Establishing marine-protected areas (MPA) is important for maintaining biodiversity and protecting endangered species. MPAs can also effectively inhibit human interference, such as tourism and pollution. MPA implementation is a feasible measure to fulfill the sustainable management goal that is to conserve marine habitats for achieving an integrative ecosystem and higher biodiversity. However, how to design an MPA under these concerns remains an important research issue to be explored. A spatial resource allocation modeling approach has been developed by integrating some earlier study efforts. The integer linear programming technique was used to develop the models, and they account for the spatial attribute of the compactness in terms of boundary length. Two models (the minimal protected area model and the maximal biodiversity conservation model) were formulated to address 2 major MPA design objectives. The models show the reciprocal relationship between MPA compactness and the 2 objectives: area size and the biodiversity value. In particular, the modeling results show that a more compact MPA contains less biodiversity. This tradeoff relationship, which is easily explored using the spatial resource allocation modeling approach, enables an MPA design to allow decision makers to select various non-inferior solutions to meet their decision preferences. This study uses the Kaomei coastal wetland in Taiwan as a case study to demonstrate the proposed modeling approach.