VIC参数定义

annual_prec
VIC uses the regression published by Kimball et. al. to improve the estimation of the dew point temperature in dry regions. This regression requires the average annual precipitation in mm for each grid cell. If the value of annual_prec is negative, the model will use the daily minimum temperature without modification for the dew point temperature.

avg_T
This parameter is the temperature of the soil at the damping depth. This temperature is often assumed to be the same as the average annual air temperature. This temperature is used as the bottom boundary of all thermal flux calculations made for the soil column.

b_infilt
The b_infilt parameter is the parameter used to describe the Variable Infiltration Curve. This is typically a value that is adjusted during the calibration of the VIC model. Parameter values range from 10-5 to 0.4. Higher values will produce more runoff. 0.2 is often used as a starting value.

bubble
The bubble parameter is the bubbling pressure, h, for the soil texture type (see, e.g., Table 5.3.2 in Rawls, et al (Handbook of Hydrology)). This parameter is necessary for running the VIC model with FULL_ENERGY==TRUEor FROZEN_SOIL==TRUE. Values must be > 0.0.

If you have a VIC soil parameter file created for water-balance mode runs, in which bubbling pressure has been set to a “nodata” value such as -99, you will not be able to use this soil parameter file for FULL_ENERGY==TRUE or FROZEN_SOIL==TRUE . However, a quick way to estimate bubbling pressure from the existing soil parameters (namely the expt parameter) is:

bubble = 0.32*expt + 4.3

This is illustrated in figure 1, generated by taking the data from table 5.3.2 in Rawls, et al (Handbook of Hydrology), computing expt from lambda, and performing a linear regression.

Reference: Rawls et al., Infiltration and Soil Water Movement, In: Handbook of Hydrology, D. Maidment (ed.), 1993

bulk_density
The bulk density in layers used in the VIC model can be compiled from data available through Penn State University at
http://dbwww.essc.psu.edu/geotree/dbtop/amer_n/us_48/data/soilprop/statsgo_geo/bd/doc.html

This reports bulk density for each of 11 individual layers, which cover a depth of 2.5 meters. The majority of data does not extend beyond 1.5 meters, which includes the first 9 of their 11 layers. The Penn State data is available in gridded form at a resolution of 30 arc-seconds.

The gridded bulk density data is imported into Arc/Info, where it is averaged for the appropriate individual layers and aggregated to the grid cell size.

cellnum
Cell numbers are assigned to each grid cell beginning with 1 at the upper left corner of the basin, and numbering consecutively by rows. Cell numbers will be referenced in soil files, as well as in the vegetation parameter file.

depth
Depth is the depth of each layer in meters. Though this can be changed in the calibration process, initially it can be set to the breakdown of layer depths selected by combining the 11 individual soil layers of the Penn State University STATSGO data (for example, the Ksat data).

dp
This is the soil theraml damping depth. It is defined as the depth in the soil column at which the soil temperature remains nearly constant annually. This is the depth to which soil thermal flux calculations will be made, and is often set to 4m. The constant temperature at this boundary is defined with the parameter avg_T.

Ds
The soil parameter Ds represents the fraction of the Dsmax parameter at which non-linear base-flow occurs. This is typically a parameter that is adjusted during the calibration of the VIC model. An initial value of 0.001 may be used. Typically this value is small (less than 1).

Dsmax
The parameter Dsmax is the maximum velocity of baseflow for each grid cell. This can be estimated using the saturated hydraulic conductivity, Ksat, for each grid cell multiplied by the slope of the grid cell. The values for Ksat can be averaged for the layers for which baseflow will be included. When working in decimal degrees, the elevation data for the basin should be projected to an equal area map projection, in order to have horizontal dimensions in the same units as the vertical dimensions so that the slopes computed in Arc/Info are meaningful values.

expt
The Exponent (expt) parameter is the exponent, n, from the Brooks-Corey relationship (see, e.g., Table 5.1.1 in Rawls, et al). Values can be estimated from the table of soil hydraulic properties indexed by soil texture (see table here). Values should be > 3.0.

Reference: Rawls et al., Infiltration and Soil Water Movement, In: Handbook of Hydrology, D. Maidment (ed.), 1993

init_moist
Initial moisture content of each layer can be set at any reasonable value. One approach is to use fractional soil moisture content (expressed as a fraction of the maximum soil moisture; max. soil moisture = porosity * layer depth) at the critical point, Wcr, which can be computed for each layer as a depth in millimeters by multiplying Wcr by the thickness of the layer in meters, and then multiplying by 1000.

Ksat
The parameter Ksat is the saturated hydraulic conductivity for each layer used in a VIC model. Ksat is approximated using data on soil textures available for the basin.

Soil textures are from the STATSGO database (USDA NRCS), which has been compiled from a state-by-state format into a U.S. database, available through Penn State University at http://dbwww.essc.psu.edu/geotree/dbtop/amer_n/us_48/data/soilprop/statsgo_geo/soiltext/doc.html

This reports 16 different soil texture classes for 11 layers, which cover a depth of 2.5 meters. The majority of data does not extend beyond 1.5 meters, which includes the first 9 of their 11 layers. The Penn State data is available in gridded form at a resolution of 30 arc-seconds.

The soil texture data is then imported into Arc/Info, where it is indexed to values of Ksat according to an index table soil hydraulic properties (see sample table here). The data in this table are placed into an INFO file in Arc/Info, and the individual relates are then created to be used in an Arc/Info macro to create the Ksat grid for each layer.

MIN_WIND_SPEED
MIN_WIND_SPEED defines the minimum observable wind speed, which is then used to correct the model forcing data. Anemometers typically have a minimum observable wind speed on the order of 0.5 m/s. When forcing the VIC model with hourly wind speeds, the model’s surface flux estimates have been observed to “hiccup” (especially in dry climates), while this does not impact the long term energy balance, it will be very noticeable when looking at daily fluxes. Therefore, VIC will use MIN_WIND_SPEED to restrict the minimum value of wind speed used to force the model.

Default value is set to 0 m/s, which works well when the model is forced with daily or monthly winds.

NLAYER
NLAYER is the number of soil moisture layers to be used by the model. Typically the number of layers is 2 or 3. The model cannot be run in energy balance mode with fewer than 3 layers (the top layer is thin 5-15cm), but the water balance model used to have only 2 layers. The model has not been tested with more than 3 layers, so problems may develop.

NODES
NODES defines the number of soil thermal nodes used by the model to explicitly solve soil thermal fluxes. At this time the explicit solution of the soil thermal fluxes occurs only when the model is run with the frozen soil algorithm activated. The model uses a node at the surface, a node at the bottom of the thin (5-15cm) top layer, and a node at the damping depth. If more nodes are defined they are distributed evenly between the bottom of the top layer and the damping depth. More nodes improves the accuracy of the soil heat flux solution, but also increases the computational time. We recommend at least 5 nodes.

off_gmt
off_gmt is the number of hours difference from GMT (aka UTC). These are the number of hours to be added to GMT/UTC to obtain local time for each grid cell. Sample offsets are listed in the table below. since UTC does not use any daylight savings adjustment, the gmt offset will decrease by one hour for those areas using daylight savings. For example, Eastern Standard Time (EST) changes from GMT-5 hours to Eastern Daylight Time (EDT) at GMT-4 hours.

Table of GMT Offsets in hours

Zone	City or Area Zone Crosses
GMT-12	Eniwetok
GMT-11	Samoa
GMT-10	Hawaii
GMT-9	Alaska
GMT-8	PST, Pacific US
GMT-7	MST, Mountain US
GMT-6	CST, Central US
GMT-5	EST, Eastern US
GMT-4	Atlantic, Canada
GMT-3	Brazilia, Buenos Aries
GMT-2	Mid-Atlantic
GMT-1	Cape Verdes
GMT	Greenwich Mean Time, Dublin
GMT+1	Berlin, Rome
GMT+2	Israel, Cairo
GMT+3	Moscow, Kuwait
GMT+4	Abu Dhabi, Muscat
GMT+5	Islamabad, Karachi
GMT+6	Almaty, Dhaka
GMT+7	Bangkok, Jakarta
GMT+8	Hong Kong, Beijing
GMT+9	Tokyo, Osaka
GMT+10	Sydney, Melbourne, Guam
GMT+11	Magadan, Soloman Is.
GMT+12	Fiji, Wellington, Auckland

• Table copied from Texas A&M University Institute for Scientific Computation Web Page.

phi_s
The parameter phi_s is the unitless soil moisture diffusion coefficient. This parameter is designed for the future inclusion of soil moisture diffusion in the VIC model moisture trasport equations. Currently this feature has not been implemented so a value of -999 can be used as a placeholder.

resid_moist
Residual moisture content is the amount of soil moisture that cannot be removed from the soil by drainage or evapotranspiration. Values are provided to the model as soil moisture contents (volume of residual soil moisture content / total volume of soil) [mm/mm]. When residual soil moisture is defined as 0 mm/mm the soil hydraulic conductivity relationship collapses to Campbell (1974), otherwise it follows Brooks and Corey (1964).

Brooks, R. H. and A. T. Corey, Hydraulic Properties of Porous Media, Hydrology Paper 3, Colorado State University, Fort Collins, Colo, 1964.

Campbell, G. S., A Simple Method for Determining Unsaturated Conductivity from Moisture Retention Data, Soil Sci., vol. 117, pp. 311-314, 1974.

ROOT_ZONES
ROOT_ZONES is the number of root zones used in the vegetation parameter file to define the root distribution. This allows the rooting distribution to be defined independently of both vegetation type and soil layer depth. When soil moisture layer depths are changed during calibration, the model uses linear distribution based on the defined root zones to redistribute root fractions.

No default value is set, this parameter must be defined in the global parameter file.

rough
Surface roughness of bare soil, expressed in meters, can be set to a value 0.001, and adjusted according to local data.

snow_rough
The surface roughness of the snowpack, expressed in meters, can be set to an initial value of 0.0005, and can then be adjusted according to local data.

Wcr_FRACT
The parameter Wcr_FRACT (Wcr) is the fractional soil moisture (expressed as a fraction of the maximum soil moisture; max. soil moisture = porosity * layer depth) at the critical point, which is the water content below which hydraulic conductivity begins to fall below saturated values, as does transpiration. This is set at 70% of the field capacity, in accordance with the different soil textures. Field Capacity is defined as the water content at a tension of -33kPa.

Soil textures are from the STATSGO database (USDA NRCS), which has been compiled from a state-by-state format into a U.S. database, available through Penn State University at http://dbwww.essc.psu.edu/geotree/dbtop/amer_n/us_48/data/soilprop/statsgo_geo/soiltext/doc.html

This reports 16 different soil texture classes for 11 layers, which cover a depth of 2.5 meters. The majority of data does not extend beyond 1.5 meters, which includes the first 9 of their 11 layers. The Penn State data is available in gridded form at a resolution of 30 arc-seconds.

The soil texture data is then imported into Arc/Info, where it is indexed to values of Wcr according to the table SOILTEXT. The data in this table are placed into an INFO file in Arc/Info, and the individual relates are then created to be used in an Arc/Info macro to create the Wcr grid.

wind_h
The measurement wind height is the height above the ground at which wind observations were recorded. Typically wind observations are taken at a height of 2 m in clearings. However, the VIC model assumes wind observations are made above the vegetation cover in each grid cell and that wind speeds through and below the canopy can be found using exponential and logarithmic wind profiles. For short vegetation types (grasses, shrubs, etc.), the wind measurement height should be ~2 m above the top of the vegetation. For tall vegetation types (trees), especially those defined with an overstory, the wind measurement height should be set much higher than the vegetation. This is physically justified because a forest canopy impacts atmospheric stability to a much higher elevation, than shorter vegetation. If the model is required to run with an overstory defined, and the wind height set to just above the vegetation (for example the tree height + 2m) large negative evaporations and sublimations may be found during the winter months, as the snow interception algorithm is forced to use unrealistically strong wind speeds.

When defining a vegetation type with overstory, it is strongly recommended that wind_h be defined at least 10 m higher than the height of the vegetation.

Wpwp_FRACT
The parameter Wpwp_FRACT (wp) is the fractional soil moisture (expressed as a fraction of the maximum soil moisture; max. soil moisture = porosity * layer depth) at the wilting point. Wilting Point is set at the water content at a tension of 1500 kPa, and is approximated for the different soil textures.

Soil textures are from the STATSGO database (USDA NRCS), which has been compiled from a state-by-state format into a U.S. database, available through Penn State University at http://dbwww.essc.psu.edu/geotree/dbtop/amer_n/us_48/data/soilprop/statsgo_geo/soiltext/doc.html

This reports 16 different soil texture classes for 11 layers, which cover a depth of 2.5 meters. The majority of data does not extend beyond 1.5 meters, which includes the first 9 of their 11 layers. The Penn State data is available in gridded form at a resolution of 30 arc-seconds.

Ws
The parameter Ws is the fraction of maximum soil moisture where non-linear baseflow occurs. As with the Ds parameter, this is generally adjusted during the calibration phase of applying the VIC model. Values for Ws are typically greater than 0.5. An initial value of 0.9 can be used.