A combinatorial harvest scheduling model was developed to improve the biodiversity conservation from plantation forests by integrating the connectivity of remnant native forest patches to allow for better movement of wildlife species while considering economic efficiency due to changes in the harvest schedule. The biodiversity goals included in the model are: (1) to create a shifting mosaic of closed-canopy stands that connect the remnant native forest patches, (2) to maximize the amount of interior forest habitat created in the shifting mosaic in each period, and (3) to limit the maximum clear fell area that can be created for any 3-year period. The economic goals are to maximize net present value from timber products while meeting a delivery schedule for the forest products.

The model was demonstrated using 8 arbitrary scenarios on a 1905 ha forested area using a simulated data set. The minimum shifting mosaic areas used in these scenarios were: 650, 850, 1050 and 1250ha. The goal for maximizing interior habitat was expressed as a perimeter goal, which compared the ratio of the perimeter of the shifting mosaic area with the perimeter of a circle with equal area. Two values were used for this analysis; one was twice the perimeter of a circle with equal area while the other set the value to 3.33 times the value of a circle with equivalent area. The highest level of protection from these scenarios, a minimum shifting mosaic area of 1250 ha with either perimeter goal, produces approximately 50% less discounted net revenue than the non spatially constrained solution. The smallest shifting mosaic area, 650 ha, lowered discounted net revenue between 16% for the 3.33 perimeter goal and 27% for the 2.0 perimeter goal. These models are useful for policy makers to begin to develop the cost and benefits from various levels of biodiversity protection.