INTEGRATING PEOPLE AND FOR A SUSTAINABLE FUTURE

JOHN A. BISSONETTE

and

PAUL R. KRAUSMAN

The Wildlife Society Bethesda, Maryland

1995 436 International Wildlife Management Congress 10.6

10.6 In common usage, corridor has been defined as 1) “A gallery or passageway. . . one into which compartments or rooms open,” 2) "A gallery or TOWARDS A DEFINITION OF passageway connecting several apartments of a BIOLOGICAL CORRIDOR building,” 3) “. . . a narrow passageway or route” (Merriam Webster and Co. 1961) and as numerous DANIEL K. ROSENBERG similar definitions. The common elements of these Oregon Cooperative Wildlife Research Unit. Department definitions most relevant to their ecological appli- of Fisheries and Wildlife, Oregon State University, cation are the terms passageway and connecting. Corvallis. OR 97331, USA With “passageway” there is an implicit concept that the corridor is narrow relative to the BARRY R. NOON being interconnected. In the ecological literature. U.S.D.A. Forest Service, Redwood Sciences Laboratory, corridors have been defined as 1 of 3 major land- 1700 Bayview Drive, Arcata, CA 9552 1, USA scape elements: patch. matrix. and corridor (For- man and Godron 1986:23). In a thorough discus- E. CHARLES MESLOW sion of the principles of landscape ecology. U.S. Fish and Wildlife Service. Oregon Cooperative corridors were defined as “. . . narrow strips of land Wildlife Research Unit, Department of Fisheries and which differ from the matrix on either side. Corri- Wildlife. Oregon State University. Corvallis. OR dors may be isolated strips. but are usually attached 97331, USA to a patch of somewhat similar vegetations” (For- man and Godron 1986:123). This definition char- Abstract: Lack of clear, unambiguous criteria that acterizes corridors in terms of their shape and spa- distinguishes a linear patch as a corridor tial context. but does not explicitly ascribe a contributes to controversy over the value of corri- functional role. Earlier in their discussion. Forman dors for wildlife conservation. The definitions of and Godron (1986:121) emphasized the possible biological corridors have been vague or inconsis- transport function i.e.. movement of objects) of tent, and often they confound form and function. corridors. arising as a consequence of their shape Explicit criteria that can differentiate between a lin- and context. rather than as a necessary condition to ear habitat patch and a biological corridor have not ascribe the term “corridor“ to a linear element. been formulated. We reviewed the use of the term Given the above definitions. the necessary cri- “corridor-’ in the ecological literature. and attempt- teria for determining if a linear landscape element ed to clarify the concept of biological corridor. is a corridor are ambiguous. One definition empha- sizes function (passageway from one location to an- Resumen: La falta de claridez y criterios ambiguos other) while others stress form and context (narrow, para distinguir un parche de habitat linear de un and contrasting with the environment on its edges,. corredor contribuye a la controversia de1 valor de Thus. when issues such as the significance of cor- los corredores en la conversacion de de fauna sil- ridors to the maintenance of biological diversity are vestre. Las definiciones de corredores biol@cos debated (Noss 1987. Simberloff and Cox 1987. han sido vagas e inconsistentes y frecuentemente Saunders and Hobbs 1991a). disagreement may confunden la forma y funcion. Un criterio explicito arise simply as a consequence of divergent under- que pueda diferenciar entre un parche de habitat standings of the corridor concept. linear y un corredor biol6gico no han sido aun for- Our work on corridors has been supported by the mulados. Nosotros revisamos el uso de1 termino U. S. Forest Service. Redwood Sciences Labora- "corredor" en la literatura ecoltigica en un intento tory. Arcata, California. and by the Oregon Coop- de clarificar el concepto de corredor biol&ico. erative Wildlife Research Unit. which is operated with the cooperation of Oregon State University. Key words: biological corridor. connectivity, cor- Oregon Department of Fish and Wildlife. U.S. Fish ridor, habitat corridor. habitat patch. landscape. lin- and Wildlife Service, and the Wildlife Management ear patch. Institute. We thank S. DeStefano and L. D. Harris for helpful comments on improving this manu- script. The inclusion of linear landscape elements into wildlife conservation plans has gained wide accep- USE OF THE TERM “CORRIDOR” IN tance as an important aspect of conservation strat- PUBLISHED STUDIES egy (Noss 1987, Saunders and Hobbs 199la, Mann Corridors have been described as linear patches of and Plummer 1993). Arguments for the importance natural vegetation that provide habitat for wildlife, of corridors as landscape elements can be found in either as temporary use areas (part of a home range) natural resource ecology and human ecology (For- or as a place of permanent residence (an inclusive man and Godron 1986:121). The use of the term home range). For example. Maelfait and De Keer corridor in a broad spectrum of disciplines has con- (1990) in their study of field edges in Belgium, con- tributed to vague and sometimes contradictory def- cluded that corridors were effective in the conser- initions, and has incited vigorous debate over their vation of invertebrates. Their conclusion was based importance to conservation. on the observation that the “corridor” provided

. 10.6 Towa Definition of Corridor. Ro.renbe r!: et al. 437 habitat (both temporary and permanent) for many a corridor is a type of landscape element that es- species that were not adapted to the surrounding tablishes connectivity via ‘a continuous narrow pasture. They recognized the possible importance patch of vegetation that facilitates movement of the corridor for migration, yet their conclusion among larger habitat patches and prevents their iso- of corridor value was based exclusively on the role lation (Merriam 1984). Soule’ and Gilpin (l991:3) of providing, habitat. In a similar example, Van provide a clear and concise definition: “. . . a linear Dorp and Opdam (1987) assert the functional role two-dimensional landscape element that connects of corridors as connecting-networks, yet describe two or more patches of wildlife (animal) habitat corridors in terms of their habitat composition, not that have been connected in historical time; it is in terms of any effects on animal movement. meant as a conduit for animals." Bennet (1990: Linear landscape elements arising from human 109) defined habitat corridors as “. . . narrow con- design, such as power-lines and roadside vegetation necting strips of favored habitat.” Szacki (1987) are sometimes referred to as corridors with an im- limits the discussion of corridor effectiveness to the plicit assumption of ecological value. For example. frequency of movement. without considering its Kroodsma (1987) described densities and dis- vaIue as habitat. Dmowski and Kozakiewicz (1990) tribution in brushy power-line habitat and along the defined corridors similarly, and explicitly equated edge of a forest in Tennessee (U.S.). Although data corridors with connectivity, and discussed the role were not presented that demonstrated enhanced of a narrow belt of shrubs in enhancing movement movement of . the habitat was referred to as of birds between 2 (different) habitats and in di- “brush>, corridor vegetation’ (Kroodsma 1987: recting movement. Merriam and Lanoue (1990: 282) consistent with the habitat definition of For- 124) restricted their use of the term corridor by man and Godron (1986). Similarly. roadside vege- calling it “movement corridor.” thus implicitly sug- tation is often considered as a corridor. For exam- gesting other functions for other types of corridors. ple. several papers included in Saunders and Hobbs A functional definition was adopted by Reh and (1991a) on corridors discussed the advantages of Seitz (1990) in their discussion of corridors as con- managing roadside vegetation as habitat. nectors among otherwise isolated of There are numerous examples of “corridor” frogs. In this case. corridors were discussed as fa- used to signify its structural attributes as linear hab- cilitating and directing movement. itat and its functional role as a dispersal conduit. In This brief Iiterature review should reveal that reply to a paper questioning the merits of corridors corridors mean different things to different authors. (Simberloff and Cox 1987). Noss (1987) listed first and this ambiguity has contributed to the current those criteria associated with the enhanced move- controversy over their efficacy as conservation ment function. Secondarily. factors associated with tools. Without a clear definition of corridors. de- habitat attributes were described and the discussion fined in terms of their functional effects on animal of these factors imply the habitat criteria (i.e. form) behavior, it is impossible to determine their value for corridors: “Scenery, recreation, pollution abare- as management tools. ment. and land value enhancement are what usually motivate planners to draw corridors into their de- A MODEL TO CLARIFY THE MEANING signs" (Noss 1987:162). Although a focus on form OF BIOLOGICAL CORRIDORS does not preclude a functional detinition of corri- To distinguish between a linear landscape element dor. it suggests that either set of criteria. facilitated as habitat or as a biological corridor, we need to movement or spatial structure, context, and com- clarify the function of such patches for the species position is sufficient. In summarizing the role of that occupy them. Specifically, we will focus on corridors. Saunders and Hobbs 1991b). followed corridors as facilitators of movement between hab- the definition of Forman and Godron (1986). in- itat patches. To do so. we offer the foliowing op- cluding both the habitat (form) and movement erational definitions of 2 landscape elements and (function) role of linear patches; emphasis, how- note that each may need to be defined on a species- ever, was placed on facilitated movement. Merriam specific basis. (1991:137) stated that “Corridors may or may not Habitat: a patch that provides for survivorship, be involved in achieving connectivity among patch- natality, and movement. If average survivorship es or fragments”, thus emphasizing a habitat defi- and natality rates allow a stable or growing pop- nition that may include, but does not require, a ulation that produces emigrants, it is a source functional role of facilitating movement. Laan and patch; otherwise, it is a sink that is dependent Verboom (1990) are among the few researchers upon immigrants to sustain its populations (Pul- who recognized that the role of a strip of vegetation liam 1988). as habitat or as a facilitator of movement are not Corridor: a linear landscape element that pro- necessarily equivalent, and are difficult to differ- vides for survivorship and movement, but not entiate. Failure to reconcile these 2 definitions of necessarily natality, between other habitats. “corridor” have contributed to she controversy Thus, not a11 of a species life-history require- over their value. ments may be met in a corridor. The facilitated movement function of a linear landscape element is the most commonly assumed Given the above operational definition, a corridor distinguishing characteristic of a corridor. That is, can be characterized by ~3 key parameters: 438 International Wildlife Management Congress 10.6

edge-to-edge) is modeled by a declining exponen- tial, Pr (survive to time t) = exp[-k(d/V)], where k is a constant as defined above with units of l/time, d is the distance between the source and a sink patch, and v” is the animals average speed of movement between these 2 patches. II Under the simplest case of equally-spaced. equal ! r,-d,-fz-CJ size patches, and no selectivity among possible cor- I_I- dc - ridors, we have the following total likelihood of successful dispersal.

pr (disperse from Source to Sink Patch) = 2θ/360 Pr(dispersal success)

‘2 θ = arctan - Lf@7- c 2 Fig. 1. Method of estimating the angle of intersection between a source patch and an adjacent sink patch in the i = 1, 2, 3, 4th corridor, and H 5 30’. heuristic corridor model. The null probability of dispers- ing in the correct direction is a function of the combined 1, 2, 3, 4th corridor, and (5 5 70’. angles of intersection of all neighboring sink patches. If we further assume that: 1 ) only 1 path can be chosen. 2) the 4 possible dispersal paths differ in their likelihood of success. and 3) the animal is able Selectivity (s): the degree to which a dispersing to discriminate among these paths. the disperser animal can discriminate among possible corri- will maximize its probability of success by select- dors (pathways) between habitat patches so as ing the corridor (pathway) that minimizes ki/i. That to maximize its likelihood of successful dispers- is. for fixed d, survival probability is increased by al. selecting the corridor with the least resistence (sur- Resistence (k): a measure of the resistence. or vival cost per unit time) and the corridor where an survival costs, per unit of time spent in a given animal achieves the largest average velocity. Im- corridor. portantly, these factors can have compensatory ef- Velocity N): the average rate of an animal’s fects. We expect selectivity to evolve as natural se- movement in a corridor patch. lection should favor those dispersers that chose These parameters can be combined into a simple corridors with minimum X-,/r/, values. model. Consider a hypothetical landscape with 3 If we allow non-random dispersal from the elements: a circular source habitat patch. 4 sur- source patch and differential selection among pos- rounding circular sink-habitat patches (1 aligned in sible corridors, equation (3) is changed to each of the cardinal directions from the source Pr(dispersal success) patch). and a landscape matrix in which these hab- itat patches are embedded. Each straight-line path- way from the center of the source patch that inter- sects an adjacent sink patch represents a possible corridor (Fig. 1). A pool of emigrants is available where si = selectivity coefficient associated with in the source patch. and due to density-dependent corridor i, effects. these individuals are forced to disperse. Dispersal between source and sink patches is mod- Σ θ 0 < si < 180/q, and 45 si 5 180/ . eled as a straight-line path moving away from the i=1 source patch at a random azimuth. Assuming the disperser selects a pathway that in- There are 2requirements for successful dispersal. tersects an adjacent patch, si can be thought of as First. the dispersing animal must move in a direc- the probability (p ) of selecting corridor i, with tion that will intersect a neighboring sink patch. In i our model, the probability (Pr) of intersection with an adjacent sink patch is governed by the following equation (Fig. 1): In this case, equation (4) becomes Pr (disperse from source patch to a given adjacent sink patch) = 2θ/360, Pr(dispersa1 success) = S (pi)exp[ -k,W~;)]. θ 2 2 1/2 where = arctangent [r2/dc - r2 ) ]. i=1 Second, if a correct direction is chosen, the animal An individual animal maximizes its likelihood of must move successfully through the corridor. The successful dispersal by choosing, with probability likelihood of successful travel to an adjacent sink 1, the corridor with minimum k@,, and avoiding all patch lying a distance d units away (measured other possible corridors. Additional complexity (re- 10.6 Towards a Definition of Corridor l Rosenberg et al. 439 ality ?) can be added by allowing k or Q to show Dmowski, K., and M. Kozakiewicz. 1990. Influence of a positive, or negative, density-dependent effects. shrub corridor on movements of passerine birds to a Thus, immieration rate to a sink patch is a func- lake littoral zone. Landscape Ecol. 4:99-108. tion of the number of animals directed to a corridor Forman, R. T. T., and M. Godron. 1986. Landscape ecol- ogy. John Wiley & Sons, New York, N.Y. 619pp. and the number that can successfully traverse the Harris, L. D., and J. Scheck. 1991. From implications to chosen corridor. By these criteria, a linear land- applications: the dispersal corridor principle applied scape element functions as a corridor when the im- to the conservation of biological diversity. Pages migration rate to the target patch is increased over 189-220 in D. A. Saunders and R. J. Hobbs, eds. what it would be if the linear patch was not present. conservation 2: the role of corridors. Surrey Corridors thus exist along a continuum, defined in Beatty & Sons, Chipping Norton, Australia. terms of s, k, and F, and they may have equal ef- Kroodsma, R. L. 1987. Edge effect on breeding birds ficiencies by compensatory relationships among along power-Iine corridors in east Tennessee. Am. MidI. Nat. 118:275-283. these variables. Laan, R., and B. Verboom. 1990. Effects of pool size and isolation on communities. Biol. Conserv. 54:251-262. This preliminary conceptual model of a corridor Maelfait, J. P., and R. De Keer. 1990. The border zone emphasizes its functional role as a facilitator of of an intensively grazed pasture as a corridor for spi- movement. Further, we have partitioned the overall ders Araneae. Biol. Conserv. 54:223-238. functional response into its behavioral components Mann, C. C., and M. L. Plummer. 1993. The high cost to provide a focus for future research. Estimation of . Science 260:1868-1871. of the model parameters that determine the degree Merriam, G. 1984. Connectivity: A fundamental ecolog- to which a linear patch functions as a corridor will ical characteristic of landscape pattern. Pages 5-15 in be difficult. Consideration of these parameters, J. Brandt and P. Agger. eds. Proc. first int. seminar on however, will provide a theoretical construct for as- methodology in landscape ecolo. res. and planning. Int. Assoc. for Landscape Ecol., Roskilde Univ., Ros- sessing the degree to which a linear patch may kilde. Denmark. function primarily as a biological corridor. -. 1991. Corridors and connectivity: animal popu- Our definition is similar to that of Soule’ and lations in heterogeneous environments. Pages 133- GiIpin (1991). However, we do not require a 2- I42 in D. A. Saunders and R. J. Hobbs, eds. Nature dimensional limitation. nor do we require that conservation 2: the role of corridors. Surrey Beatty patches being linked to have been connected his- & Sons, Chipping Norton, Australia. torically. This later requirement may be important -. and A. Lanoue. 1990. Corridor use by small when considering the original conservation value of : field measurement for three experimental a specific landscape pattern, but the functional sig- types of Peromyscus leucopus. Landscape Ecol. 4: 123-131. nificance of an extant corridor is unreiated to his- Merriam Webster and Co. 1961. Webster’s third new in- torical uses. ternational dictionary. Unabridged. G. and C. Merri- The approach we use to clarify the corridor con- am Co. Springfield. Mass. 2,662pp. cept is similar to that of Merriam (1991) in that we Noss, R. F. 1987. Corridors in real landscapes: a reply to emphasize the functional aspect of corridors as me- Simberloff and Cox. Conserv. Biol. 1:159-l 64. diated through their effects on an animal's behavior. Pulliam, H. R. 1988. Sources. sinks, and reg- In contrast. we view natality as an infrequent be- ulation. Am. Nat. 132:652-661 havior in corridors. but a necessary condition for a Reh, W., and A. Seitz. 1990. The influence of land use habitat patch. on the genetic structure of populations of the common Our definition differs from Forman and Godron‘s frog Rana temporaria. Biol. Conserv. 54:239-250. (1986) by restricting our use of the term "corridor" Saunders, D. A., and R. J. Hobbs. eds. 1991a. Nature conservation 2: the role of corridors. Surrey Beatty to its function as a facilitator of movement and by & Sons. Chipping Norton, Australia. 442 pp. not specifying vegetation characteristics relative to -. and -. 1991b. The role of corridors in con- vegetation of other landscape elements. SimilarIy, servation: what do we know and where do we go? our definition differs from Harris and Scheck’s Pages 421-427 in D. A. Saunders and R. J. Hobbs. (1991) definition that requires the linear patch to be eds. Nature conservation 2: the role of corridors. Sur- comprised of native vegetation and to be similar to rey Beatty & Sons, Chipping Norton, Australia. the connected tracts (i.e., target patches). Our func- Simberloff, D., and J. Cox. 1987. Consequences and tional definition has no such restrictions, and, sim- costs of conservation corridors. Conserv. Biol. 1:63- ilar to Dmowski and Kozakiewicz (1990), the term 71. “corridor” does not necessarily specify what is be- Sou 12. M. E., and M. E. Gilpin. 1991. The theory of capability. Pages 3-8 in D. A. Saun- ing connected. Our definition simply requires that ders and R. J. Hobbs, eds. Nature conservation 2: the immigration to the target patch via the corridor be role of corridors. Surrey Beatty & Sons, Chipping greater than if the corridor were absent, and this is Norton, Australia. the key criterion asserted. Szacki, J. 1987. Ecological corridor as a factor determin- ing the structure and organization of a bank vole pop- LITERATURE CITED ulation. Acta Theriol. 32:31-44. Bennet, A. F. 1990. Habitat corridors and the conserva- Van Dorp, D.. and P. F. M. Opdam. 1987. Effects of patch tion of small mammals in a fragmented forest envi- size, isolation and regional abundance on forest bird ronment. Landscape Ecol. 4:109-l22. communities. Landscape Ecol. 1:59-73.