthermore, the accurate estimation of the runoff threshold can not only lead to more accurate estimates of the design flood and reduce the costs of flood protection and damage, but it is also important in the optimal use and management of rainfall (Ghazvinian & Karami 2023a; Karami et al. 2023). Regarding the lack of direct runoff measurement data, rainfall and storm intensity data are used to estimate the amount and intensity of floods (Ebrahimi et al. 2015). At the beginning of the conversion of rainfall into runoff, some of the rainfall is absorbed by the soil or penetrates deep into the ground, and the rest becomes surface runoff (Ebrahimi et al. 2015; Saber et al. 2019). After determining the threshold for the start of runoff in each region, the use of biological methods allows for taking the necessary measures and preventing the transformation of runoff into floods with correct estimation and management, and engineering operations (Saber et al. 2019).

By providing outstanding and unique design features, low impact development (LID) methods can reduce the effects of urban runoff and climate change (Saadat Foomani & Malekmohammadi 2020). LID can restore the hydrology of urban areas to pre-development functions by using flood control and natural hydrological features (Zhao & Meng 2020). LID helps to achieve the goal of sustainable development, as it promotes effective urban flood management (Luan et al. 2017). Infiltration trenches, vegetated filters, and bioretention basins are three best low impact development management practices (LID-BMP) that can provide nature-based solutions to control runoff volume, runoff onset time, PF, and removal of stormwater pollutants. Sun et al. (2011), Davis (2008), Muha et al. (2014), and Mangangka et al. (2015) discussed the appropriate efficiency of bioretention basins in flood control. As part of nature, turfs are low-cost structures. By using turfs, the total volume of runoff can be reduced by infiltration and storage (Ghazvinian & Karami 2023b). The effectiveness of grass in investigating the process of flood and runoff control has been investigated by Green et al. (2021) and Hamel & Tan (2022). Also, the gravel structure of the infiltration trench can reduce the volume of surface runoff and the amount of peak discharge and increase the groundwater supply. Meena & Gupta (2018) and Kumar et al. (2022) investigated the effectiveness of the infiltration trench in controlling the volume of runoff and the hydrological parameters affecting the runoff.

The field research reviewed in the following has investigated the relationship between rainfall and runoff, as well as the appropriate effectiveness of LIDs such as turf and infiltration trenches in runoff reduction. Arnaez et al. (2007) studied the factors affecting runoff by creating artificial rainfall in gentle slopes. To this aim, 22 rainfalls with an intensity of 30–117 mm/h were applied on a loam texture with surface and non-homogeneous gravel pavement, and the values of surface runoff were equal to 7.2–104 mm/h. In addition to the rainfall intensity, they concluded that the kinetic energy of the cloudburst, the resistance of soil particles to the impact of raindrops, the amount of vegetation, and the slope can affect the amount of runoff and its threshold.

By studying the effects of rainfall intensity with values of 0.67, 1, and 1.67 mm/h and slopes of 9, 27, and 36% on spring corn vegetation, Mu et al. (2015) concluded that an increase in rainfall intensity and the degree of slope increase the runoff coefficient. Alyaseri & Zhou (2016) conducted a field study on the effect of different types of permeable pavements on urban runoff. In this research, surfaces paved with permeable concrete, permeable asphalt, and paving could reduce runoff by 12.5, 9.6, and 7.9%, respectively. dos Santos et al. (2017) investigated the effects of the characteristics of 176 rainfall events, including intensity, duration, and frequency on runoff in three watersheds in Brazil under different uses. They stated that the rainfall structure can greatly affect the changes in surface runoff.

Alyaseri et al. (2017) conducted a field-scale study on the effect of rain gardens on reducing stormwater in combined sewages in a densely populated urban area. Compared to conditions before the installation of green infrastructure, rain gardens have resulted in a 76% reduction in stormwater runoff in sewers. Sun et al. (2019) investigated the effect of site and precipitation patterns on the performance of bioretention cell (BRC). The results showed that although the small and medium rainfall events were dominant, they contributed less to the total rainfall depth than the large rainfall events. The ratio of runoff coefficient to imperviousness can be used as an indicator to explain why BRCs perform differently with the same design strategy under the same rainfall events. Rainfall patterns had significant impacts on the hydrologic performance of BRCs by influencing the overflow and underdrain flow. BRCs performed better for rainfall events with a longer duration and lower rainfall intensity because they generated smoother runoff processes into the BRCs.

Song et al. (2021) used the geocell structure to strengthen the soil to grow vegetation and improve slope stability. The composition of geocell and soil was investigated in two small and large sizes and two slopes under three different rainfall intensities. The results revealed the proper efficiency of the geocell. Demirezen & Kazezyılmaz-Alhan (2022) investigated the hydrological performance of the infiltration trench in the laboratory dimension using a rainfall simulator. Based on the results, rainfall intensity, gravel size, and the distance between the pavement and the runoff outlet can significantly affect the magnitude and delay time of the hydrograph and runoff velocity. Alyaseri et al. (2023) addressed the practical application of green infrastructure for runoff management. Conducting experiments on green infrastructures reported a reduction of the runoff volume by 62%. Furthermore, researchers stated the consistency of the amount of flow reduction to the findings of other related studies, although the overall percentage of runoff reduction from this study was lower than the reduction percentage reported in many other studies.

According to research, vegetation and infiltration trenches are effective in controlling floods and improving hydrological parameters. However, to the best of our knowledge, few studies have been conducted on the effect of rainfall with very high intensity such as 200 mm/h, or the combined effect of rainfall and runoff on the pavements. In this regard, this research seeks to experimentally investigate the hydrological parameters by creating vertical rainfall, horizontal runoff, and the combination of rainfall and runoff using a rainfall simulator on the treatments of test control (O), sandy loam soil (SL), gravel (G), gravel with a geocell (GGE) layer, rosemary (R), rosemary with a geocell layer (RGE), turf (T), and turf with a geocell layer (TGE) in a 5% slope. The hydrological parameters of time to start runoff (TR), time to end runoff (TER), time to peak (TP), time base (TB), C, and PF rate were compared in treatments of infiltration trench, grass, and rosemary (with and without geocell) with impervious surface (control) and bed soil using a rainfall simulator.

Concerning the basins installed in the rainfall simulator, geotextile sheets were considered at the bottom and then, 15 cm of sandy loam soil (substrate soil) was placed at the bottom of the basin. Table 2 presents the details of sandy loam soil. It is worth noting that sandy loam soil conditions were considered based on research conducted by Dehghani et al. (2019). A geotextile layer was placed on the sandy loam soil and then, 15 cm of experimental treatments were placed along with their soil. For tests containing geocell, geocell layers were placed in layers of sandy soil or plant soil. After each test, all the soils used in each test were removed from the basin and completely dry soil was used for the next test.

In this research, 24 physical models were created at the laboratory level and on a small scale and classified into eight groups. The groups were impervious pavement as the O, substrate layer (SL), G, GGE, R, RGE, T, and TGE. Furthermore, a slope of 5% was considered for the treatments based on research (Li et al. 2020; Hou et al. 2022). Note that three rainfall and runoff conditions were considered for each of the treatments. In the first case, rainfall with an intensity of 200 mm/h was considered in the treatments. In the second case, a horizontal runoff of 3 l/min was tested on the treatments and the third case was the combination of vertical rainfall and horizontal runoff for all samples with a rainfall intensity of 200 mm/h and horizontal runoff of 3 l/min. Table 1 presents details of the experiments. It should be noted that the rainfall intensity of 200 mm/h and horizontal runoff of 3 l/h were considered based on the initial tests and the most critical state of flooding using a rain-making device in all treatments. Rainfall intensity of 200 mm/h was selected based on the return period of 10 years and 10 min of intensity-duration-frequency curves (IDF) of Rasht city. With an average annual rainfall of 1,337.5 mm, Rasht is one of the rainiest cities in Iran. Therefore, to create a critical state, this rainfall intensity was investigated for this city.

Rosemary is a plant with fragrant, evergreen, needle-like leaves and white, pink, purple, or blue flowers (de Macedo et al. 2020). This plant belongs to the mint family, which also includes many plants (de Macedo et al. 2020).

Most grasses are perennial plants, although annual grasses are available in nature. Turf is mostly propagated through seed cultivation; some types of turfs are also propagated by non-seed methods. Among other characteristics of turf is the continuous production of fresh leaves from its growing center, which is performed throughout the growing season. However, the leaves are worn out and destroyed in cases like old age. Regardless of the natural and agricultural conditions, turf grows fast in any case and it is always more than the amount of seed sprayed on the land. Thus, turf has been widely used in green spaces because of its characteristics such as uniform, continuous, and fast growth.

In this research, gravel (sand particles) was used to investigate an infiltration trench in the laboratory dimension. Table 3 gives the details of sandy soil.

By using a rainfall simulator, this study aims to evaluate the performance of impervious pavement, substrate soil, porous gravel, rosemary, and turf in two conditions with and without geocell in terms of TR, TER, C, PF, TP, and TB and under different rainfall intensities and slopes. To this aim, we measured the effective runoff parameters mentioned in the research (Liu et al. 2020).

As shown in Figure 5(c) and 5(d), for treatments G and GGE, the maximum cumulative runoff volume was estimated to be 126.4 and 113.7 l, respectively, while the lowest cumulative runoff volume for the mentioned treatments was estimated at 92.8 and 85.4 l, respectively. Based on Figure 4(e), the lowest and highest cumulative runoff volume observed for treatment R was 123.4 and 164.8 l, respectively. On the other hand, the minimum and maximum cumulative runoff volume for RGE treatment is reported as 116.5 and 157.1 l, respectively (Figure 5(f)). According to Figure 5(g) and 5(h), the lowest cumulative runoff volume for treatments T and TGE was estimated at 143.8 and 134.1 l, respectively, while the maximum cumulative runoff volume for the mentioned treatments was estimated at 186.4 and 172.2 l, respectively.

Tables 5–10 give the results of the factorial analysis of parameters TR, TER, TP, TB, C, and PF based on the independent variables of the test mode (TM) and treatments O, SL, G, GGE, R, RGE, T and TGE (sample treatment, ST). The results were then categorized by Duncan's method. Analyzes were carried out in SPSS 21 software, similar to the studies conducted by Dadrasajirlou et al. (2022) and Samii et al. (2023). The factorial results of the effect of TM and ST parameters on TR, TER, TP, TB, C, and PF (individually and in pairs) reveal the existence of a significant difference at the 5% level between the data in all cases. Table 5 shows the F-statistics and associated p-values for the main effects of ‘ST’ and ‘TM’, as well as their interaction effect ‘ST: TM’. All p-values are less than 0.05, indicating that both ‘ST’ and ‘TM’ have significant main effects on ‘TR’, and there is also a significant interaction effect between ‘ST’ and ‘TM’.

In Table 6, the p-values for both ‘ST’ and ‘TM’ are less than 0.05, which suggests that these variables have a statistically significant effect on ‘TER’. The F-statistic is a measure of how much a model improves the prediction of the data compared to a model with no predictors. A larger F-statistic indicates a more significant improvement. For the ‘ST’ variable, the F-statistic is approximately 3.12. This suggests that including ‘ST’ in the model improves the prediction of ‘TER’ compared to a model without ‘ST’. For the ‘TM’ variable, the F-statistic is approximately 5.94. This suggests that including ‘TM’ in the model improves the prediction of ‘TER’ even more than ‘ST’ does. The p-value is the probability of observing a result as extreme as the one that was actually observed, assuming that the null hypothesis is true. In this case, the null hypothesis is that the variable has no effect on ‘TER’. A smaller p-value indicates stronger evidence against the null hypothesis. For the ‘ST’ variable, the p-value is approximately 0.007. This is less than the commonly used significance level of 0.05, so we reject the null hypothesis and conclude that ‘ST’ has a significant effect on ‘TER’. For the ‘TM’ variable, the p-value is approximately 0.004. This is also less than 0.05, so we reject the null hypothesis and conclude that ‘TM’ also has a significant effect on ‘TER’. In summary, both ‘ST’ and ‘TM’ have a statistically significant effect on ‘TER’. The ‘TM’ variable appears to have a stronger effect than the ‘ST’ variable, as indicated by the larger F-statistic and the smaller p-value.

In Table 7, the F-statistic for ‘ST’ is 669.15 and the p-value is extremely small (7.08 × 10^–56), indicating that the differences between the means of the different groups in ‘ST’ are statistically significant. Similarly, the F-statistic for ‘TM’ is 197.96 and the p-value is also extremely small (1.20 × 10⁻²⁷), indicating that the differences between the means of the different groups in ‘TM’ are statistically significant. This suggests that both ‘ST’ and ‘TM’ have a significant effect on the dependent variable.

Table 8 provides the sum of squares (sum_sq), degrees of freedom (df), F-statistic (F), and the p-value (PR(>F)) for each of the independent variables (‘ST’ and ‘TM’) and the residual. The p-values for both ‘ST’ and ‘TM’ are very small, indicating that both variables have a significant effect on the dependent variable ‘PF’. In our case, the p-values for both ‘ST’ and ‘TM’ are very small (almost zero), which indicates that both variables have a significant effect on the dependent variable ‘PF’. In other words, changes in ‘ST’ and ‘TM’ are associated with significant changes in ‘PF’. This is a high-level interpretation of the results. For a more detailed interpretation, you would need to look at the specific coefficients of the model, which are not provided in this table.

Table 9 shows the sum of squares (sum_sq), degrees of freedom (df), F-statistic (F), and the p-value (PR(>F)) for each of the independent variables (ST and TM) and the residual.

The p-value for both independent variables is less than 0.05, indicating that they significantly affect the dependent variable. In summary, both ST and TM have a significant effect on TR, as indicated by their small p-values. The F-statistics also suggest a significant effect, with TM having a larger effect than ST, as indicated by its larger F-statistic.

In Table 10, the p-values for both independent variables are less than 0.05, suggesting that both variables significantly affect the dependent variable (TB). The p-values for both independent variables (ST and TM) are less than 0.05, which typically indicates that there is a statistically significant relationship between the variable and the response variable (TB). In other words, changes in ST and EC are associated with changes in TB.

Table 11 categorizes EC and ST parameters for the dependent data of TR, TER, TP, TB, C, and PF based on Duncan's method. Classification for EC in all independent parameters of TR, TER, TP, TB, C, and PF was placed in different groups, indicating that hydrological parameters behave differently for the conditions of rainfall, horizontal runoff, and the combination of rainfall and runoff. In the TR parameter, TGE and TGE treatments are placed in one group, while the rest of the treatments are located in different categories. Concerning the TER parameter, GGE and G treatments, as well as RGE and TGE treatments and T and G treatments, are classified in the same category. However, all treatments are placed in different groups for hydrological parameters C, PF, and TP. In the TB parameter, TGE and RGE treatments are categorized into one group, while the rest of the treatments are in other groups.

The research described in this article focuses on the analysis and comparison of different types of coatings and their performance in reducing rainfall-runoff and reducing urban flooding. The use of LID methods, such as vegetation and permeable trenching, is a smart climate adaptation strategy that aims to minimize the negative effects of urbanization and protect water resources.

By studying the parameters affecting urban floods under different rainfall and runoff conditions, this research contributes to climate-smart adaptation by providing insights into effective strategies for managing and controlling floods in urban areas. Understanding the factors that influence rainfall-runoff, such as soil permeability, vegetation type, and pavement characteristics, can help in the design and implementation of sustainable stormwater management practices. This can be seen in the result of Sarker's (2022) research.

Furthermore, the research investigates the performance of different pavement types, including gravel, vegetation (such as rosemary and turf), and geocell-reinforced pavements, in reducing floods and improving hydrological parameters. These findings can contribute to river floodplain erosion control by identifying effective measures to reduce surface runoff and prevent erosion in urban and peri-urban areas. Vegetation and geocell-reinforced pavements, in particular, have shown promise in reducing stormwater runoff and improving slope stability, which can help prevent erosion and sedimentation in river systems. In the research results of Sarker et al. (2023), these issues have been investigated.

In terms of environmental protection, the research explores the use of LID methods to restore the hydrology of urban areas and mitigate the adverse environmental impacts of urbanization. By promoting natural hydrological processes and reducing impervious surfaces, LID methods can help maintain water quality, enhance groundwater recharge, and protect aquatic ecosystems. The findings of this study can inform urban planning and stormwater management practices that prioritize environmental protection and sustainability based on the general topic and objectives outlined in the available sections. The research has the potential to contribute to climate-smart adaptation, river floodplain erosion control, and environmental protection through the evaluation of different pavement types and their impact on urban floods and hydrological parameters.

This research sought to quantitatively investigate the effect of turf and rosemary plants and gravel on some parameters of the catchment basin, such as the amount of runoff volume reduction, TR and TER, C, TP, TB, and PF rate in the laboratory dimension under three conditions of rainfall, horizontal runoff, and combinations of rainfall and horizontal runoff. Generally, 24 tests were conducted on 8 treatments of control (impermeable), sandy loam, gravel, gravel with a geocell layer, rosemary, rosemary with a geocell layer, turf, and turf with a geocell layer at a slope of 5% using a rainfall simulator. In a certain treatment, using the geocell layer leads to the lowest cumulative runoff volume and the highest reduction percentage in the runoff volume in all test situations. Gravel treatment with geocell can reduce the runoff volume for the test conditions of rainfall, horizontal runoff, and the combination of rainfall and horizontal runoff by approximately 68, 62, and 58%, respectively, which is the best treatment in this regard. The presence of turf and gravel can also delay the TP flow rate. By comparing the turf and rosemary treatment and gravel with sandy loam treatment, the percentage of runoff volume reduction is higher, while the PF rate is less. Concerning the GGE treatment, the TR was reported greater and the C was lower than that of the other treatments. Among the limitations of this study, the lower surface of urban substrates may have a much greater soil thickness than the thickness determined in this experiment. But considering that all the treatments and tests were done in completely identical conditions, and then it is possible to compare the treatments in runoff control and the parameters affecting it. Future research is suggested to investigate the soil erosion rate and the quality of runoff with and without geocell under the soil conditions studied in this research.

This research received no external funding.

H.G. carried out data curation, performed formal analysis and methodology, investigated the work, found resources and software, visualized the work, and wrote the original draft. H.K. conceptualized the work, did funding acquisition, administered the project, supervised the work, wrote the review for the article, and edited the manuscript.

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

The authors declare there is no conflict.