Difference between revisions of "Infiltration:Parameter Estimates"

From Gsshawiki
Jump to: navigation, search
 
Line 45: Line 45:
 
Standard practice in developing GSSHA models is to obtain digital soil textural classification data and use these data to develop an index map of soil types.  Soil textural maps may be combined with land use or vegetation maps.  Land use and vegetation can strongly influence soil hydraulic properties.  The Mapping Table is used to assign initial parameters to the soil types in the index map.  One or more of these parameters, typically ''K<sub>s</sub>'' and [[Image:Psi.gif]]''f'' or [[Image:Psi.gif]]''<sub>b</sub>'', are used as calibration parameters.  As discussed by Senarath et al (2000), calibration is best done using an automated calibration method, such as SCE (Duan et al, 1992), combined with long term simulations.  The possible parameter values are bounded by the range found in literature values, unless other factors, such as land use or vegetation, dictate otherwise.  The range of values may be narrowed by making field and laboratory measurements of parameters.
 
Standard practice in developing GSSHA models is to obtain digital soil textural classification data and use these data to develop an index map of soil types.  Soil textural maps may be combined with land use or vegetation maps.  Land use and vegetation can strongly influence soil hydraulic properties.  The Mapping Table is used to assign initial parameters to the soil types in the index map.  One or more of these parameters, typically ''K<sub>s</sub>'' and [[Image:Psi.gif]]''f'' or [[Image:Psi.gif]]''<sub>b</sub>'', are used as calibration parameters.  As discussed by Senarath et al (2000), calibration is best done using an automated calibration method, such as SCE (Duan et al, 1992), combined with long term simulations.  The possible parameter values are bounded by the range found in literature values, unless other factors, such as land use or vegetation, dictate otherwise.  The range of values may be narrowed by making field and laboratory measurements of parameters.
  
Impervious areas can be accounted for by modifying the infiltration parameters as described above, or the area available for infiltration into a cell can be adjusted by specifying the impervious fraction using the '''AREA_REDUCTION''' process and including an index map for impervious area, and a mapping table with the fraction impervious for each value in the associated index map. See the section on Mapping Tables.  
+
Impervious areas can be accounted for by modifying the infiltration parameters as described above, or the area available for infiltration into a cell can be adjusted by specifying the impervious fraction using the '''AREA_REDUCTION''' process and including an index map for impervious area.  Include the '''AREA_REDUCTION''' card in the project file and an '''AREA_REDUCTION''' mapping table with the fraction impervious for each value in the associated index map.   Also see the section on Mapping Tables.  
 
<noinclude>
 
<noinclude>
 
{{Nav|Nav7}}
 
{{Nav|Nav7}}
 
</noinclude>
 
</noinclude>

Latest revision as of 19:07, 7 August 2020

It is best to use soil property and Green-Ampt infiltration parameters derived from field and laboratory measurements of infiltration on the study watershed. Even under controlled conditions hydraulic soil property measurements are very difficult. Hysteresis effects and the extremely non-linear behavior of soil water retention make it very difficult to uniquely identify soil infiltration parameters. Hydrologic studies seldom have budgets sufficient to determine the needed parameters in the field.

Considerable prior research has been performed to relate soil infiltration parameter values to textural classification. Some highly relevant references are Rawls and Brakensiek (1983) and (1985), and Rawls et al. (1982) and (1983). Table 9 summarizes Rawls and Brakensiek soil parameter estimates as a function of United States Department of Agriculture (USDA) textural classification. It is important to note that the values listed in Table 9 were derived from the geometric mean of tests on a large number of soil samples. Hydraulic conductivies for all GA based approaches are half of the saturated values listed in Table 9 (Rawls, et al., 1982). The variance of these values is large, indicating significant uncertainty or low correlation between textural classification and soil texture. However, these values are useful because they provide an initial estimate of infiltration parameters. The variances of the values in Table 9 are listed in the original papers, and are published in Maidment (1993).

USDA
Textural
Classification
Total Porosity/Saturation  
θs
  (cm3/cm3)
Effective Porosity/Saturation  
θe
  (cm3/cm3)
Field Capacity Saturation  
θf
  (cm3/cm3)
Wilting Point Saturation  
θwp
  (cm3/cm3)
Residual Saturation  
θr
  (cm3/cm3)
Bubbling Pressure Geometric Mean  
Psi.gifb
(cm)
Pore Size Distribution Arithmetic Mean  
λ
  (cm/cm)
Saturated Hydraulic Conductivity (multiply by 0.5 for GA methods)  
Ks
(cm/h)
Wetting Front Suction Head (Capillary Head)  
Psi.gifƒ
(cm)
Sand 0.437 0.417 0.091 0.033 0.02 7.26 0.694 23.56 4.95
Loamy sand 0.437 0.401 0.125 0.055 0.035 8.69 0.553 5.98 6.13
Sandy loam 0.453 0.412 0.207 0.095 0.041 14.66 0.378 2.18 11.01
Loam 0.463 0.434 0.27 0.117 0.027 11.15 0.252 1.32 8.89
Silt loam 0.501 0.486 0.33 0.133 0.015 20.79 0.234 0.68 16.68
Sandy clay loam 0.398 0.330 0.255 0.148 0.068 28.08 0.319 0.30 21.85
Clay loam 0.464 0.390 0.318 0.197 0.075 25.89 0.242 0.20 20.88
Silty clay loam 0.471 0.432 0.366 0.208 0.040 32.56 0.177 0.20 27.30
Sandy clay 0.430 0.321 0.339 0.239 0.109 29.17 0.223 0.12 23.90
Silty clay 0.479 0.423 0.387 0.250 0.056 34.19 0.150 0.10 29.22
Clay 0.475 0.385 0.396 0.272 0.090 37.30 0.165 0.06 31.63

Table 9 - Rawls & Brakensiek soil parameter estimates.

In the table soil moistures θ, are listed for saturation (s), effective saturation (e), field capacity (f), wilting point (wp), and residual (r). These values are applicable for all approaches. These are followed by the bubbling pressure Psi.gifb (used for RE), the pore distribution index (used for RE, GAR and multi-layer GA in continuous mode), saturated hydraulic conductivity Ks (used directly for RE, halved for all GA approaches), the wetting front suction head Psi.giff (used for all GA approaches).

Standard practice in developing GSSHA models is to obtain digital soil textural classification data and use these data to develop an index map of soil types. Soil textural maps may be combined with land use or vegetation maps. Land use and vegetation can strongly influence soil hydraulic properties. The Mapping Table is used to assign initial parameters to the soil types in the index map. One or more of these parameters, typically Ks and Psi.giff or Psi.gifb, are used as calibration parameters. As discussed by Senarath et al (2000), calibration is best done using an automated calibration method, such as SCE (Duan et al, 1992), combined with long term simulations. The possible parameter values are bounded by the range found in literature values, unless other factors, such as land use or vegetation, dictate otherwise. The range of values may be narrowed by making field and laboratory measurements of parameters.

Impervious areas can be accounted for by modifying the infiltration parameters as described above, or the area available for infiltration into a cell can be adjusted by specifying the impervious fraction using the AREA_REDUCTION process and including an index map for impervious area. Include the AREA_REDUCTION card in the project file and an AREA_REDUCTION mapping table with the fraction impervious for each value in the associated index map. Also see the section on Mapping Tables.

GSSHA User's Manual

7 Infiltration
7.1     Richards’ Equation
7.2     Green and Ampt (GA)
7.3     Multi-layer Green and Ampt
7.4     Green and Ampt with Redistribution (GAR)
7.5     Parameter Estimates