SEARCH

SEARCH BY CITATION

Keywords:

  • Species protection;
  • conservation planning;
  • costs;
  • priority setting;
  • resource allocation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Conservation spending in California, USA exceeds conservation expenditures in many countries. To date, there has been no objective method to assess the efficiency of such spending for achieving species conservation outcomes. We conducted the first such retrospective analysis of conservation spending, examining the distribution of $2.8 billion spent on land protection by the state of California and partners from 1990 to 2006. Using a return on investment algorithm with species protection as the sole objective, we describe a “cost-efficient” funding scenario that would have protected four times more distinct species and three times more threatened and endangered species compared to the observed allocation. Differences between the species-diversity spending and the observed spending patterns reflect the myriad funding objectives, beyond protecting species, of the state. Identifying cost-effective conservation strategies are essential given the need to maintain species diversity in the face of global change.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

California, USA, is one of the most species-rich regions in the world (Dallman 1998). It is also relatively populous and wealthy, with large expenditures on conservation. In the last 16 years with funding from state, federal, and other sources, California has spent at least $2.8 billion—an average of $177 million per year—on acquiring land and conservation easements that protect habitat and provide other ecosystem services. By comparison, conservation expenditure by the federal government in Brazil was $88 million in 2000 (Young & Roncisvalle 2002). Such expenditures could result in extraordinary conservation outcomes for species protection.

Allocation of funds to date have been based on multiple criteria, including the mission and objectives of the funding agencies, existence of predetermined biodiversity priorities (Halpern et al. 2006), effectiveness of past conservation effort (Kareiva & Marvier 2003), the presence of opportunity and ability to leverage other efforts, and limitations of the funding sources themselves. In practice, decisions are rarely based on species protection alone. Regardless of the intended goal, however, methods to measure progress toward goals are needed—especially in places like California where the public is a major source of funding.

Conservation spending for land acquisition in the USA tends to correlate with factors such as population or proximity to coast (Lerner et al. 2007). This may not necessarily deliver the greatest species protection gains. Often, conservation decisions are made without adequate data (Doak & Mills 1994) or rigorous evaluation of the inputs highly relevant to the investment decision, like economic cost. The inclusion of cost in evaluating conservation investments is of crucial importance as the demand for conservation resources exceeds the supply (Balmford et al. 2000; Odling-Smee 2005; Naidoo et al. 2006). Several studies have demonstrated that incorporating information on costs and biodiversity benefit could in theory provide a more cost-efficient allocation of limited conservation resources for achieving that specific objective (Ando et al. 1998; Costello & Polasky 2004; Wilson et al. 2006; Murdoch et al. 2007).

We present the first comparison between observed and optimal patterns of conservation expenditures for achieving species protection outcomes to gauge the cost-efficiency of conservation investments. We identify investment priorities for the protection of species diversity in California, at the scale of the state's counties, between 1990 and 2006. Specifically, we (1) measure the cost-efficiency of observed conservation spending for protecting the number of threatened and endangered and distinct species, as captured in a current state program and compare it to an optimally efficient funding scenario; and, (2) investigate possible explanations for the observed patterns using population density, coastal proximity, and the species diversity and rarity characteristics of each county. This study advances previous return on investment (ROI) applications by conducting a retrospective comparison and using available data on species, costs, and current protection at a scale that is appropriate for informing practical conservation decisions.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

In California, public funding is often initiated through the legislative process and voted for by the populace in the form of bonds. Bond funds support a number of state agencies and programs, with mandates that range from providing recreation opportunity to protecting wildlife habitat and watersheds. Many funds are directed to land acquisition and conservation easements, which can also provide species conservation returns. Expenditure of state funds typically requires that there be matching contribution(s) from other sources and partners. Here, we examine patterns of conservation spending, using public records that detail location, size, and price of the majority of the land acquisition projects from 1990 to 2006 by one state entity and its partners (Wildlife Conservation Board 2007).

The state of California comprises 58 counties. For each county that received land protection funding (n= 52), we summarized the funds spent by the state and its partners (e.g., nongovernmental organizations, local, state, and federal agencies) from 1990 to 2006 and conducted a pairwise correlation of conservation spending by these two funding sources. Observed conservation spending in each county was regressed against (1) the population density of each county (U.S. Census Bureau 2000), (2) the distance of the midpoint of each county from the coast, (3) the number of all plant and vertebrate species, and (4) the number of threatened and endangered species in the county (California Department of Fish and Game 2005; CalFlora 2007). Spending and (log) species data were corrected for the size of each county and nonspecies variables were also log transformed to satisfy the demands of parametric statistical tests (analyses were conducted in JmpIn5.1, SAS Institute, Cary, NC, USA).

We used two species metrics to measure the efficacy of conservation spending: the number of native vascular plants and vertebrates, and the number of state-listed threatened and endangered plants and vertebrates. Both lists were compiled for each county at the species level. For each county, data on plants were compiled from CalFlora (CalFlora 2007) and filtered by native species limited in distribution to California. Native vertebrate data were compiled from the Wildlife Habitat Relations database (California Department of Fish and Game 2005). Both datasets were further filtered to determine the presence or absence of California-listed threatened and endangered species by county.

For each county we compiled (1) the area (km2) currently in public ownership in protected areas, or in conservation easements held by nongovernmental conservation organizations (California Resources Agency Legacy Project 2005; The Nature Conservancy 2005); (2) the cost of acquiring land (dollars per km2) based on the total project acquisition costs and project area contained in the database; (3) the area of land available for acquisition (i.e., land which is neither converted to urban, nor high intensity agriculture, protected, nor in public ownership, calculated from land cover data (California Department of Forestry and Fire Protection 2002); and (4) the projected rate of habitat loss for each county determined using two sources of land cover change detection data between two dates (Table 1 and see Appendix S1 and Table S1 in Supplementary Material). To estimate the amount of protection in 1990, we subtracted the amount of land purchased by the state between 1990 and 2006 in each county from the total public and protected land in 2006. We assumed the proportional rate of habitat loss remained constant over the 16 years, that the effectiveness of conservation investment was similar across all counties, and that the species protection return with increasing area acquired was represented by the species-area relationship using a "z" exponent of 0.2 (Rosenzweig 1995).

Table 1.  Characteristics of priority counties for conservation investment in California, USA at the outset of applying a cost-efficient funding approach from 1990 to 2006 with an annual budget of $177 million
CountyHabitat loss rate (%yr−1)Area (km2)% Natural (1990 Estimate)% Available (1990 Estimate)% Protected (1990 Estimate)Cost/km2Spp. RichnessT&E Spp. RichnessROI Obj.
  1. "Spp. richness" is the number of native plant and vertebrate species and "T&E Spp. richness" is the number of state-listed plant and vertebrate species. "Obj" (Objective) codes: 1 = selected with species richness as objective, 2 = T&E species richness, and 3 = selected by both.

Calaveras0.122,68399%80%19%$144,164610171
Plumas0.436,76996%26%70%$80,760475121
Siskiyou0.1316,436 95%33%61%$113,456519161
Tehama0.097,66092%62%30% $76,933657171
Lake0.003,44493%47%46%$338,265598182
Napa0.302,04281%70%11%$766,770612212
Sacramento0.362,53139%39%<1%$451,826392152
Sonoma0.334,10076%71% 6%$903,518701312
Yolo0.102,64540%38% 2%$272,675432162
Colusa0.082,99653%39%14% $57,427523143
Mariposa0.013,786100% 48%52%$102,909694163
Monterey0.148,56585%58%27%$335,958843323
San Luis Obispo0.268,57692%71%20%$330,122814313

To determine the most cost-efficient allocation of funding over the 16 years, we used one of the ROI allocation algorithms described in Wilson et al. (Wilson et al. 2006). The “maximize gain” algorithm directs investment at each time-step to the county where the most new distinct species can be protected given a fixed budget (without consideration of the rate of loss of species). By accounting for the complementarity of species (Underwood et al. 2008), we avoid the double-counting of species that is prevalent in other studies (e.g., Wilson et al. 2006; Murdoch et al. 2007). For each year, the output provides the amount of funds allocated, area acquired, and the number of species protected in each county—assuming that land acquisition results in immediate and successful protection of species. We specify a yearly conservation budget that reflects the average amount of conservation spending by the state and its partners for land acquisition from 1990 to 2006—$177 million per annum.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

The number of native vascular California-endemic plants and vertebrates totals 2,274 species, of which 162 are state-listed as threatened or endangered (Table S1). Approximately 190,500 km2 (51% of the total study area) is public or protected land. Using the species-area curve, the current public and protected land is estimated to support 1,930 distinct species (85% of the total) and 132 threatened and endangered species (81% of the total) in the study area.

Observed conservation spending and species returns

Between 1990 and 2006, the state spent $1.6 billion on land acquisitions and easements that leveraged an additional $1.3 billion from partners—a total of $2.8 billion. This spending resulted in a total of 3,254 km2 of land acquired in 723 separate projects. Fee title acquisitions comprised 83% the projects and 96% of the dollars spent, with the remainder being conservation easements. Levels of investment in each county by partners was similar to that invested by the state—for example, Humboldt County accounted for 17% of the state's total investment and 21% of the partners' investment. There was evidence for a pairwise correlation between state and partner spending by county (r= 0.67 and P < 0.0001).

Over the last 16 years, five of the six counties with the highest funding allocation were located in densely populated southern California: Los Angeles (10% or $289 million), San Diego (9% or $248 million), Riverside (8% or $223 million), Ventura (7% or $188 million), and Orange (5% or $152 million) (Figure 1). Funding allocation varied considerably between counties: 31 counties (53%) received less than 1% each of the total funding over the 16 years. We estimate that the observed spending pattern of the $2.8 billion secured the protection of 12 distinct plant and vertebrate species and two threatened and endangered species.

image

Figure 1. Observed conservation spending for land acquisition in California, USA (conservation easement and fee title) from 1990 to 2006. Numbers in counties are dollars ($ millions) spent.

Download figure to PowerPoint

Conservation spending by the state and partners was best explained by the proximity of each county to the coast (r2= 0.58, P < 0.0001) with coastal counties receiving greater funding, along with more densely populated counties (r2= 0.33, P < 0.0001). A positive relationship was also found between conservation spending and the number of threatened and endangered species (r2= 0.24, P= 0.0002), but not between conservation spending and the total species richness.

Cost-efficient conservation spending to maximize species returns

In contrast to observed spending, if the conservation objective was to maximize the protection of distinct native species, then applying the Wilson et al. (2006) algorithm for “cost-efficiency” would allocate funding to 8 of the 52 counties over the 16 years, acquiring 18,524 km2 (Figures 2a and 3). This scenario would protect 52 distinct plant and vertebrate species, compared to the estimated 12 species protected by the observed allocation. Priority counties would include Monterey (receiving $1.1 billion, 38%) and San Luis Obispo and Siskiyou (both receiving $487 million, 17%). The remaining five counties receive between 2% and 13% of the total funding after 16 years. Alternatively, if the conservation objective was to maximize the protection of threatened and endangered species, then the most cost-efficient allocation would result in 7,813 km2 acquired across nine counties (Figure 2b). After 16 years, almost 40% of the total funding would be allocated to Sonoma County ($1.1 billion) followed by Monterey ($487 million, 17%) with the remaining counties receiving between 2% and 13% of total funding. This funding scenario would protect five distinct threatened and endangered species compared to the estimated two species protected by the observed spending.

image

Figure 2. Cost-efficient funding scenario for land acquisition in California, USA from 1990 to 2006 with an annual conservation budget of $177 million. Counties selected with an objective of maximizing the gain of; (A) distinct native plant and vertebrate species and (B) California-listed threatened and endangered species. Numbers in counties are dollars ($ millions) spent.

Download figure to PowerPoint

image

Figure 3. Comparison of the observed conservation spending and a cost-efficient funding scenario which seeks to maximize the protection of all native species and threatened and endangered species from 1990 to 2006 with an annual budget of $177 million.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Our retrospective analysis measured the cost-efficiency of species conservation investments from 1990 to 2006 in the state of California. With partners, the state spent US$2.8 billion to protect conservation lands that provided open space, habitat, recreation, and other ecosystem services in California. Our analyses indicated that this spending tended to densely populated coastal counties which, possibly as a result of this density, also have high numbers of threatened and endangered species. If, over the same timeframe, the investment objective was solely to maximize the number of distinct species protected, then the same expenditure might have resulted in over 15,000 km2 more land and 40 more distinct species protected. If, alternatively, the objective was solely to maximize the protection of threatened and endangered species, then the same expenditure might have resulted in 1,500 km2 more land and three more species protected. A striking difference between the observed and the “cost-efficient” scenario for species conservation is the focused geographic distribution of funding (see Figure 1 versus Figure 2a and 2b). On the one hand, spreading investments has benefits: protecting local populations of species; providing recreation opportunities to more people; providing direct evidence of land conservation for voters. On the other hand, the cost-efficient scenario illustrates how high species protection gains can be made by investment in a limited number of places. Over 90% of the observed spending ($2.8 billion) was spent in counties that would be considered nonpriorities if the objective was solely to maximize the protection of distinct species richness at the least cost.

Priority counties in the cost-efficient scenario are not necessarily characterized by highest species richness but feature reasonable land costs and low existing protection, resulting in substantial conservation returns. For example, Colusa County received funding in both scenarios (Figure 3) despite having relatively low species richness (ranked 32nd and 17th for all species richness and threatened and endangered species richness, respectively). However, this county has both low pre-existing protection (14%) and the lowest land costs of any county. Returns on each dollar spent are therefore very high. In contrast, a combination of high land costs and/or high protection means some counties that harbor high species richness do not receive funding. For example, San Bernardino, characterized by the eighth highest species richness and moderate land costs, did not receive funding since 80% of the county was already under federal or state management. The diminishing returns incorporated into the species-area relationship mean that the benefit of additional protection in well-protected counties is limited. Alternatively, Los Angeles County (ranked seventh in species richness) was not prioritized by the cost-efficient allocation scenario for species protection because, among other factors, land costs are the second highest of any county in the state: $12 million/km2. High land cost represents high demand for land, and therefore may reflect greater immediate risks to existing biodiversity.

The discrepancy between the observed conservation spending and the cost-efficient scenario is likely to result from a number of factors. Observed spending reflects the multiple mandates of the funding agencies; species protection is rarely the exclusive priority. Oftentimes species conservation efforts must incorporate complementary outcomes achieved with funds dedicated to other purposes, such as the protection of watersheds, agricultural resources, recreational areas, open space, and the protection of other ecosystem services. The California Resources Agency's Wildlife Conservation Board, for example, has a mandate to promote recreation and as a result strategic land acquisitions occur close to population centers or in scenic areas that provide recreational opportunities. The funding allocated to Humboldt County, for example, was to secure the world's largest remaining stand of unprotected old-growth coastal redwoods (Sequoia sempervirens); this provided not only habitat for endangered species, but also recreational and aesthetic values to the public. Other bond funds were earmarked for the protection of specific ecosystems or properties; funds allocated within such specific programs, for example, the Oak Woodland Program, would only be dispensed to counties that meet the funding requirements. We also note that we only considered a subset of the total money spent on conservation in California. If we had data from all public and private sources, it would likely provide a modified picture of conservation spending in California. Similarly, using units of analysis at the scale of the acquisitions rather than county jurisdictions would modify the results and better reveal correlates of spending. Further, we emphasize that we excluded six counties from our analysis because they received no funding from the state; these counties could potentially offer even better investment opportunity for species returns than those identified here.

Future applications of this approach should incorporate other considerations relevant to conservation investment decision making. For example, counties could be prioritized according to threats from development (i.e., in terms of minimizing species loss), which may be especially important in rapidly developing regions such as California. Because most funding agencies have multiple objectives that are mandated by law, further refinements of tools like that presented here must focus on maximizing multiple objectives for funding allocation (e.g., species protection and recreation opportunity, or species protection and ecosystem services).

The retrospective analysis presented here provides a systematic method for evaluating the cost-efficiency of conservation investments for the purposes of protecting unique species. This is particularly important in California, given the public source of funds and that state spending often leverages additional funding from other entities. While a general assumption by the conservation community is that spending will, among other things, protect California's unique species, this return is generally not presented in conservation spending reports. The approach we offer here can provide a means to quantify outcomes (e.g., number of species protected) rather than just inputs (e.g., dollars invested) (Ferraro & Pattanayak 2006). Our approach also provides a transparent and repeatable decision support tool at a scale appropriate for informing future conservation investments by any entity with a large geographic purview. In most situations where the allocation of funds depends on multiple criteria, and if the proportion designated to species protection can be determined, our approach can help ensure funds are spent strategically (Shaffer et al. 2002; Lerner et al. 2007) to achieve the greatest outcomes for species protection. And, in instances where criteria other than species protection determine spending allocation, additional optimization models could be developed and tested. This is critical not only for the utilitarian, recreational, spiritual, and ethical values that species protection provides today (Noss & Cooperrider 1994), but is also of utmost importance for securing the protection of that diversity in the face of future global change.

Editor : Stephen Polasky

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

This study was funded by The Nature Conservancy of California, the Rodney Johnson/Katherine Ordway Stewardship Endowment, and The Oracle Corporation. We thank the California Department of Fish and Game for data, and James Quinn for valuable discussions, and Peter Kareiva, Robin Cox, and Stacey Solie for reviewing an early draft of the article.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Ando, A., Camm J., Polasky S., Solow A. (1998) Species distributions, land values, and efficient conservation. Science 279, 21262128.
  • Balmford, A., Gaston K.J., Rodrigues A.S.L., James A. (2000) Integrating costs of conservation into international priority setting. Conserv Biol 14, 19.
  • CalFlora (2007) CalFlora database project. Available from: http://www.calflora.org/species/. Accessed July 2007.
  • CalFlora. (2007) CalFlora database project. Available from: http://www.calflora.org/species/. Accessed July 2007.
  • California Department of Fish and Game, (2005) California Interagency Wildlife Task Group. California Wildlife Habitat Relations version 8.1 personal computer program. Sacramento , CA .
  • California Department of Forestry and Fire Protection (2002) Multi-source Land Cover Data. 100m (2002 v2). Digital data.
  • California Resources Agency Legacy Project (2005) Public Conservation and Trust Lands. Updated by TNC March 2006. Digital data.
  • Costello, C., Polasky S. (2004) Dynamic reserve site selection. Resource and Energy Economics 26, 157174.
  • Dallman, P.R. (1998) Plant life in the world's mediterranean climates. University of California Press, Sacramento , USA .
  • Doak, D.F., Mills L.S. (1994) A useful role for theory in conservation. Ecology 75, 615526.
  • Ferraro, P.J., Pattanayak S. (2006) Money for nothing? A call for empirical evaluation of biodiversity conservation investments. PLoS Biol 4, e105.
  • Halpern, B.S., Pyke C.R., Fox H.E., Haney C.J., Schlaepfer M.A., Zaradic P. (2006) Gaps and mismatches between global conservation priorities and spending. Conserv Biol 20, 5664.
  • Kareiva, P., Marvier M. (2003) Conserving biodiversity coldspots: recent calls to direct conservation funding to the world's biodiversity hotspot may be bad investment advice. Am Sci 91, 344352.
  • Lerner, J., Mackey J., Casey F. (2007) What's in Noah's wallet? Land conservation spending in the United States. BioScience 57, 419423.
  • Murdoch, W., Polasky S., Wilson K.A., Possingham H.P., Kareiva P., Shaw M.R. (2007) Maximizing return on investment in conservation. Biol Conserv 139, 375388.
  • Naidoo, R., Balmford A., Ferraro P.J., Polasky S., Ricketts T.H., Rouget M. (2006) Integrating economic costs into conservation planning. Trends Ecol Evol 21, 681686.
  • Noss, R.F., Cooperrider A.Y. (1994) Saving nature's legacy. Island Press, Washington , D.C .
  • Odling-Smee, L. (2005) Dollars and sense. Nature 437, 614616.
  • Rosenzweig, M. (1995) Species diversity in space and time. Cambridge University Press, Cambridge , U.K .
  • Shaffer, M.L., Scott J.M., Casey F. (2002) Noah's options: initial cost estimates of a national system of habitat conservation areas in the United States. BioScience 52, 439443.
  • The Nature Conservancy (2005) TNC property title interests in California. Digital data.
  • Underwood, E.C., Shaw M.R., Wilson K.A. et al . (2008) Protecting biodiversity when money matters: maximizing return on investment. PLoS ONE 3(1):e1515.doi:DOI: 10.1371/journal.pone.0001515.
  • U.S. Census Bureau (2000) Census 2000 data PLoS ONE 3(1):e1515.doi:DOI: 10.1371/journal.pone.0001515. for the state of California. Available from: http://www.census.gov/census2000/states/ca.html. Accessed December 2007.
  • Wildlife Conservation Board (2007) Project Data database extract 1990 to 2006. Provided June 2007.
  • Wilson, K.A., McBride M.F., Bode M., Possingham H.P. (2006) Prioritising global conservation efforts. Nature 440, 337340.
  • Young, C.E.F., Roncisvalle C.A. (2002) Expenditures, investment and financing for sustainable development in Brazil. In U.N. Comisión Económica para América Latina, Santiago , Chile .

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
CONL_018_sm_SuppMat.doc121KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.