The habitat suitability analysis generates suitability surface for winter and summer Northern Bobwhite Quail habitats. It
is based on a research paper "Habitat Model to Predict Landscape Use of Northern Bobwhite in Missouri"
by Loren W. Burger, Eric Kurzejeski and others, 1998.

The habitat model and online interface implementation are described in four sections:

- Bobwhite Quail Winter Habitat Model
- Bobwhite Quail Summer Habitat Model
- Bobwhite Quail Overall Habitat
- Habitat Model Implementation

Habitat suitability of a point location is determined by certain landscape parameters within the specie's mean home range. The mean home range of Northern Bobwhite during winter season is a 345-meter radius area (or 38 hectares). The contributing landscape parameters include:

- shape index of row crop fields (RC_LSI),
- edge density of woody patches (WOOD_ED),
- edge density of CRP fields (CRP_ED).

The landscape parameters are calculated through the following equations based on the FRAGSTATS model:

RC_LSI = TE_{row_crop_fields}/ [ 2 * √ ( π * TA_{row_crop_fields})]

WOOD_ED = TE_{woody_patches}/ TA_{all_landscape_patches }CRP_ED = TE_{CRP}/ TA_{all_ landscape_patches}All variables are calculated within the mean home range, where TE is the total edge length in meters and TA is the total area in hectares.

The winter habitat is estimated as below:

Y = -13.9995 + 3.1823 * RC_ LSI + 0.0542 * WOOD_ED + 0.0629 * CRP_ED

The habitat suitability index (HSI) surface depicts the suitability probability by scaling the Y value to 0 - 1 through the following equation:

HSI = Exp (Y) / [ 1 + Exp (Y) ]

Habitat suitability of a point location is determined by certain landscape parameters within the specie's mean home range. The mean home range of Northern Bobwhite during summer breeding season is a 535-meter radius area (or hectares). The contributing landscape parameters include the number of grass waterway patches (GRA_NUMP), the number of all landscape patches (LAN_NUMP), mean patch edge for CRP patches (CRP_MPE), landscape edge density of all landscape patches (LAND_ED), and mean perimeter to area ratio of fallow patches (FA_MPAR).

CRP_MPE = (TE_{CRP }) / (TN_{CRP})

LAND_ED = TE_{all_patches}/ TA_{all_patches}

FA_MPAR = [ ∑ (E_{fallow_patch}/ A_{ fallow_patch}) ] / TN_{fallow_patch }All variables are calculated within the mean home range, where TE is the total edge length in meters, TN is the total number of the patches, TA is the total area in hectares, E is the edge length in meters of a patch, and A is the area in hectares of a patch.

The summer habitat of the point location is estimated as bellow:

Y = -19.1142 - 0.2277*GRA_NUMP + 0.1471*LAN_NUMP + 0.00261*CRP_MPE + 0.0440*LAND_ED - 0.00344*FA_MPE

The habitat suitability index (HSI) surface depicts the suitability probability by scaling the Y value to 0 - 1 through the following equation:

HSI = Exp (Y) / [ 1 + Exp (Y) ]

The overall habitat of a location for Bobwhite quails is estimated by calculating the average value of its winter and summer habitats.

The habitat model algorithms are implemented using ArcView GIS software and Avenue scripting language. In order to generate a habitat suitability surface, a study area is converted to grid cells and habitat suitability is estimated for each cell. The analysis output therefore is in GRID format. In turn, the sizes of the study area and grid cells determine the processing time. While a larger study area obviously requires longer processing time, it is also time consuming when a small grid cell size is used.

The resultant habitat suitability index surface contains values ranging from 0 to 1. The values are scaled to integer values ranging from 0 to 100 for display purpose in the interface. The values are shown in 4 classes: 1) 0-25 for 'Not suitable', 2) 26-50 for 'Marginally suitable', 3) 51-75 for 'Moderately suitable', and 4) 76-100 for 'Highly suitable'.