Advertisement

Biodiversity and Conservation

, Volume 20, Issue 8, pp 1611–1623 | Cite as

Habitat quality of source patches and connectivity in fragmented landscapes

  • Robert L. Schooley
  • Lyn C. Branch
Original Paper

Abstract

Because spatial connectivity is critical to dispersal success and persistence of species in highly fragmented landscapes, the way that we envision and measure connectivity is consequential for biodiversity conservation. Connectivity metrics used for predictive modeling of spatial turnover and patch occupancy for metapopulations, such as with Incidence Function Models (IFM), incorporate distances to and sizes of possible source populations. Here, our focus is on whether habitat quality of source patches also is considered in these connectivity metrics. We propose that effective areas (weighted by habitat quality) of source patches should be better surrogates for population size and dispersal potential compared to unadjusted patch areas. Our review of a representative sample of the literature revealed that only 12.5% of studies incorporated habitat quality of source patches into IFM-type connectivity metrics. Quality of source patches generally was not taken into account in studies even if habitat quality of focal patches was included in analyses. We provide an empirical example for a metapopulation of a rare wetland species, the round-tailed muskrat (Neofiber alleni), demonstrating that a connectivity metric based on effective areas of source patches better predicts patch colonization and occupancy than a metric that used simple patch areas. The ongoing integration of landscape ecology and metapopulation dynamics could be hastened by incorporating habitat quality of source patches into spatial connectivity metrics applied to species conservation in fragmented landscapes.

Keywords

Colonization Connectivity Fragmentation Habitat heterogeneity Metapopulation Patch quality Neofiber alleni 

Introduction

Habitat connectivity is crucial to dispersal success, persistence, and genetic diversity of species in fragmented landscapes. Exactly how we conceptualize and quantify connectivity is important for understanding spatial population processes and for conserving biodiversity. Patch connectivity, the inverse of patch isolation, should index the likelihood of immigration into a focal habitat patch (Tischendorf and Fahrig 2000; Moilanen and Hanski 2006). Metapopulation studies frequently use patch connectivity metrics to model recolonization of patches and demographic rescue of small populations (Moilanen and Hanski 2001; Moilanen and Nieminen 2002). In contrast, landscape connectivity applies to whole landscapes (Tischendorf and Fahrig 2000) and is the degree to which landscape structure promotes or impedes movements of individuals among resource patches (Taylor et al. 1993). Our emphasis is on patch connectivity as used for predictive modeling of metapopulations in highly fragmented environments. Another less subtle distinction is between structural connectivity and functional connectivity (Taylor et al. 1993; Tischendorf and Fahrig 2000; Baguette and Van Dyck 2007). Structural connectivity measures the spatial patterning of habitat based on mappable landscape features. Functional connectivity is a species-specific measure that considers how movement behavior and dispersal capacity of species interact with structural connectivity.

Previous assessments of spatial connectivity metrics have produced valuable insights. Moilanen and Nieminen (2002) demonstrated that the nearest neighbor measure (i.e., distance from focal patch to nearest source patch) can be inferior to buffer metrics and Incidence Function Model (IFM) metrics for predicting colonization (see also Moilanen and Hanski 2006). Buffer metrics sum all occupied habitat areas within a specified distance of a focal patch to index potential sources of dispersers. However, buffer measures are sensitive to the selected buffer radius and do not consider explicit spatial locations of source patches within the buffer (Moilanen and Nieminen 2002). IFM connectivity metrics perform well by incorporating the distances to and sizes of potential source populations. Patch areas of source populations are used as a surrogate for local population sizes. Simulation studies also support the superiority of area-informed connectivity metrics relative to distance-only measures (Bender et al. 2003; Tischendorf et al. 2003). In contrast, Prugh (2009) recently evaluated connectivity measures for 24 metapopulations and concluded that nearest neighbor measures can perform as well as IFM metrics if distance to the nearest occupied patch is used as a source instead of distance to any nearest patch. Winfree et al. (2005) determined that none of the simple connectivity metrics were good predictors of colonization for several empirical systems and suggested that more detailed biological information should be included in metrics. Relative performance of connectivity metrics also can depend on spatial scaling of the system (i.e., dispersal ability of species relative to interpatch distances; Matter et al. 2005). Finally, existing connectivity metrics can be ordered so as to display a progression from structural to functional measures and to represent a tradeoff between information content and data requirements (Calabrese and Fagan 2004; Fagan and Calabrese 2006).

Our focus is on habitat quality of suitable patches and its effect on patch connectivity. Researchers now recognize that quality of focal patches can affect local extinctions, colonizations, and occupancy (Thomas 1994; Hokit et al. 1999; Fleishman et al. 2002; Franken and Hik 2004; Armstrong 2005; Thornton et al. 2011). Patch areas can be weighted by habitat quality to create a more ecologically relevant measure termed ‘effective patch areas’ (Hanski 1994; Hokit et al. 2001). However, we posit that habitat quality of source patches typically is ignored when patch connectivity metrics are constructed. For instance, IFM metrics use patch areas of source patches and not their effective patch areas. Habitat-weighted areas often should be better surrogates for population sizes and thus better predictors of dispersal potential.

Our objectives were twofold. First, we reviewed relevant literature to determine how often habitat quality of source patches was incorporated into patch connectivity metrics applied to fragmented landscapes. Second, we analyzed empirical data for a mammal metapopulation and directly compared the abilities of IFM connectivity metrics that either used raw areas or effective areas of source patches to predict colonization and patch occupancy. Previous reviews have discussed possible extensions of current connectivity metrics but have not underscored the importance of habitat quality of source patches (Moilanen and Nieminen 2002; Bender et al. 2003; Calabrese and Fagan 2004; Winfree et al. 2005; Fagan and Calabrese 2006; Moilanen and Hanski 2006; Kindlmann and Burel 2008; Kadoya 2009; Prugh 2009; but see Visconti and Elkin 2009).

Mini-review

Our goal was to survey a representative sample of literature in spatial ecology in which IFM-type connectivity metrics were applied. We excluded from our review any studies that used a nearest-neighbor measure or a habitat buffer because our experience was that habitat quality of sources was rarely included in these simpler connectivity metrics. We were particularly interested in whether there were changes over time in how researchers dealt with habitat quality of patches when measuring connectivity. Therefore, we first revisited the literature survey conducted by Moilanen and Nieminen (2002) in which they reviewed all 1998–2000 issues of six representative journals in ecology and conservation: American Naturalist, Conservation Biology, Ecology, Journal of Animal Ecology, Landscape Ecology, and Oikos. Moilanen and Nieminen (2002) excluded studies of population genetic structure. We inspected 20 papers from Moilanen and Nieminen’s (2002) review in which researchers had used some type of IFM connectivity metric. We removed five papers from our survey in which connectivity was measured using only summed distances to source patches but not patch areas (e.g., Nekola 1999), or by weighting source patches with direct estimates of population sizes instead of with indirect estimates from patch areas (e.g., Thomas and Kunin 1999). Hence, we reviewed 15 papers from 1998 to 2000 (Appendix).

Next, we repeated the literature survey procedures of Moilanen and Nieminen (2002) for the same six journals for all 2005–2008 issues. We included papers that used some form of IFM metric. As before, we did not review studies in which the response variable was genetic similarity. We reviewed 25 papers from 2005 to 2008 (Appendix). A quantitative meta-analysis (e.g., Prugh et al. 2008) was not possible because information needed to estimate effect sizes—relative performances of connectivity metrics with and without quality of source patches—was missing from nearly all studies (see also Mortelliti et al. 2010). Moreover, a meta-analysis was not necessary to make our main points (see “Results” section).

For the total of 40 papers in our mini-review, we asked two main questions. (1) Was habitat quality of source patches included in connectivity metrics through the use of effective patch areas or some similar weighting? (2) Was habitat quality of focal patches considered in studies? An affirmative answer to the second question would indicate researchers were aware of the potential importance of habitat quality of suitable patches for their system. We classified papers as belonging to the early period (1998–2000) or the late period (2005–2008). We also classified papers based on whether the response variable was typical of metapopulation studies (occupancy, extinction, colonization, metapopulation persistence) or some other response (see Appendix). We used logistic regression (Proc Genmod, SAS 2002) to examine whether the proportion of studies that included habitat quality of source patches depended on time period and the type of response variable. We conducted a separate analysis with logistic regression to ask if the likelihood that a study included habitat quality of focal patches depended on time period and the response variable. We used an Akaike Information Criterion (AICc) to select the best model for each response variable and to rank the others (Burnham and Anderson 2002). We present results as AIC differences (Δi = AICi − minimum AIC) so that the best model has Δi = 0. We also report Akaike weights (wi) that are normalized relative likelihoods that model i is the best model.

Effect of matrix heterogeneity on connectivity was not the focus of our review. Nevertheless, we kept track of whether investigators used Euclidean distances or effective distances modified by the matrix (Ricketts 2001; Chardon et al. 2003; Verbeylen et al. 2003) to measure interpatch distances as an additional gauge of whether habitat heterogeneity was being incorporated into spatial ecology.

Mini-review results

Overall, only 5 of 40 (12.5%) studies incorporated habitat quality of source patches into connectivity metrics using effective patch areas or a related approach (Appendix). There was little evidence that time period or response type had an effect on whether studies included source patch quality (Table 1). The top-ranked model was the intercept-only model. The only other competitive model (Δi ≤ 2) included time period as a predictor, but this model did not alter the log-likelihood much compared to the intercept-only model (Table 1).
Table 1

Ranking of logistic regression models to evaluate how habitat quality of patches was treated in published papers in spatial ecology (see Appendix)

Response variable and modela

K

Log-likelihood

AICc

Δi

wi

Source patch

 Intercept-only

1

−15.07

0.00

0.50

 Time

2

−14.67

1.41

0.25

 Response

2

−15.03

2.14

0.17

 Time, response

3

−14.56

3.54

0.08

Focal patch

 Intercept-only

1

−27.68

0.00

0.37

 Time

2

−26.70

0.26

0.32

 Response

2

−27.53

1.59

0.16

 Time, response

3

−26.31

1.83

0.15

Source patch refers to whether habitat quality of sources was included in connectivity metrics. Focal patch refers to whether habitat quality of target patches was considered. K =  no. estimable parameters, Δi = AICci − minimum AICc, and wi are Akaike weights

aPredictor variables include time period of study that was either 1998–2000 or 2005–2008 (time), and whether the response variable was typical of metapopulation studies or not (response)

Nineteen of 40 (47.5%) studies included habitat quality of focal patches in their analyses (Appendix). We detected no strong effects of time period or response type on whether studies included focal patch quality (Table 1). The top-ranked model again was the intercept-only model. The Δi values and log-likelihoods, however, suggested a weak effect of time period. There was an increase over time in the likelihood that studies considered focal patch quality (1998–2000 = 33.3%, 2005–2008 = 56.0%). Overall, in 14 of 19 (73.7%) studies in which habitat quality of focal patches was considered, habitat quality of source patches was ignored when measuring connectivity (Appendix). We are responsible for such an inconsistent treatment of patch quality in some of our own research (Hokit et al. 1999; Schooley and Wiens 2005; but see Hokit et al. 2001; Schooley and Branch 2007, 2009).

Only 3 of 40 (7.5%) reviewed studies used effective distances modified by intervening matrix type or topography to measure inter-patch distances (Moilanen and Hanski 1998; Schooley and Wiens 2005; Snäll et al. 2008). Euclidean distances remained the norm for measuring patch isolation.

Empirical example: source quality and connectivity for muskrats

To determine whether including habitat quality of source patches improved predictions of patch occupancy and colonization, we investigated spatial dynamics of round-tailed muskrats (Neofiber alleni) at Avon Park Air Force Range in central Florida (see Schooley and Branch 2009). Suitable habitat for muskrats covered <5% of the landscape and consisted of small (median = 0.9 ha), geographically isolated depression marshes. The round-tailed muskrat is a rare, nocturnal, semi-aquatic herbivore with a restricted geographic distribution (Florida and southern Georgia) that has been proposed as a Species of Special Concern due to extensive loss of wetland habitat (Lefebvre and Tilmant 1992). Muskrats displayed metapopulation dynamics at Avon Park in which patch occupancy was stable between years, despite substantial local turnover, because extinctions were balanced by recolonizations (Schooley and Branch 2009).

In central Florida, round-tailed muskrats typically occur within zones of maidencane grass (Panicum hemitomon) in wetlands, often use maidencane to build lodges, and maidencane is their main diet item (Birkenholz 1963; Schooley and Branch 2009). Hence, we developed a maidencane index as a measure of patch quality (Schooley and Branch 2009). For each suitable wetland, we recorded cover and density of maidencane grass and then calculated an index that ranged from 0 (no maidencane) to 1 (pure maidencane with an estimated cover of 81–100%). The mean maidencane index for wetlands was 0.18 (SD = 0.16, min. = 0, max. = 0.88). Our study area included a strong gradient in habitat quality of patches (Schooley and Branch 2007; Mortelliti et al. 2010).

We surveyed 457 wetlands for presence-absence of muskrats during fall–winter in each of 2 years (2002–2003, 2003–2004). Patch occupancy was determined reliably based on presence of active lodges constructed by muskrats. False absences were unlikely (Schooley and Branch 2005) so use of occupancy modeling techniques was unnecessary (MacKenzie et al. 2003). Instead, we used ordinal logistic regression (Proc Genmod, SAS 2002) to model patch occupancy in which the response variable was the number of years a wetland was occupied (0, 1, 2). We used binary logistic regression to model colonization probability. Local extinctions were not strongly related to connectivity (Schooley and Branch 2009), so we do not evaluate extinction models here.

We focused on two key predictors for metapopulations—patch area and connectivity—in our logistic regression models. We included either patch area or effective patch area of the focal patch. Effective area (EA i ) for patch i was calculated as Q i A i /Q*, where Q i was the habitat quality (maidencane index) of patch i and Q* was the maximal habitat quality of any patch (Hanski 1994; Hokit et al. 2001). Our spatial connectivity metrics either used patch areas of source patches (Connectivity_A) or effective patch areas of source patches that incorporated habitat quality (Connectivity_EA). Connectivity_A (Ca i ) of patch i was defined as
$$ {{Ca}}_{{i}} = \sum\limits_{{{{j}} \ne {{i}}}} {\mathop {{p}}\nolimits_{{j}} \exp ( - \alpha {{d}}_{{ij}} } ){{A}}_{{j}}^{{b}} $$
(1)
where p j is the likelihood that patch j was occupied the previous year, α is a parameter that scales the effect of distance on migration, d ij is the distance (km) between centers of patches i and j (for d ij  ≤ 2 km), A j is the area of patch j, and b is a scaling parameter relating abundance to area. We set p j equal to 0 for patches unoccupied in both years, 0.5 for patches occupied in 1 year, and 1 for patches occupied in both years. We estimated α as 0.5 by using logistic regression to model colonization of patches in 2003–2004 as a function of exp(−αd), where d was distance to nearest potential source patch (Hokit et al. 2001), and selecting the value of α that minimized the deviance of model fit. Emigration is unlikely to scale linearly with patch area, so we assumed a value of 0.5 for b, which is a more reasonable expectation (Moilanen and Nieminen 2002). Model performance is relatively insensitive to changes in b (Moilanen and Nieminen 2002; Prugh 2009). We calculated connectivity using only source patches within 2 km because our study area was not an isolated patch network and a minor edge adjustment was needed (Schooley and Branch 2009). We used a buffer radius of 2 km because only 2% of vacant wetlands located >2 km from a source patch were colonized between years (Schooley and Branch 2007).

We calculated Connectivity_EA (Cea i ) for patch i where all variables are as defined in Eq. 1 except that we replaced area of source patches (A j ) with effective area of source patches (EA j ). As before, effective area (EA j ) for patch j was calculated as Q j A j /Q*, where Q j was the habitat quality of patch j and Q* was the maximal patch quality.

We used AICc to rank the logistic regression models for occupancy and colonization (Burnham and Anderson 2002). For each response variable, we evaluated eight models with covariates (Area, Effective Area, Connectivity_A, Connectivity_EA, Area + Connectivity_A, Area + Connectivity_EA, Effective Area + Connectivity_A, Effective Area + Connectivity_EA) plus the intercept-only model.

Results

In each year, 26% of suitable patches (n = 457) were occupied. The only competitive model for predicting occupancy (Table 2) included the effective area of the focal patch (β = 1.624, SE = 0.270) and Connectivity_EA (β = 0.185, SE = 0.027). The spatial connectivity metric that included habitat quality of source patches was superior to the metric that ignored habitat quality. The best patch occupancy model explained a reasonable amount of variation (R 2 = 0.34). All models with a single covariate (not shown in Table 2) had no support (Δi ≥ 48.27, wi = 0.00).
Table 2

Ranking of logistic regression models for predicting patch occupancy and colonization of wetlands for a metapopulation of round-tailed muskrats

Response variable and modela

K

Log-likelihood

AICc

Δi

wi

Occupancy

 Effective area, Connectivity_EA

4

−332.53

0.00

0.98

 Effective area, Connectivity_A

4

−336.26

7.45

0.02

 Area, Connectivity_EA

4

−349.54

34.00

0.00

 Area, Connectivity_A

4

−353.49

41.90

0.00

 Intercept-only

2

−409.60

150.07

0.00

Colonization

 Effective area, Connectivity_EA

3

−110.18

0.00

0.84

 Effective area, Connectivity_A

3

−111.86

3.36

0.16

 Area, Connectivity_EA

3

−116.33

12.30

0.00

 Area, Connectivity_A

3

−118.47

16.59

0.00

 Intercept-only

1

−134.75

45.09

0.00

K = no. estimable parameters, Δi = AICci − minimum AICc, and wi are Akaike weights. The top four models plus the intercept-only model are shown

aPredictor variables include patch area of focal patch (Area), effective patch area of focal patch (Effective area), spatial connectivity measured using patch areas of source patches (Connectivity_A), and spatial connectivity measured using effective patch areas of source patches (Connectivity_EA). Effective patch areas are weighted by habitat quality

Of the 340 wetlands that were vacant in 2002–2003, 46 (13.5%) were colonized by 2003–2004. Again, the only competitive model for predicting patch colonization (Table 2) included effective area of the focal patch (β = 2.206, SE = 0.499) and Connectivity_EA (β = 0.188, SE = 0.042). Larger and more connected patches were more likely to be colonized between years (Fig. 1; R 2 = 0.25). All models with a single covariate had no support (Δi ≥ 18.75, wi = 0.00).
Fig. 1

Combined influences of effective patch area and spatial connectivity on probability of vacant wetlands being colonized by round-tailed muskrats. Effective patch area (modified by habitat quality) of focal patches is plotted on a natural log scale. Connectivity_EA is a metric that includes effective patch areas of source patches. Colonization probabilities are predicted values from a logistic regression model [logit (Colonization) = −3.38 + 2.206 (effective area) + 0.188 (Connectivity_EA)]

Discussion

Our literature review of spatial ecology studies revealed that habitat quality of source patches generally was not included in connectivity metrics even when variation in quality of focal patches was recognized. This omission occurred even though effective patch areas of sources should be better predictors of patch colonization, occupancy, and abundance for most systems. Recently, Visconti and Elkin (2009) demonstrated via simulation modeling that connectivity metrics that included patch quality performed better in all scenarios, especially in environments with high variance in patch quality and spatial clustering of good quality patches. Our empirical example of round-tailed muskrats demonstrated the superiority of a habitat-informed connectivity metric for a metapopulation that includes substantial variation in patch quality and positive spatial autocorrelation of patch quality (Schooley and Branch 2007).

Habitat quality of source patches can be included in IFM connectivity metrics merely by replacing patch areas with effective patch areas (Hanski 1994) as in our empirical example. Alternatively, Jaquiéry et al. (2008) included habitat quality of source patches instead of their areas, but this substitution only marginally improved explanatory power of occupancy and did not account for variance in colonization. For graph-theoretic approaches to connectivity, which might provide an ideal balance between information content and data requirements (Calabrese and Fagan 2004), nodes can be habitat patches described by their areas or by some index of habitat quality or productivity (Urban and Keitt 2001; Visconti and Elkin 2009). However, inclusion of effective areas when applying graph theory to conservation is not the norm. Buffer connectivity metrics can be reasonable predictors of colonization if buffer radii reflect dispersal abilities of focal species (Moilanen and Nieminen 2002). Habitat quality can be considered when measuring isolation with buffers (e.g., González-Varo et al. 2008), but this practice is rare. Finally, connectivity models derived from circuit theory could be extended by giving greater weight to high-quality source patches that release more current (McRae et al. 2008).

The simplest way to convert patch areas to effective patch areas is to use the relative weighting method suggested by Hanski (1994) and applied to our muskrat metapopulation. This approach is perhaps best suited to systems in which one, or a few, environmental variables can be identified as important drivers of habitat quality. For instance, Hokit et al. (2001) calculated effective patch areas for the Florida scrub lizard (Sceloporus woodi) using percentages of open, sandy habitat within patches. Likewise, Moilanen and Hanski (2006) created effective patch areas by summing coverages of two host-plant species within patches for the Glanville fritillary butterfly (Melitaea cinxia), and Rabasa et al. (2007) used modified patch areas that accounted for fruit production by the host plant for the butterfly Iolana iolas. More complicated functions of environmental variables could be used to index patch quality (Moilanen and Hanski 1998) as could measures from ecological-niche factor analysis (Jaquiéry et al. 2008) or occupancy modeling that accounts for imperfect detection (Cosentino et al. 2010).

Conservation of species in fragmented landscapes often is guided by metapopulation theory (Hanski and Gaggiotti 2004). A long-awaited fusion of metapopulation biology and landscape ecology has begun (Hanski and Gaggiotti 2004; With 2004; Armstrong 2005). Our literature review indicated researchers integrate habitat quality of focal patches into simple patch-matrix perspectives fairly often. However, habitat quality of source patches generally is disregarded when quantifying connectivity. Moreover, effects of the matrix on dispersal are ignored in favor of using straight-line distances to measure patch isolation. The unresolved question centers on how much we can simplify landscapes (Pellet et al. 2007; Prugh et al. 2008), and dispersal behavior (Baguette and Van Dyck 2007), and still retain predictive measures of spatial connectivity (Calabrese and Fagan 2004; Winfree et al. 2005). In our view, overlaying habitat quality of suitable patches onto binary landscapes is a complexity that will improve predictions and conservation efforts for many metapopulations. Nevertheless, a patch-matrix approach remains a practical starting point for evaluating connectivity if habitat quality is not well understood for a species, obtaining data on habitat quality is too costly, or patch area is correlated with resource quantity and population size (Rabasa et al. 2007).

Notes

Acknowledgments

We are grateful to S. Cardiff, J. Christopoulos, R. Gilbreath, M. McDermott, A. Pries, L. Showen, M. Shumar, and C. Wolf for assistance with fieldwork. We also thank J. Bridges, P. Ebersbach, P. Margosian, S. Orzell, S. Penfield, and P. Walsh for facilitating our study at Avon Park Air Force Range. Our research on Neofiber was funded by a grant from the US Department of Defense.

References

  1. Armstrong DP (2005) Integrating the metapopulation and habitat paradigms for understanding broad-scale declines of species. Conserv Biol 19:1402–1410CrossRefGoogle Scholar
  2. Baguette M, Van Dyck H (2007) Landscape connectivity and animal behavior: functional grain as a key determinant for dispersal. Landsc Ecol 22:1117–1129CrossRefGoogle Scholar
  3. Bastin L, Thomas CD (1999) The distribution of plant species in urban vegetation fragments. Landsc Ecol 14:493–507CrossRefGoogle Scholar
  4. Bender DJ, Tischendorf L, Fahrig L (2003) Using patch isolation metrics to predict animal movement in binary landscapes. Landsc Ecol 18:17–39CrossRefGoogle Scholar
  5. Birkenholz DE (1963) A study of the life history and ecology of the round-tailed muskrat (Neofiber alleni True) in north-central Florida. Ecol Monogr 33:255–280CrossRefGoogle Scholar
  6. Brooks CP, Antanovics J, Keitt TH (2008) Spatial and temporal heterogeneity explain disease dynamics in a spatially explicit network model. Am Nat 172:149–159PubMedCrossRefGoogle Scholar
  7. Burnham KP, Anderson DR (2002) Model selection and multimodel inference: a practical information-theoretic approach. Springer-Verlag, New YorkGoogle Scholar
  8. Calabrese JM, Fagan WF (2004) A comparison-shopper’s guide to connectivity metrics. Front Ecol Environ 10:529–536CrossRefGoogle Scholar
  9. Chardon JP, Adriaensen F, Matthysen E (2003) Incorporating landscape elements into a connectivity measure: a case study for the Speckled wood butterfly (Pararge aegeria L.). Landsc Ecol 18:561–573CrossRefGoogle Scholar
  10. Cosentino BJ, Schooley RL, Phillips CA (2010) Wetland hydrology, area, and isolation influence occupancy and spatial turnover of the painted turtle, Chrysemys picta. Landsc Ecol 25:1589–1600CrossRefGoogle Scholar
  11. Doak P (2000) Habitat patchiness and the distribution, abundance, and population dynamics of an insect herbivore. Ecology 81:1842–1857CrossRefGoogle Scholar
  12. Ewers RM, Thorpe S, Didham RK (2007) Synergistic interactions between edge and area effects in a heavily fragmented landscape. Ecology 88:96–106PubMedCrossRefGoogle Scholar
  13. Facon B, David P (2006) Metapopulation dynamics and biological invasions: a spatially explicit model applied to a freshwater snail. Am Nat 168:769–783PubMedCrossRefGoogle Scholar
  14. Fagan WF, Calabrese JM (2006) Quantifying connectivity: balancing metric performance with data requirements. In: Crooks KR, Sanjayan M (eds) Connectivity conservation. Cambridge University Press, Cambridge, pp 297–317CrossRefGoogle Scholar
  15. Fleishman E, Ray C, Sjögren-Gulve P, Boggs CL, Murphy DD (2002) Assessing the roles of patch quality, area, and isolation in predicting metapopulation dynamics. Conserv Biol 16:706–716CrossRefGoogle Scholar
  16. Foppen RPB, Chardon JP, Liefveld W (2000) Understanding the role of sink patches in source-sink metapopulations: reed warbler in an agricultural landscape. Conserv Biol 14:1881–1892CrossRefGoogle Scholar
  17. Franken RJ, Hik DS (2004) Influence of habitat quality, patch size and connectivity on colonization and extinction dynamics of collared pikas Ochotona collaris. J Anim Ecol 73:889–896CrossRefGoogle Scholar
  18. González-Varo JP, López-Bao JV, Guitián J (2008) Presence and abundance of the Eurasian nuthatch Sitta europaea in relation to size, isolation and the intensity of management of chestnut woodlands in the NW Iberian Peninsula. Landsc Ecol 23:78–89CrossRefGoogle Scholar
  19. Hanski I (1994) A practical model of metapopulation dynamics. J Anim Ecol 63:151–162CrossRefGoogle Scholar
  20. Hanski I (1998) Connecting the parameters of local extinction and metapopulation dynamics. Oikos 83:390–396CrossRefGoogle Scholar
  21. Hanski I (1999) Habitat connectivity, habitat continuity, and metapopulations in dynamic landscapes. Oikos 87:209–219CrossRefGoogle Scholar
  22. Hanski I, Gaggiotti OE (2004) Metapopulation biology: past, present, and future. In: Hanski I, Gaggiotti OE (eds) Ecology, genetics, and evolution of metapopulations. Elsevier Academic Press, Amsterdam, pp 3–22CrossRefGoogle Scholar
  23. Hanski I, Alho J, Moilanen A (2000) Estimating the parameters of migration and survival for individuals in metapopulations. Ecology 81:239–251CrossRefGoogle Scholar
  24. Hokit DG, Stith BM, Branch LC (1999) Effects of landscape structure in Florida scrub: a population perspective. Ecol Appl 9:124–134CrossRefGoogle Scholar
  25. Hokit DG, Stith BM, Branch LC (2001) Comparison of two types of metapopulation models in real and artificial landscapes. Conserv Biol 15:1102–1113CrossRefGoogle Scholar
  26. Jaquiéry J, Guélat J, Broquet T, Berset-Brändli L, Pellegrini E, Moresi R, Hirzel AH, Perrin N (2008) Habitat-quality effects on metapopulation dynamics in greater white-toothed shrews, Crocidura russula. Ecology 89:2777–2785PubMedCrossRefGoogle Scholar
  27. Johnson DM (2005) Metapopulation models: an empirical test of model assumptions and evaluation methods. Ecology 86:3088–3098CrossRefGoogle Scholar
  28. Johnson DM, Horvitz CC (2005) Estimating postnatal dispersal: tracking the unseen dispersers. Ecology 86:1185–1190CrossRefGoogle Scholar
  29. Kadoya T (2009) Assessing functional connectivity using empirical data. Popul Ecol 51:5–15CrossRefGoogle Scholar
  30. Kindlmann P, Burel F (2008) Connectivity measures: a review. Landsc Ecol 23:879–890Google Scholar
  31. Kuussaari MI, Saccheri I, Camara M, Hanski I (1998) Allee effect and population dynamics in the Glanville fritillary butterfly. Oikos 82:384–392CrossRefGoogle Scholar
  32. Laaksonen M, Peuhu E, Várkonyi G, Siitonen J (2008) Effects of habitat quality and landscape structure on saproxylic species dwelling in boreal spruce-swamp forests. Oikos 117:1098–1110CrossRefGoogle Scholar
  33. Lefebvre LW, Tilmant JT (1992) Round-tailed muskrat (Neofiber alleni). In: Humphrey SR (ed) Rare and endangered biota of Florida. Volume I. Mammals. University Press of Florida, Gainesville, pp 276–286Google Scholar
  34. Lei G, Hanski I (1998) Spatial dynamics of two competing specialist parasitoids in a host metapopulation. J Anim Ecol 67:422–433CrossRefGoogle Scholar
  35. Lindenmayer DB, McCarthy MA, Pope ML (1999) Arboreal marsupial incidence in eucalypt patches in southeastern Australia: a test of Hanski’s incidence function metapopulation model for patch occupancy. Oikos 84:99–109CrossRefGoogle Scholar
  36. MacKenzie DI, Nichols LD, Hines JE, Knutson MG, Franklin AB (2003) Estimating site occupancy, colonization, and local extinction when a species is detected imperfectly. Ecology 84:2200–2207CrossRefGoogle Scholar
  37. Matter SF, Roslin T, Roland J (2005) Predicting immigration of two species in contrasting landscapes: effects of scale, patch size, and isolation. Oikos 111:359–367CrossRefGoogle Scholar
  38. McRae BH, Dickson BG, Keitt TH, Shah VB (2008) Using circuit theory to model connectivity in ecology, evolution, and conservation. Ecology 89:2712–2724PubMedCrossRefGoogle Scholar
  39. Menéndez R, Thomas CD (2000) Metapopulation structure depends on spatial scale in the host-specific moth Wheeleria spilodactylus (Lepidoptera: Pterophoridae). J Anim Ecol 69:935–951CrossRefGoogle Scholar
  40. Moilanen A (1999) Patch occupancy models of metapopulation dynamics: efficient parameter estimation using implicit statistical inference. Ecology 80:1031–1043CrossRefGoogle Scholar
  41. Moilanen A (2000) The equilibrium assumption in estimating the parameters of metapopulation models. J Anim Ecol 69:143–153CrossRefGoogle Scholar
  42. Moilanen A, Hanski I (1998) Metapopulation dynamics: effects of habitat patch area and isolation, habitat quality and landscape structure. Ecology 79:2503–2515CrossRefGoogle Scholar
  43. Moilanen A, Hanski I (2001) On the use of connectivity measures in spatial ecology. Oikos 95:147–151CrossRefGoogle Scholar
  44. Moilanen A, Hanski I (2006) Connectivity and metapopulation dynamics in highly fragmented landscapes. In: Crooks KR, Sanjayan M (eds) Connectivity conservation. Cambridge University Press, Cambridge, pp 44–71CrossRefGoogle Scholar
  45. Moilanen A, Nieminen M (2002) Simple connectivity measures in spatial ecology. Ecology 83:1131–1145CrossRefGoogle Scholar
  46. Moilanen A, Smith AT, Hanski I (1998) Long-term dynamics in a metapopulation of the American pika. Am Nat 152:530–542PubMedCrossRefGoogle Scholar
  47. Mortelliti A, Boitani L (2008) Interaction of food resources and landscape structure in determining the probability of patch use by carnivores in fragmented landscapes. Landsc Ecol 23:285–298CrossRefGoogle Scholar
  48. Mortelliti A, Amori G, Boitani L (2010) The role of habitat quality in fragmented landscapes: a conceptual overview and prospectus for future research. Oecologia 163:535–547PubMedCrossRefGoogle Scholar
  49. Nekola JC (1999) Paleorefugia and neorefugia: the influence of colonization history on community pattern and process. Ecology 80:2459–2473CrossRefGoogle Scholar
  50. Pellet J, Fleishman E, Dobkin DS, Gander A, Murphy DD (2007) An empirical evaluation of the area and isolation paradigm of metapopulation dynamics. Biol Conserv 136:483–495CrossRefGoogle Scholar
  51. Price SJ, Marks DR, Howe RW, Hanowski JM, Niemi GJ (2005) The importance of spatial scale for conservation and assessment of anuran populations in coastal wetlands of the western Great Lakes, USA. Landsc Ecol 20:441–454CrossRefGoogle Scholar
  52. Prugh LR (2009) An evaluation of patch connectivity measures. Ecol Appl 19:1300–1310PubMedCrossRefGoogle Scholar
  53. Prugh LR, Hodges KE, Sinclair ARE, Brashares JS (2008) Effect of habitat area and isolation on fragmented animal populations. Proc Natl Acad Sci USA 105:20770–20775PubMedCrossRefGoogle Scholar
  54. Rabasa SG, Gutiérrez D, Escudero A (2007) Metapopulation structure and habitat quality in modeling dispersal in the butterfly Iolana iolas. Oikos 116:793–806CrossRefGoogle Scholar
  55. Ranius T, Kindvall O (2006) Extinction risk of wood-living model species in forest landscapes as related to forest history and conservation strategy. Landsc Ecol 21:687–698CrossRefGoogle Scholar
  56. Renfrew RB, Ribic CA (2008) Multi-scale models of grassland passerine abundance in a fragmented system in Wisconsin. Landsc Ecol 23:181–193CrossRefGoogle Scholar
  57. Ricketts TH (2001) The matrix matters: effective isolation in fragmented landscapes. Am Nat 158:87–99PubMedCrossRefGoogle Scholar
  58. SAS (2002) SAS for Windows, Version 9.0. SAS Institute Inc, CaryGoogle Scholar
  59. Schooley RL, Branch LC (2005) Survey techniques for determining occupancy of isolated wetlands by round-tailed muskrats. Southeast Nat 4:745–756CrossRefGoogle Scholar
  60. Schooley RL, Branch LC (2007) Spatial heterogeneity in habitat quality and cross-scale interactions in metapopulations. Ecosystems 10:846–853CrossRefGoogle Scholar
  61. Schooley RL, Branch LC (2009) Enhancing the area-isolation paradigm: habitat heterogeneity and metapopulation dynamics of a rare wetland mammal. Ecol Appl 19:1708–1722PubMedCrossRefGoogle Scholar
  62. Schooley RL, Wiens JA (2005) Spatial ecology of cactus bugs: area constraints and patch connectivity. Ecology 86:1627–1639CrossRefGoogle Scholar
  63. Schtickzelle N, Wallis de Vries MF, Baguette M (2005) Using surrogate data in population viability analysis: the case of the critically endangered cranberry fritillary butterfly. Oikos 109:89–100CrossRefGoogle Scholar
  64. Schtickzelle N, Mennechez G, Baguette M (2006) Dispersal depression with habitat fragmentation in the bog fritillary butterfly. Ecology 87:1057–1065PubMedCrossRefGoogle Scholar
  65. Schultz CB, Crone EE (2005) Patch size and connectivity thresholds for butterfly habitat restoration. Conserv Biol 19:887–896CrossRefGoogle Scholar
  66. Snäll T, O’Hara RB, Ray C, Collinge SK (2008) Climate-driven spatial dynamics of plague among prairie dog colonies. Am Nat 171:238–248PubMedCrossRefGoogle Scholar
  67. Steffan-Dewenter I, Schiele S (2008) Do resources or natural enemies drive bee population dynamics in fragmented habitats? Ecology 89:1375–1387PubMedCrossRefGoogle Scholar
  68. Taylor PD, Fahrig L, Henein K, Merriam G (1993) Connectivity is a vital element of landscape structure. Oikos 68:571–573CrossRefGoogle Scholar
  69. Thiele J, Schuckert U, Otte A (2008) Cultural landscapes of Germany are patch-corridor-matrix mosaics for an invasive megaforb. Landsc Ecol 23:453–465CrossRefGoogle Scholar
  70. Thomas CD (1994) Extinction, colonization, and metapopulations: environmental tracking by rare species. Conserv Biol 8:373–378CrossRefGoogle Scholar
  71. Thomas CD, Kunin WE (1999) The spatial structure of populations. J Anim Ecol 68:647–657CrossRefGoogle Scholar
  72. Thornton DH, Branch LC, Sunquist ME (2011) The influence of landscape, patch, and within-patch factors on species presence and abundance: a review of focal patch studies. Landsc Ecol 26:7–18CrossRefGoogle Scholar
  73. Tischendorf L, Fahrig L (2000) On the usage and measurement of landscape connectivity. Oikos 90:7–19CrossRefGoogle Scholar
  74. Tischendorf L, Bender DJ, Fahrig L (2003) Evaluation of patch isolation metrics in mosaic landscapes for specialist vs. generalist dispersers. Landsc Ecol 18:41–50CrossRefGoogle Scholar
  75. Tremlová K, Münzbergová Z (2007) Importance of species traits for species distributions in fragmented landscapes. Ecology 88:965–977PubMedCrossRefGoogle Scholar
  76. Urban D, Keitt T (2001) Landscape connectivity: a graph-theoretic perspective. Ecology 82:1205–1218CrossRefGoogle Scholar
  77. Urban MC, Phillips BC, Skelly DK, Shine R (2008) A toad more traveled: the heterogeneous invasion dynamics of cane toads in Australia. Am Nat 171:E134–E138PubMedCrossRefGoogle Scholar
  78. van Nouhuys S, Hanski I (1999) Host diet affects extinctions and colonizations in a parasitoid metapopulation. J Anim Ecol 68:1248–1258CrossRefGoogle Scholar
  79. Vellend M, Verheyen K, Jacquemyn H, Kolb A, Van Calster H, Peterken G, Hermy M (2006) Extinction debt of forest plants persists for more than a century following habitat fragmentation. Ecology 87:542–548PubMedCrossRefGoogle Scholar
  80. Verbeylen G, De Bruyn L, Adriaensen F, Matthysen E (2003) Does matrix resistance influence Red squirrel (Sciurus vulgaris L. 1758) distribution in an urban landscape? Landsc Ecol 18:791–805CrossRefGoogle Scholar
  81. Vergara PM, Marquet PA (2007) On the seasonal effect of landscape structure on a bird species: the thorn-tailed rayadito in a relict forest in northern Chile. Landsc Ecol 22:1059–1071CrossRefGoogle Scholar
  82. Verheyen K, Fastenaekels I, Vellend M, De Keersmaeker L, Hermy M (2006) Landscape factors and regional differences in recovery rates of herb layer richness in Flanders (Belgium). Landsc Ecol 21:1109–1118CrossRefGoogle Scholar
  83. Visconti P, Elkin C (2009) Using connectivity metrics in conservation planning—When does habitat quality matter? Divers Distrib 15:602–612CrossRefGoogle Scholar
  84. Winfree R, Dushoff J, Crone EE, Schultz CB, Budny RV, Williams NM, Kremen C (2005) Testing simple indices of habitat proximity. Am Nat 165:707–717PubMedCrossRefGoogle Scholar
  85. With KA (2004) Metapopulation dynamics: perspectives from landscape ecology. In: Hanski I, Gaggiotti OE (eds) Ecology, genetics, and evolution of metapopulations. Elsevier Academic Press, Amsterdam, pp 23–44CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2011

Authors and Affiliations

  1. 1.Department of Natural Resources and Environmental Sciences and Program in Ecology, Evolution, and Conservation BiologyUniversity of IllinoisUrbanaUSA
  2. 2.Department of Wildlife Ecology and ConservationUniversity of FloridaGainesvilleUSA

Personalised recommendations