White-tailed deer are extremely efficient foragers, eating highest quality vegetation first before eating that of a lower quality. An algorithm to simulate this behavior, which is based on the assumption that deer find the best quality forage [FDG+94], is illustrated below.
Each deer's daily search for food begins at its current location. If the current location contains available forage above the threshold, alpha_1, deer grazing is simulated on the constituent 100m cells within the current 500m location, effectively reducing the forage levels at both resolutions. Available forage within a 500m grid cell is a function of the actual forage level, water level, and the maximum water depth that the deer can withstand. Since maximum water depths differ for bucks, does, and does with fawns, available forage levels in the same grid cell will vary for different deer. If the available forage level in the current grid cell is less than the threshold, or there is not enough forage in the cell to satisfy the deer's maximum daily intake, which is a function of body weight, a search of the concentric squares centered at the deer's current location is performed, (see Figure 2(a)) for the nearest 500m grid cell with the maximum level of available high quality forage. If multiple grid cells along the same concentric square perimeter have the same maximum available forage level, one of them is chosen at random. The deer is then assigned to this new location and its daily travel distance incremented by the search radius (the number of grid cells between the old and new locations). If the chosen cell does not contain enough high quality forage above the threshold to satisfy the deer's maximum daily intake, grazing upon medium quality forage within the same grid cell is simulated. If the maximum forage intake is still not satisfied, the search for high quality forage will continue up to the deer's maximum moving distance.
If the maximum moving distance is reached while searching for high quality forage, but no forage above the threshold has been found within the last search perimeter, the forage search is restarted from the deer's current location. This time, however, the search will proceed among the concentric squares for a 500m cell with available medium quality forage above a new threshold, alpha_2. Due to the lower caloric content of medium quality forage, alpha_2 is larger than alpha_1.
If no grid cells with available high or medium quality forage above the thresholds are located within the maximum travel distance, the deer is forced to graze on low quality forage, and an attempt is made to place the deer randomly at a suitable habitat location on the outer perimeter of the search area (see Figure 2(b)). A suitable habitat location is a 500m grid cell with a water level below the maximum level the deer can withstand. If no suitable habitat location is found, a perimeter grid cell will be randomly selected from those not located within the habitat type ``open water''. The deer will remain at its current location only if all perimeter cells are located within this habitat type.
If the deer is not forced to consume low quality forage, the high/medium quality forage search will continue until one of three conditions is satisfied: the maximum daily intake is reached, the maximum travel distance is reached, or the maximum forage time is reached.
Each adult deer or independent fawn will forage sequentially each day of the simulation according to the above rules. Order is chosen randomly and is computed daily, so that no deer has preferential access to food sources.