Evaluation of comprehensive treatment effect of geotechnical and ecological engineering for debris flow: case of Wenchuan County, Sichuan Province

As a crucial link in debris flow disaster control, control engineering can be divided into geotechnical engineering and ecological engineering. Geotechnical engineering is less affected by climate and can play an immediate role in the treatment of debris flow in a short time. Ecological engineering has obvious regulating effect on surface runoff and erosion. Furthermore, there are few discussions on comprehensive treatment effect between geotechnical engineering and ecological engineering for debris flow. Therefore, taking the debris flow gullies with control engineering implemented in Wenchuan area of Sichuan Province as the research object, this paper puts forward an evaluation method of treatment effect comprehensively considering the geotechnical engineering and ecological engineering. It is mainly based on the impact of engineering on the formation conditions of debris flow and the damage of projects. The comprehensive treatment effect of geotechnical and ecological engineering for debris flow is analyzed. The corresponding evaluation indexes are extracted, including 5 first-class evaluation indexes such as landform factor and water source factor, as well as 11 s-class evaluation indexes such as the peak flow reduction rate, the sediment retention ratio and the LAI ratio. The Fuzzy-AHP method is used to establish the evaluation model of treatment effect. The evaluation results are graded. It is considered that the treatment effect of other debris flow gullies is medium or above except that two gullies are poor. The evaluation results are consistent with the actual investigation, which provides a reference for the control of debris flow in this area.


Introduction
According to the classification of debris flow control engineering, it can be divided into geotechnical engineering and ecological engineering (Cui et al. 2013). Geotechnical engineering for debris flow mainly refers to structural measures such as check dams, sabo structures, levees, and training channels (Mizuyama 2008). The check dam can reduce the velocity of debris flow and decrease the peak sediment discharge and stop the front part of debris flow (Ikeya 1989). Beam dam, which is widely used in Japan, has played a good role in the separation of water and stone of debris flow and accelerated the deposition of sediment (Okubo 1997). The drainage channel of the energy dissipation structure can significantly reduce the velocity of debris flow (Jiangang et al. 2015). Channel works can be used to prevent the deposition of debris flow and reduce the river bed variation (Ikeya 1989). Setting baffles in the drainage channel can greatly reduce the flow velocity and have a good control effect on the flow velocity of debris flow (Wang et al. 2021). The drainage system can reduce the superficial and subsurface runoff from the area above the slide and prevent the formation of shear surface to stabilize the unstable area (Huebl 2005) The effect of geotechnical engineering on debris flow control is remarkable. However, in environmentally unfriendly areas, it is not enough to rely only on geotechnical engineering, such as slit dams. The risk of failure of the much more flexible ecological engineering methods can also be reduced (Wu and Feng 2006). The term ecological engineering was coined by Odum in the 1960s (Odum, H.T. 1962). The concept of ecological engineering generally refers to the development of new sustainable ecosystems that have both human and ecological value (William J. Mitsch 2003). Ecological engineering generally includes agroecological restoration, riparian trees (Lin et al. 2006) and vegetation applications (Iii R 2005). The formation of debris flow is due to soil erosion, which provides sufficient conditions for its material source. Roots can effectively reduce streambank erosion by improving the shear strength of soil (Docker 2008). Compared with devegetated riverbanks, vegetated riverbanks can effectively improve the instability of slopes, so as to reduce the formation of loose solid matter and collapse (Hubble 2004). Similarly, the forest will intercept 2-5% of the canopy throughfall to the forest floor to reduce surface runoff (Harmon 1995), which will inhibit the formation of debris flow. In agroecological engineering, after the implementation of the slope to ladder project, the surface runoff is significantly lower than before, and the water diversion loss and soil loss are also greatly reduced, and the soil infiltration capacity and soil erosion resistance are significantly enhanced (Zuo 2004). Therefore, it can be found that ecological engineering is very important for the treatment of debris flow.
In the existing research on debris flow control engineering, it is not enough to consider only geotechnical engineering or ecological engineering. The evaluation of comprehensive treatment effect of the two is less involved. Furthermore, the research on the treatment effect of geotechnical engineering does not consider the factors of damage. Because the factors of the engineering itself will also have a certain impact on the treatment effect, it is necessary to make a systematic and effective analysis on this part. It is crucial to summarize the comprehensive treatment effect of geotechnical engineering and ecological engineering for debris flow, extract the corresponding evaluation indexes, and select the appropriate evaluation method to establish model. Based on the above discussion, on the basis of field investigation and relevant tests, this paper takes the existing engineering in 18 debris flow gullies in Wenchuan area of Sichuan Province as the research object to analyze the effect of engineering in the area. Combined with the operation of the engineering, the evaluation factors are extracted, and the Fuzzy-AHP method is used to quantitatively evaluate the treatment effect. It provides a reference for debris flow control in this area.

Geographical position
Wenchuan County, located in the southeast of Aba Tibetan and Qiang Autonomous Prefecture, Sichuan Province, is between E102°51′ ~ 103°44′ and N30°15′ ~ 31°43′, with a total area of about 4082.9 km 2 . G213, G317 and S303 in the county are important transportation and national defense channels for Tibet. The study area, located in Wenchuan County, is located in the middle and lower reaches of Minjiang River, including Miansi Town, Yinxing Town, Gengda Town, Yingxiu Town and other areas. The study area contains 18 debris flow gullies including Qipan, Banzi, Dengxi, Sucun, Maxi and Mozi et al.

Topographic features
Wenchuan County, located in the northwest of Sichuan Basin, is located between Qionglai Mountain System and Longmen Mountain System. It is a typical alpine valley landform with high northwest and low Southeast landforms, strong valleys longitudinally and horizontally, and large topographic gradients and relative elevation differences (Han, 2018). The Minjiang River flows through this area with deep valley and turbulent flow. The width of the river is about 70 ~ 100 m. In general, the area is divided into three levels according to elevation. The high mountains and extremely high mountains with an altitude of about 3500 ~ 5000 m are first-class stairs, and the relative height difference is about 1500 m. The middle and high mountains with an altitude of about 2000 ~ 3500 m are two-step steps, and the relative height difference is more than 1000 m. The middle mountains, low mountains and river valleys below 2000 m above sea level are three steps, with a relative height difference of about 800 m (Hu 2017). Due to the rapid downcutting of Minjiang River, the cutting on both banks is relatively strong, so the terrain is generally steeper. The high mountains and steep slopes in the study area provide sufficient conditions for the formation of debris flow.

Meteorohydrology
Wenchuan County is located in the earthquake extreme earthquake area, wet in the South and dry in the north, with obvious vertical zoning. The south of Supodian, Yinxing Town, is the rainy central area in Western Sichuan, with an average annual rainfall of 1285.1 mm. The north belongs to the semi-arid valley area in the upper reaches of Minjiang River, with an average annual rainfall of 526.3 mm. Unique natural geographical environment and complex climatic conditions provide good conditions for the formation and development of mountain disasters such as landslide and debris flow (Han 2018). According to the statistics of precipitation data in recent 30 years , it is found that the average annual precipitation in Wenchuan County is 847.6 mm. The maximum annual precipitation is 1213.3 mm, which occurs in 2018. The minimum annual precipitation is 537.3 mm, which 1 3 occurs in 1996. According to the monthly average rainfall data of Wenchuan County, from 1991 to 2020, the precipitation is concentrated in May to September, of which the maximum monthly rainfall is 166.1 mm and the minimum monthly rainfall is 5.8 mm. Due to the complex terrain in mountainous areas and the concentrated rainfall in summer, debris flow disasters occur frequently in this period. In particular, on August 20, 2019, 00:00-07:00 (Beijing time), the maximum accumulated rainfall in Wenchuan County was 65 mm. There were 22 dangerous areas where flash flood warning occurred, leading to debris flows in many places in Wenchuan County. The field investigation of this paper is mainly based on the background of "8.20" debris flow.

Engineering overview
The study area includes 18 debris flow gullies, such as Qipan, Banzi, Dengxi and Sucun et al. The debris flow gullies are numbered in turn along the middle and lower reaches of Minjiang River from GD01 to GD18. Geotechnical engineering in each debris flow gully is investigated on the spot, including 46 check dams and 11 drainage channels. The basin area, main gully length, gully bed gradient and other parameters of each debris flow gully are counted, as shown in Table 1. As the Wenchuan earthquake in 2008 had a great impact on vegetation, only the vegetation coverage after the earthquake was statistically investigated and the vegetation coverage in the ecological engineering was interpreted by remote sensing. However, as an important indicator of vegetation coverage, NDVI value compares the changes of NDVI value from 2008 to 2018, as shown in Fig. 1. The position  Fig. 2. The field investigation of some debris flow gullies is shown in Fig. 3. Taking the Chediguan gully as an example, we can see that the bedrock on both sides of the gully is exposed and is severely scoured by the debris flow. Similarly, the mud marks are also extremely obvious. According to the site conditions of Taoguan gully and Niujuan gully, the scouring

Construction of evaluation model
The selection of evaluation indicators of comprehensive treatment effect can be considered from the impact on the formation conditions of debris flow, combined with the engineering ontology factors and considering the impact on reducing the peak flow of debris flow, so as to construct the evaluation index system. The Fuzzy-AHP method is used to determine the weight of the evaluation index, and the grade of the evaluation result is finally determined by constructing the membership function.

Selection of evaluation factors
The comprehensive treatment effect of geotechnical engineering and ecological engineering is reflected in the impact on the formation conditions of debris flow in the whole basin after the implementation of the engineering. At the same time, considering the effect on the peak flow of debris flow and the own factors of the engineering, the evaluation index system of comprehensive treatment effect is established. The 18 debris flow gullies in the study area have adopted geotechnical control works, such as check dams or drainage channels. The research of ecological engineering in the study area can be considered from the vegetation changes before and after treatment.

Effect on peak discharge of debris flow
The reduction of peak flow of debris flow by geotechnical and ecological engineering is reflected in that the check dams in geotechnical engineering can intercept coarse particles in debris flow. Under the action of gravity separation, fine particles and water enter the downstream of check dam from drainage hole or overflow outlet, resulting in the reduction of peak flow (Chen, 2013). The forestry measures in the ecological engineering can effectively intercept the debris flow, resulting in the accumulation of debris flow . The vegetation can affect the velocity of debris flow by increasing the surface roughness, so as to reduce the peak flow of debris flow. As the basic parameter of engineering design, the peak flow of debris flow is also an important factor to measure the destructive power of debris flow. For the calculation of the peak flow of debris flow, the velocity can be calculated first. In each debris flow gully, 7 typical sections are taken from upstream to downstream. The typical section should be selected at the place where the channel is straight and the section changes little. There is no blockage, and there is no confluence and backflow, and the mud trace is relatively clear. Taking the Chediguan gully as an example, it can be seen from Fig. 3 that the debris flow has left obvious scratches on the bank wall. Moreover, the gully at this location is straight, and the cross section has little change. Therefore, it can be used as a typical vertical section. Take this as the principle of investigating the typical section, and then investigate the next typical section within 50 m along the channel. The reliability of the mud level is judged according to the continuity of the mud level on the vertical section. And the average mud depth, flow section area and hydraulic gradient are calculated accordingly. The average width and slope of the gully are measured with infrared rangefinder. Then, the empirical formula method is used to select different velocity calculation formulas according to the nature of debris flow. Combined with the mud depth data, the morphological investigation method is used to calculate the peak flow of the section, which can represent the peak flow reduction rate of debris flow: where Q max is the maximum flow in 7 typical sections, m 3 /s; Q 1 is the flow reduced by geotechnical engineering and ecological engineering, which is the calculated flow of Sect. 7, m 3 /s.

Effect on landform
After the implementation of geotechnical engineering in the study area, the check dam can slow down the longitudinal slope of the ditch bed, so as to form a back silting slope, which can effectively reduce the erosion of the ditch bed. Therefore, the impact on the landform can be analyzed from the slope of intercepted sediments. Through field investigation, infrared rangefinder is used to measure the slope of intercepted sediments and the original ditch slope, respectively. Because the original ditch slope is not easy to measure, select the place downstream of the check dam that is not obviously affected by debris flow. Taking its slope as the original ditch slope. The slope upstream of the check dam is taken as the slope of intercepted sediments. Then, it can calculate the longitudinal slope ratio of ditch bed: where I 1 is the slope downstream of the check dam; I 2 is the slope upstream of the check dam.

Effect on substance source of debris flow
Influence of comprehensive action on substance source conditions of debris flow is mainly reflected in that geotechnical engineering can effectively retain sediment. It can reduce the substance source of debris flow. Ecological engineering can improve the shear strength of soil, so that it is not easy to form loose solid matter, so as to inhibit the generation of substance source.
The amount of sediment intercepted is determined by the dam height and the topography at the upstream of the dam, which can be calculated as a triangular wedge: where V is the amount of sediment intercepted, m 3 ; H is the clear height of the dam, m; l is the length of back siltation, m; b is the average ditch width, m; I is the original ditch slope; I 0 is the slope of intercepted sediments. The total amount of debris flow process is calculated according to the maximum flow and duration of debris flow: where Q is the total amount of a debris flow process, m 3 ; Q c is the maximum flow of debris flow, m 3 /s; T is the duration of debris flow, s. According to the field investigation, the duration of debris flow is 1 h, taking 3600 s.
Then, the amount of solid material washed out by debris flow: where c is the gravity of debris flow, and the average value of gravity is calculated according to the measured soil samples on site, taking 1.82 t/m 3 ; w is the gravity of water, taking 1 t/m 3 ; H is the gravity of solid matter of debris flow, taking 2.65 t/m 3 . The sediment retention ratio is defined as ratio of the amount of sediment retained to the amount of solid material washed out by debris flow: where V is the amount of sediment retained by check dams, m 3 ; Q H is the amount of solid material washed out by debris flow, m 3 . (2) The shear strength of soil is an important indicator of the consolidation of grass planting soil. Plant roots can strengthen the shear strength of soil (Manbeian 1973). Therefore, the root soil samples on the soil surface in ecological measures are selected. The soil samples are prepared according to the water content and dry density of undisturbed soil samples, and finally, the direct shear test is carried out. Cohesion, as an index of soil characteristics, is the standard to measure the shear strength of soil. According to the cohesion c value of each debris flow gully obtained from the test, it is normalized to define the shear strength ratio: where c is the cohesion of the soil sample taken from the debris flow gully, kPa; c max is the maximum cohesion of all soil samples, kPa; c min is the minimum cohesion of all soil samples, kPa.

Effect on water source of debris flow
(i) Soil bulk density When the soil texture is similar, the bulk density can better reflect the tightness of the soil. The bulk density is thinner and more porous than the novel soil, and its porosity is higher. The water storage capacity of the soil is related to the porosity. Therefore, the soil with smaller bulk density can store more water and give better play to its water storage benefits. By comparing the difference of bulk density between root soil and soil without root, the improvement effect of ecological engineering on soil bulk density is characterized. On site, the soil samples of each debris flow gully are sampled with a ring knife. The six samples of root soil and soil without root are taken, respectively. The root soil samples are 10 cm away from the soil surface. Finally, the bulk density of root soil and soil without root is obtained. The average value is taken to define the degree to which the soil bulk density is lower than that of roots: where 1 is the bulk density of root soil, g/cm 3 ; 2 is the bulk density of soil without root, g/ cm 3 .
(ii) Soil porosity The total water storage of soil depends on the total porosity of soil. Therefore, the soil with large total porosity can store more water, which can better reduce the runoff forming debris flow. Calculation formula of total soil porosity: where ds is the bulk density of soil particles. It is generally considered that the bulk density of soil particles of soil topsoil is 2.65 g/cm 3 . rs is the bulk density of soil and can be measured by the ring knife.
The increase degree of soil porosity is expressed by soil porosity ratio: where 2 is the porosity of root soil; 1 is the porosity of soil without root.
(iii) Leaf Area Index LAI (Leaf Area Index) has a wide range of uses and important significance in the study of forest hydrology. It refers to the ratio of the total forest area in the stand to its forest area. It is one of the important indicators to measure whether the stand structure is reasonable or not. Canopy interception increased linearly with the increase in LAI (Zeng, 1996). Therefore, LAI can reflect the ability of vegetation to intercept rainwater. LAI can be obtained indirectly through optical remote sensing technology. NDVI (Normalized Difference Vegetation Index) is one of the commonly used vegetation indexes. The correlation between LAI and NDVI is established (Hui et al. 2003): where X is the value of NDVI; Y is the value of LAI. The NDVI data of each debris flow gully in recent ten years (2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018) are obtained, and then, the changes of LAI are obtained. The LAI ratio is defined as: where Y 2 is the LAI in 2018; Y 1 is the LAI in 2008.

Influencing factors of the engineering itself (i) Normalized difference vegetation index
As a widely used vegetation index, NDVI is the best indicator of plant growth state and spatial distribution density, which is linearly correlated with plant distribution density. Through remote sensing interpretation of the study area, the average value of NDVI of each debris flow gully in recent ten years (2008-2018) is obtained, and the NDVI ratio is defined as: where S 1 is the annual average NDVI in 2008; S 2 is the annual average NDVI in 2018.
(ii) The siltation degree of engineering The sediment deposition of the drainage channels in the study area is less, so the siltation degree of the engineering in this study is only for the check dams. Due to the limited sediment regulation capacity of each check dam, the sediment deposition in the dam will greatly weaken the sediment regulation capacity of the dam, so as to shorten the service life of the check dam. Therefore, the sediment deposition in the check dam is closely related to the treatment effect. Conduct on-site investigation on the check dams in each debris flow gully in the study area and count their respective siltation degree. Since the design storage (10) capacity data of the dam are not easy to obtain, the siltation thickness is used to characterize the siltation degree of the engineering. The siltation degree of engineering can be expressed as: where h is the siltation thickness of sediment, m; H is the clear height of check dam, m. For the case that there are multiple check dams in the same debris flow gully, the mean value of the calculation results is taken as the siltation degree.
(iii) The opening degree of drainage holes In order to reduce the hydrodynamic pressure in front of the dam, increase the drainage of the middle and lower parts of the dam and reduce the flow of the overflow outlet on the dam crest, drainage holes will be set on the dam to adjust the function of transporting sediment and prolong the service life of the check dam. However, when the drainage hole is blocked, it will have a great impact on the sand transport capacity of the check dam. If the dredging is not carried out in time, the subsequent debris flow will cause dam break, resulting in greater losses. Therefore, the blocking of drainage holes will have a certain impact on the treatment effect of the engineering. The ratio of the number of drainage holes not blocked to the total number of drainage holes is used to represent the opening degree of drainage holes: where ΔC is the number of drainage holes blocked; C is the total number of drainage holes.
(iv) The damage degree of engineering The geotechnical engineering in the study area mainly includes check dams and drainage channels. Considering that the damage of engineering will also have a great impact on the treatment effect, the damage factors are summarized. The check dams include the damage of dam foundation, the damage of dam abutment and the damage of dam body. The drainage channels include foundation scouring, structural impact and slope thrust, and their damage degrees are taken quantitatively. The main basis is shown in Tables 2 and  3. Therefore, the damage of geotechnical engineering in the area can be comprehensively evaluated. Generally, it can be divided into two levels, as shown in Fig. 3. According to the evaluation index diagram of damage degree of the engineering, the Fuzzy-AHP method is used for quantitative evaluation, and the evaluation results are expressed as A 11 .

The weight of damage degree of the engineering
The evaluation index system of damage degree of the engineering is shown in the figure below. Using analytic hierarchy process, the damage degree of geotechnical engineering is the evaluation target. The middle layer includes 2 factors such as check dam factor B 1 and drainage channel factor B 2 . The bottom layer includes 6 items such as the damage degree of dam foundation C 1 , the damage degree of dam abutment C 2 , the damage degree of dam (14)  Layer B 2 − C eigenvector: W 3 = (0.557, 0.123, 0.32) T , maximum characteristic root: max = 3.0536.

(iii) Conduct conformity inspection:
The consistency index can be obtained according to Table 4, and the consistency index can be calculated: CI = max −n n−1 , ( max is the maximum eigenvalue of the judgment matrix). Then, calculate the consistency ratio: CR = CI RI . They are 0, 0.0088 and 0.05, respectively, which are less than 1, so the consistency of the judgment matrix is acceptable.

Weight of evaluation factors of treatment effect
The evaluation index system of treatment effect established according to the above selected evaluation factors is generally divided into two layers, as shown in Fig. 5. The analytic hierarchy process is used to evaluate the comprehensive treatment effect. The middle layer includes 5 items such as characteristic parameter factor N 1 and landform factor N 2 , and the bottom layer includes 11 items such as the flow reduction rate A 1 and the longitudinal slope ratio of ditch bed A 2 .
(i) The judgment matrix is constructed by using the 1-9 scale method, and the judgment matrices of M − N layer, N 3 − A layer, N 4 − A layer and N 5 − A layer are obtained as follows: (iii) Conduct conformity inspection: The consistency index can be calculated: CI = max −n n−1 , ( max is the maximum eigenvalue of the judgment matrix). Then, calculate the consistency ratio: CR = CI RI . They are 0.033, 0, 0.027, 0.020, respectively, which are less than 1. Therefore, the consistency of the judgment matrix is acceptable, in which the eigenvector is the calculated weight.

Establishment of membership function
Fuzzy-AHP method can realize the overall evaluation covering these factors for things with diversified influencing factors. The fuzzy boundary is described by membership function and membership degree, so as to grade or classify things and evaluate them. The principle of maximum membership degree is applied to make a more detailed analysis of the attributes or related factors of things. Finally, it can determine the grade of the evaluation results, change the qualitative description into quantitative description, and make the results appear concise and intuitive. The influencing factors are divided into four grades: excellent, good, medium and poor, and the corresponding grade scores are 0.875, 0.625, 0.375 and 0.125. According to the characteristics of qualitative factors, the trapezoidal membership function is used to construct its membership function, as follows: 4 Evaluation results: taking the engineering for debris flow in Wenchuan area as an example

Quantitative calculation of evaluation factors
According to the selection and formula of the above evaluation factors, the evaluation factors of comprehensive effect treatment of geotechnical and ecological engineering in 18 debris flow gullies in the study area are quantitatively calculated, including 11 evaluation factors such as longitudinal slope ratio of gully bed, silt retention ratio and shear strength ratio. The results are shown in Table 5.

Name
The peak flow reduction rate The longitudinal slope ratio of ditch bed The sediment retention ratio The shear strength ratio The ratio of soil bulk density The soil porosity ratio The LAI ratio The NDVI ratio The siltation degree of engineering The opening degree of drainage holes The

Characteristic parameter factor
The characteristic parameter factors mainly affect the peak flow of debris flow. Table 5 shows that the reduction rate of peak flow of debris flow in the study area is different. The minimum value is 20.1%, and the maximum value can reach 54.1%, and generally it is mostly 30 ~ 50%, indicating that the engineering have played a certain role in reducing the flow of debris flow.

Landform factor
Landform factor mainly refers to the longitudinal slope ratio of ditch bed. From the calculation results, it can be seen that the slope ratio of ditch bed is generally large, indicating that the slope of back siltation formed by the check dams in the study area is obvious. It is consistent with the slope of back siltation phenomenon caused by more sediment deposition in the check dams reservoir found in the field investigation. Taking Sucun gully as an example, the longitudinal slope ratio of ditch bed is 0.827. According to the field investigation, it is found that the three check dams in the gully are full of sediment. As shown in Fig. 6, the storage capacity of 2# dam and 3# dam is full. The drainage holes are completely blocked. It results in a slow upstream slope of the check dam, which plays the role of protecting the bed and fixing the slope. Therefore, the longitudinal slope of ditch bed is relatively large.

Substance source factor
Substance source factors mainly include the sediment retention ratio and the shear strength ratio. It can be seen from the data that the check dams still play a great role in the whole, effectively intercepting the substance source of debris flow and reducing the harm of debris flow. However, in some gullies, such as Yinxingping, due to its large amount of substance source,resulting in a large amount of sediment when debris flow occurs. Therefore, the only check dams are not enough to retain so much sediment, which will lead to the sediment deposition. The site investigation found that the three check dams in the Yinxingping gully are full. Moreover, for example, according to the Fig. 6 The check dams in the Sucun gully above formula, the solid matter flushing amount of a debris flow outbreak in Xingfu gully is 353400 m 3 , and the total sediment storage amount of seven check dams is 38500 m 3 . It can be obtained that the sediment retention ratio of Xingfu gully is 0.109. The sediment retention ratio is relatively small, indicating that the sediment retention effect of the check dams in the Xingfu gully is poor. Combined with the field investigation, it is found that Xingfu gully has more loose solid materials and rich substance sources, while most of the seven check dams have been silted up, as shown in Fig. 7. It can be seen that the storage capacity of the check dams is limited, which is not enough to retain the total amount of solid matter for the outbreak of a debris flow. Consequently, the sediment retention effect is poor, which is consistent with the results of the field investigation.
According to the calculation results, the shear strength ratios of soil in Yinxingping, Liujia and Niutang gully are 1.000, 0.876 and 0.828, respectively, indicating that the overall level of soil shear strength is high. Combined with field sampling, it is found that the water content of root soil in the gully is high, and the content of roots is also large. The root content has a great impact on the shear strength of soil, which can greatly enhance the cohesion of soil. It can effectively inhibit the formation of substance sources of debris flow.

Water source factor
Water source factor mainly include the ratio of soil bulk density, the soil porosity ratio and the LAI ratio. From the calculation results of the ratio of soil bulk density and the soil porosity ratio, it can be seen that the physical properties of soil can be better improved after the implementation of ecological engineering. Reducing the bulk density of soil and increasing the porosity of soil can effectively reduce the formation of water source of debris flow. From the calculation results, it can be seen that there is a small gap in the effect of reducing soil bulk density in each debris flow gully. But in terms of improving soil porosity, the soil porosity of Hongchun gully and Guji gully is relatively large, which are 0.682 and 0.622, respectively, indicating that the effect is obvious.
The data of the LAI ratio show that the LAI of each debris flow gully in the study area increases with time, and the growth rate is mostly 20%. However, the growth rate of some gullies is small, such as Banzi gully, Liujia gully and Niutang gully, whose LAI ratios are 0.036, 0.097 and 0.052, respectively.

Engineering ontology factor
The engineering ontology factors mainly include the NDVI ratio, the siltation degree of engineering, the opening degree of drainage holes and the damage degree of engineering.
(i) The NDVI ratio.
Through remote sensing interpretation of the study area, the NDVI distribution value in recent ten years (2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018) is obtained, and the annual mean value of NDVI of each debris flow gully is obtained, as shown in Fig. 8. It can be seen from the figure that the NDVI value in the study area increases with time. According to the calculation data, the growth rate of most debris flow gullies is about 20%, and a few debris flow gullies are quite different, such as Banzi gully, Liujia gully and Niutang gully. The growth rate of vegetation coverage is small, less than 10% in recent ten years.
(ii) The siltation degree of engineering.
According to the field investigation, most of the dam bodies in the study area are in the state of half reservoir or full reservoir. For example, there is more sediment deposition in the check dams in Banzi gully and Cutou gully, as shown in Fig. 9, so the siltation degree of the dam is small. Through the calculation data, it can also be found that the siltation degree of engineering in the two gullies is 0.060 and 0.048, respectively. Therefore, the sediment regulation ability of the dam body is greatly reduced.
However, in Liujia gully, it is found that the check dam has less siltation and less blockage of drainage holes, so the dam body can still play a good role of "blocking coarse and discharging fine." For the check dams in Mozi gully, the dam body is also mostly in the state of half reservoir or empty reservoir, as shown in Fig. 10. By calculating the respective siltation degrees, they are 0.744 and 0.528. It can be seen that the check dams in the gully can still play a good role. It can be seen from the calculation results that the opening degree of some check dams are small, such as Banzi gully and Sucun gully, with an opening degree of 0.010. Therefore, the ability of sediment regulation of the check dams is greatly weakened. It can only play its sediment retaining benefit, which has a great impact on the adjustment of debris flow particle size and the reduction of hydrodynamic force in front of the dam. In the gullies such as Hongchun gully, Liujia gully and Mozi gully, the opening degrees are 0.750, 0.800 and 0.640, respectively. It can be seen that the opening degree of the drainage hole of the dam body is large, as shown in Fig. 11, indicating that the drainage hole is less blocked, and the dam body can still play its role in regulating sediment.
(iiii) The damage degree of engineering Taking 1# dam in the Sucun gully as an example, according to the field investigation, it can be seen that the dam foundation has been seriously eroded by running water, resulting in the exposure and suspension of the dam foundation. At the moment, the dam foundation is in danger of toppling and deformation at any time. However, the dam abutment and dam body have high integrity and low damage degree, as shown in Fig. 12. As the dam body has been filled with sediment, its function must be affected to a certain extent. The classification and quantitative value of each main control factor are described, in which the dam foundation damage is severe damage, and the value is 0.1. Since the dam abutment is slightly damaged, we determine its damage value as 0.9. The dam body is slightly damaged, and its value is 0.9.
According to the calculation results, Banzi gully, Dengxi gully and other engineering in the study area are seriously damaged, while the damage of Cutou gully, Chediguan gully are moderate, and the damage of engineering in other debris flow gullies is less. According to the field investigation, the integrity of these prevention and control projects is high, so the functionality of the engineering is less affected, which is beneficial to the treatment effect.
The grille e-dam in the B Banzi gully The check dam in the C Cutou gully Fig. 9 The majority of sediment in the check dams

Evaluation of comprehensive treatment effect
Based on the analysis of the comprehensive treatment effect of geotechnical and ecological engineering in 18 debris flow gullies, 11 evaluation factors are obtained. The calculation results of each factor are shown in Table 5. Since the physical meaning of each evaluation index is different, and the influence of dimension is eliminated. Consequently, the samples are normalized using the following formula: The factors are divided into four grades: excellent, good, medium and poor, and the corresponding grade scores are 0.875, 0.625, 0.375 and 0.125. The membership function refers to the above formula, so as to determine the fuzzy comprehensive evaluation vector of the treatment effect of each debris flow gully, and determine its membership evaluation grade according to the principle of maximum membership degree. The evaluation results are shown in Table 6. According to the evaluation results, the treatment According to the field investigation, the debris flow in Banzi gully has seriously damaged the engineering and destroyed the houses near the gully mouth, resulting in huge economic losses, as shown in Fig. 13. After the debris flow broke out in Dengxi gully, the villagers were visited on site. The houses were seriously damaged, and the mud marks outside the wall could reach about 2 m. In addition, the projects in the gully were seriously damaged, most of check dams were damaged, and the functions were seriously affected, as shown in Fig. 14. In conclusion, it can be seen that the evaluation results are in good agreement with the actual investigation.

Conclusions and discussion
Based on the field investigation of 18 debris flow gullies with engineering implemented in Wenchuan area, Sichuan Province, the comprehensive treatment effect of geotechnical and ecological engineering is analyzed and evaluated, and the following conclusions are drawn: 1. The comprehensive treatment effect can be considered from the aspects of landform, substance source and water source forming debris flow and reducing the peak flow of debris flow. 2. Considering the effect of the engineering's own factors on the treatment effect, the firstclass evaluation indexes of comprehensive treatment effect are selected as five factors: landform factor, water source factor, substance source factor, engineering ontology factors and characteristic parameter factor. The first-class evaluation indexes are refined to summarize the second-class evaluation indexes, including 11 items such as longitudinal slope ratio of ditch bed, damage degree of the engineering. 3. Through the combination of field investigation and indoor experiment, the quantitative calculation of each selected evaluation index can better reflect the actual situation of the site. 4. The Fuzzy-AHP method is used to establish the membership function and then determine the level of treatment effect. Among the engineering in the 18 debris flow gully, the Although the quantitative evaluation of the treatment effect of geotechnical and ecological engineering has made some progress, there are still some deficiencies. It is hoped that further discussion can be made in the future research: a. In this paper, different types of engineering are not discussed in detail. Therefore, in the follow-up research, it is necessary to distinguish the differences of treatment effects between different types of engineering, so as to provide reference basis for the treatment effect. b. In the damage research of geotechnical engineering, this paper only qualitatively describes and uses Fuzzy-AHP method for evaluation. A quantitative index should be put forward for a certain damage factor, so as to make the quantitative evaluation and make the results more intuitive. c. Part of the data acquisition in this paper has certain experience and randomness, such as the calculation of peak flow of debris flow, and the source of data is based on field investigation, which lacks high accuracy. Therefore, a series of experiments are needed to support these theories in the subsequent work. d. At present, there is little research on the mechanism of comprehensive action of geotechnical-ecological engineering, and subsequent experiments are needed to study the mechanism of synergy and the mode of optimal allocation of the two. e. The Fuzzy-AHP method used in this paper has the influence of human factors in calculating the weight, so it needs to be improved in the follow-up work.

Conflict of interest
The authors have not disclosed any competing interests.