TW200419045A - Sediment yield estimation expert system (SedExpert) - Google Patents

Abstract

In the Soil and Water Conservation Technical Regulations published by Council of Agriculture, Executive Yuan, Republic of China, chapter 2 basic data investigation and analyses, section 8 soil erosion, 62th rule indicates the soil erosion estimation of slope land may use Universal Soil Loss Equation (USLE), the formula can be calculated as: Am=Rm.Km.L.S.C.P, where Am is the average annual soil loss (t/ha per year), Rm the rainfall erosivity factor (MJ/hammh), Km the soil erodibility factor (t/MJ h mm), L the slope length factor, S the slope steepness factor, C the cover-management factor and P is the supporting practice factor. During an early stage, soil loss in small agricultural plots could be obtained through the field survey. With the fast progress in computer technology, soil erosion can be scaled up from slope to the large-scale watershed by combining the USLE and the techniques of GIS and RS. The objective of the patent is to integrate USLE, mentioned-above techniques and digital elevation to automatically extract the soil erosion factors and calculate sediment delivery ratio, and then developed the sediment yield estimation system to calculate the spatial information and distribution of sediment yield in the watershed. The analyzed result can be used as the reference of sediment control in the watershed.

Description

200419045

V. Description of the invention (1) The silt amount estimation expert system (Sed Expert) of the present invention is based on the automatic division theory of catchment area, combined with GIS, RS technology and DE M data, and automatically extracts soil °° erosion factor, and combines The concept of the mud-sand transfer rate on the slope is used to estimate the software of mud yarn production in the catchment area. The technical fields, contents and implementation methods are described as follows: Xiyi Technical Field The technical field of the present invention belongs to soil and water conservation planning and design and geographic information systems. Software invention in the field of spatial analysis. Second, the prior art

Universal Soil Loss Equation (Universal Soil Loss Equation,

U S L E) is currently the most widely used soil cash forecasting formula in countries around the world. The soil erosion process is quantified and estimated by a series of factors. It has been applied successfully in the United States. Its theoretical model (such as WEPP) is different and cannot be applied to any region. Therefore, this formula is used to evaluate soil erosion balance in Taiwan. When measuring, it is advisable to make appropriate corrections to the erosion parameters to match the natural environmental conditions in Taiwan. Early soil loss assessment was based on small test plots, and most of its factors were obtained from field measurements. With the advancement of computer technology, the soil surplus model combined with GIS and RS technology has been widely used in large-scale watersheds. Evaluation, but the current assessment method still uses the entire catchment area or sub-catchment area as the evaluation unit, and its factors are relatively rough. In addition, its parameter factors cannot be reasonably derived, and more accurate quantitative information cannot be calculated (Wischmeier , 1 9 7 6; Wilson, 1 986; Moore and Wilson, 1 9 92). In order to improve the above shortcomings, the silt production estimation system of the catchment area of the present invention is based on the theory of automatic division of catchment area, combined with the USLE model and the sediment transfer rate.

200419045

* 1 I I V. Description of the invention (2) Software for estimating the amount of mud and sand. 3. Contents of Estimation of Slope and Sand Yield in Catchment Area (I) Calculation of Slope and Sand Yield The soil erosion model can be divided into empirical model (EmP iri cal Model) and theoretical model (Physical Model). υπϋ is the formula under test. Its soil surname process is quantified and estimated by a series of factors. It is quite successful in the United States. It is different from the theoretical model (such as WEPP) and cannot be applied to any region. Therefore, it is used in Taiwan. This formula evaluates soil

In the case of erosion, appropriate modification of erosion parameters should be made to match the natural environmental conditions in Taiwan. Early soil loss assessment was based on small test plots. Most of its factors were obtained from field measurements. With the advancement of computer technology, the soil impact model combined with G IS and RS technology has been widely used in large-scale watersheds. In the current evaluation method, the entire catchment area or sub-catchment area is still used as the evaluation unit. The representativeness of the factors is relatively rough. If the above model is combined with the grid characteristics of G IS, RS technology and DEM data, and the use of Catchment ^ = water flow direction concept can effectively establish the sensitive area of the second sand disaster in the erosion factor database.泥 I mud is a theory of automatic division of water catchment area that can establish localized water catchment area. The concept of USLE and mud-sand transfer rate, development evaluation system, and graphical operating environment are used to facilitate the operation environment and sensitive area. Decision information. The sand production evaluation system of the present invention utilizes a self-propelled water catchment area as an evaluation unit, and the soil loss amount and the mud and sand production on the slopes are developed for the development platform. Water area sediment estimation is based on concepts such as

Page 9 200419045

V. Description of the invention (3) Figure 1. The estimation process is explained as follows: 1 Based on the current grid size of 40 × 40 m in D EM data in Taiwan, based on the theory of automatic division of catchment area, calculation The flow direction of non-depression, delineate the catchment area and extract the distribution of real water network, and analyze the geological and hydrological information of the catchment area as the source of spatial information needed for silt estimation. 2. Establish a soil erosion factor database using DEM and satellite imagery data, combined with G IS and RS technologies, including interpolation of rainfall erosion index and soil erosion index in the catchment area by linear Kriging method; use spatial analysis method to automatically Extract the diffuse flow length and slope of the grid and substitute it into the formula of Wischmeier and Smith (1978) to calculate the slope length and slope factor. Satellite image data and land use data are used to establish the vegetative cover factor (C) and Shuai. 1 conversion formula to quickly obtain the C-factor database. 3. Calculate the amount of soil loss in the catchment area by using the soil loss formula (USLE) and multiply it by the mud-sand transfer rate to obtain the mud-sand yield on the slope of the catchment area. The spatial distribution and quantitative information can be classified and coordinated. Satellite image classification of main land-use data in the catchment area (forest land, fruit trees, vegetables, bare land, waters and clouds), and to explore the sensitive locations of mud and sand sources as decision-making parameters for mud and sand control (I) Soil erosion model and mud transfer Rate _ 1 · Soil erosion model (1) Rainfall erosion index (Rm) The calculation of rainfall erosion index is based on the erosion index data from Huang Junde (1 79), and it is used to establish the rainfall in Taiwan by geostatistical method ° σ < annual average drop

200419045 V. Description of the invention (4) Equivalent graph of rain wash surplus index. (2) Soil erosion index (Km) The calculation of soil erosion index is based on the value of the soil erosion index in Taiwan slope land surveyed by Wan Xinsen and Huang Junyi (丨 98 i, 1 89). Methods Establish the equivalent map of soil erosion index in Taiwan. (3) Slope length factor (L)

Wi schmeier (1976) pointed out that the effective slope length is the distance from the start of the diffuse flow to the place where the slope slows down or to the channel, which usually does not include the part of the channel, and the slope length is susceptible to factors such as slope slowdown or land use change. Truncate. wi lsoon (1986) pointed out that the effective slope length of USLE is often affected by buildings, roads or artificial drainage systems, and its length will not be too long. Generally speaking, the slope of the catchment area is affected by ground storage and infiltration, as well as the interception of ravines, road drainage side ditches, etc. The length of the diffuse surface flow is rarely more than 100m, plus the general safe drainage of slopes. If it exceeds 100m, drainage is needed in sections, otherwise runoff concentration is easy to form eroded ditch. Therefore, ^ the invention uses numerical elevation model data in conjunction with the observation of drainage flow direction f 'to extract the diffuse and canal flow in the catchment area (for safety and convenience calculation, the ground If the stream length exceeds 1001 ", it can be regarded as the channel distribution). = The length of the diffuse flow, using the formula of the USLE slope length factor, from the slope length factor + of the district, a reasonable assessment of the amount of soil loss and silt production on the slope surface (Lin Zhaoyuan, Lin Wenci, 1 9 9 9). Block ΐ Because the terrain of the catchment area is estimated at 5 levels 丨 I > "" Using a numerical elevation model in Taiwan, the resolution of the grid is 40 m X 4 0 m. ^,, Week k π > Xuan can be regarded as a diffuse flow to estimate the slope length distribution of the weird water area; the cumulative flow E is considered as a channel flow when the initial heterogeneous set length exceeds two grids.

200419045 I '.. V. Description of the invention (5) (4) Slope factor (S) The slope factor is used to calculate the average slope of each grid (0) (Burrough, 1986), and then substitute S = 65.4sin20 + 4.56sin 0+ 0.0654 formula to find the S value. (5) Coverage and management factor (C)

Since there is no database available for the C value in Taiwan, the current method of obtaining the C factor has been gradually replaced by telemetry images. The use of land use interpretation results is converted to C values. However, the interpretation of images requires professional telemetry operators and It takes time to obtain the required information. If the C. value database can be generated through the assessment of green vegetation, it will effectively reduce the time required to evaluate the c value. According to the present invention, the C factor can be obtained quickly, and the coverage and management factors are calculated using the vegetation index of the telemetry theory. The conversion formula of the NDVI value and the C value is C = {(1 -NDVI) / 2} ^ (1 + NDVI ) (6) Soil and water conservation treatment factors (p) Soil and water = Factors are considered in the most dangerous state, that is, the evaluation is performed without any soil and water conservation treatment (P = 1). Silt transfer rate The mud transfer in this study is driven by the transfer of the mud on the slope. The migration to the canal is defined by the transfer rate of mud and mud. The division rate 'assumes that the main road (often flowing water) on the slope is lost. The rate of sand migration can be regarded as a grid. If the contribution of mud and sand is used to increase the net area, the analysis method is generated. For simple calculation, it is based on the assumption that the surface water of the slope surface and the mud and sand point pair flow closest to the channel channel. The area table's runoff is also larger,

Page 12 200419045 V. Description of the invention (6) The mud and sand can be brought to the river channel more and more. Therefore, the sediment transfer rate of any grid point on the slope of the catchment area can be regarded as the ratio of the upstream inflow area (A grid) of the grid point to the upstream inflow area (A tota 1) of the grid flow to the closest channel ( (SDR = Agrid / Atotal) is shown in Figure 2. This means that the closer the waterfront is to the grid point of the channel, the easier it is for the erosion of the sand on the slope to enter the channel, and increase the sediment production of the river, so the mud transfer rate is higher. Yield (Y s) on the slope of the catchment area can be calculated from the product of the slope soil loss (Am) and the slope sediment transfer rate (SDR) (Ys = Am * SDR) in the catchment area. The drainage flow direction can calculate the sediment transfer rate and the amount of sediment in each grid. 4. Sediment estimation expert system The development of this software is based on the theory of automatic division of catchment area, using the USLE model and the concept of sediment transfer rate, combining DEM data and satellite image data, using the inference interface of the expert system, and writing the program The silt volume estimation expert system (see Figures 3, 4, 5, and 6), the results of which are mainly applied to the estimation of soil erosion and slope sediment production, and the example analysis of the watershed of Deji Reservoir (see Figure 7) , Eight, nine and ten), can be used to verify the practicality of the present invention.

Page 13 200419045

Page 14 200419045 Schematic description

? ....................................... ........ 疒 、, like you? You private. Xianjiu brother called poor secret Γ ——__ · — · .. · — ^ ……… m ••• / AWAWAWAW ···· :: * »书. :: v :::: 5 C resource? • Solitary effect · ί ^ Χ: < · Gang? 、 Heart% «« Poor Cup 3 ['3 承 上 .. ## \ \ ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, SSSSVA ιΛήΛί ····, // · > ····························· ··· '· [· Figure 3. Expert system for estimating sediment volume (Rm and Km factors)

_Figure 4. Expert system for estimating the amount of mud and sand (L and S factors) iBi Page 15 200419045 Schematic illustration

Figure 5. Expert system for estimation of sediment volume (C and P factors)

_Figure 6. Expert system for estimating the amount of mud and sand (calculation of erosion depth) IHi Page 16 200419045 Schematic description

Figure VII. C value distribution of Deji reservoir catchment area

_Figure 8. Distribution of mud and sand transfer rate in the catchment area of Deji Reservoir 1B1 page 17 200419045

Soil wave loss Or secret, ki / jc〇 < m 29mm Longitudinal Na said wzm Fox articles. ¾ 11 »! 〇 (» \ mm _ > Brain! 0 period 4 (%) M-7S% ZSi% ^ 26% 475% 3% mn% Figure IX. Distribution map of soil loss in the watershed of Deji Reservoir / rf: S4 π ^ m · · ...... jiS \ ._ ~ # fenqi body, ⑼ ————: : ㈣ :: dirty ^ SS 5: S 30 ^ mm om write 骊 S 涩 S \: / m_ lean fraction ratio (%) t, / w Figure 10: Distribution of mud-second output on the slope of Deji Reservoir watershed

Page 18 200419045

1. Analysis of USLE model factors The current analysis of USLE model factors is fm and U. Most of them use spatial analysis techniques of subcatchments or the entire catchment. + Λ "π and η 9 Μ 枓 and automatic extraction of satellite image data" "and *, and the theory of automatic division of the catchment area extends the scale to the catchment area to estimate the soil erosion space in the catchment area Distribution situation. 2 · Slope silt transfer rate _ For a long time, the calculation of silt transfer rate is based on the river ±, and represents the entire catchment area with rough values. The present invention derives the concept of area ratio from the definition of silt transfer rate. Xiao He DEM data, combined with drainage flow direction data, automatically calculates the slope sediment transfer rate in a grid analysis method and multiplies the above-mentioned soil erosion amount to estimate the slope sediment yield and calculate the soil erosion depth in the catchment area. 3. Estimation of slope mud production Using the self-derived automatic catchment area division theory, taking the catchment area as the evaluation unit and combining the concepts of USLE and mud transfer rate, develop a system for evaluating soil erosion and slope mud production, and use graphics The window interface is a development platform, providing a convenient operating environment for beginners, and managers can quickly obtain decision-making information on environmentally sensitive areas in the catchment area. 4 · Since the present invention can improve the past disadvantages, it is a reliable and easy to popularize method. It can cooperate with the concept of the expert system to write a program and establish an expert system for estimating sediment volume (SedExpert).

Page 19 200419045 VI. Scope of patent application Quantitative information and spatial distribution of soil loss in the catchment area and silt production on the slope are estimated as the reference basis for sediment control and treatment in the catchment area.

Claims (1)

200419045 VI. Scope of patent application 1. A network analysis method for calculating soil erosion factor (USLE factor). The analysis of USLE model factors is mostly based on sub-catchments or the entire catchment. The grid is used as a unit, combined with the geographic analysis system's spatial analysis technology. The DEM data and satellite image data are used to automatically extract the factors required for soil surplus and the automatic catchment theory is used to extend the scale to the catchment area. Estimate the spatial distribution of soil erosion in the catchment area. 2 · —A method for calculating the rate of mud and sand transfer on slopes For a long time, the calculation of the rate of mud and sand transfer has been dominated by rivers and represents the entire catchment area with rough values. The invention derives from the definition of mud and sand transfer rate Area ratio concept, combined with DEM data, combined with drainage flow data, automatically calculates the slope sediment transfer rate by grid analysis method, multiplies the soil erosion amount mentioned above, estimates the slope sediment production and calculates the soil in the catchment area Erosion depth. 3. Kind of method for estimating the weight of mud and sand on the slope of the catchment area. Using i: Derived theory of automatic division of catchment area, taking the catchment area as an assessment of the concepts of Tan, UsiE, and the rate of mud-sand migration, develop Soil Erosion ::: = 二 > 二 = # 二 统, and using the graphical window interface as a development platform, #niandao scholars can easily choose from the use environment for 7Jc P ^ r, and managers can quickly Obtain decision-making resources 1 ~ 4 in the sensitive area of Laishui District 4 · For example, the method of item 1 to 3 of the step-by-step reasoning mechanism park for patent application, combined with the expert system to programmatically integrate the above method to establish the amount of mud and sand Page 18 2004d9045 VI. Patent application scope SedExpert, combined with DEM data and satellite image data, can quickly estimate the quantitative information and spatial distribution of soil erosion in the catchment area and the amount of mud and sand on the slope, as the catchment The reference basis for the area's sand control and treatment. Page 19