Applying Optimization and the Analytic Hierarchy Process to Enhance Agricultural Preservation Strategies in the State of Delaware
Messer, Kent D., Allen, William L., Agricultural and Resource Economics Review
Using agricultural preservation priorities derived from an analytical hierarchy process by 23 conservation experts from 18 agencies in the state of Delaware, this research uses weighted benefit measures to evaluate the historical success of Delaware's agricultural protection fund, which spent nearly $100 million in its first decade. This research demonstrates how these operation research techniques can be used in concert to address relevant conservation questions. Results suggest that the state's sealed-bid-offer auction, which determines the yearly conservation selections, is superior to benefit-targeting approaches frequently employed by conservation organizations, but is inferior to the optimization technique of binary linear programming that could have provided additional benefits to the state, such as 12,000 additional acres worth an estimated $25 million.
Key Words: conservation optimization, farmland protection, analytic hierarchy process, binary linear programming
(ProQuest: ... denotes formulae omitted.)
In the United States, conservation groups spend an estimated $3.2 billion annually (Lerner, Mackey, and Casey 2007). While operations research techniques are frequently used in a wide variety of areas, yielding substantial success, such techniques have rarely been applied to on-theground conservation efforts despite the promise of providing more conservation benefits for the same budget constraint (Prendergast, Quinn, and Lawton 1999, Rodrigues and Gaston 2002, Azzaino, Conrad, and Ferraro 2002, Messer 2006). A partial explanation for this lack of adoption is that many of the initial analyses in operations research have focused on problem setups-such as covering problems that identify the minimum number of preserves necessary to protect a set number of endangered species or the maximum number of species that could be protected with a set of protected areas (e.g., Ando et al. 1998, Balmford et al. 2001, Polasky, Camm, and Garber- Yonts 2001, Moore et al. 2004, Strange et al. 2006, Cabeza and Moilanen 2001, ReVelle, Williams, and Boland 2002)-that have little relationship to the actual priorities and problems faced by conservation organizations. Secondly, conservation objectives and goals tend to be difficult to characterize, identify, and measure, and lack a common metric for success, such as profit in business applications. Furthermore, other obstacles exist for the use of these techniques for conservation, including how to identify the true decision-space for the conservation group, which must first locate willing sellers, develop the meaning of the measures of conservation benefit, assess the relative importance of one environmental characteristic over another, and provide reliable, arm's-length estimates of the costs involved (Strager and Rosenberger 2006). In this research, we show the benefits of applying operations research techniques in a setting where these latter obstacles have already been essentially overcome given the existing program priorities of the Delaware Agricultural Lands Preservation Foundation (DALPF) and its historical data of willing sellers' offers, parcels' market appraisals, GIS information on parcels' agricultural and ecological value, and a gathering of conservation experts to help determine the relative value of different agricultural and ecological measures.
The most common approach in the economics literature for evaluating the benefits of agricultural land preservation is willingness-to-pay (WTP) surveys of the public (e.g., Bergstrom, Dillman, and Stoll 1985, Halstead 1984, Kline and Wichelns, 1996, 1998, Duke and Ilvento 2004, Ozdemir et al. 2004, Johnston and Duke 2009, Duke and Johnston 2010).1 However, other studies have explicitly examined the public's preferences for different attribute trade-offs inevitably involved in conservation settings by employing the technique of analytic hierarchy process (AHP) (e.g., Duke and Aull-Hyde 2002, Strager and Rosenberger 2006). …