Schloss, Nuñez and Lawler (2012) projected that dispersal limitation will prevent on average 9.2% (locally up to 39%) of Western Hemisphere mammals from tracking projected shifts in suitable climate, with primates and eulipotyphlans most vulnerable. The authors explicitly flagged as unresolved that their dispersal estimates ignored landscape permeability, species-specific habitat preferences, topography, rivers, roads, habitat dynamics, dispersal mortality, and path sinuosity; that habitat-informed, least-cost movement modeling was infeasible for large numbers of species because of missing data and computational cost; and that the underlying bioclimatic models could not represent novel climates, evolution, or local adaptation. Climate-connectivity modeling has since become computationally feasible at continental and global scales, but species-specific dispersal distances and frequencies remain unavailable for most mammals, dispersal-aware projections are still rare and still rest on allometric estimates or binary dispersal scenarios, and no validation against observed modern mammal range shifts exists. The open problem is to build dispersal- and habitat-informed, demographically realistic range-shift projections that are feasible for large multi-species mammal faunas, and thereby determine whether the simplified dispersal assumptions under- or over-estimate the capacity of mammals to keep pace with climate change.

Background

Species have historically responded to climatic change by shifting their geographic ranges, and the velocity of climate change — the speed at which a species would need to move across the surface of the Earth to maintain constant climatic conditions, defined by Loarie et al. (2009) as the ratio of the temporal rate of climate change to the local spatial climate gradient — provides a benchmark for how fast such shifts must occur. Whether a species can actually keep pace depends on both this required velocity and its dispersal ability, the distance and frequency at which individuals move from their birth site or breeding site to new locations. Bioclimatic models (also called correlative species distribution models) forecast where climate will be suitable for a species by relating its current mapped distribution to climate variables and projecting that relationship onto future climate simulations from general circulation models; they typically represent the realized climatic niche and say nothing about movement. Schloss, Nuñez and Lawler (2012) combined bioclimatic projections for 493 nonvolant mammals of the Western Hemisphere (built with random-forest models under a mid-high A2 emissions scenario across 10 climate models) with dispersal velocities estimated from allometric body-mass relationships for natal dispersal — the movement of juveniles before their first reproduction — multiplied by the number of dispersal opportunities over 110 years. They found that on average 9.2% of mammals per location (up to 39% in some regions, notably the Amazon) will likely be unable to keep pace; 87% of species face projected range reductions, about 20% of which are attributable to dispersal limitation rather than loss of suitable climate; and 58% of range expansions projected under an unlimited-dispersal assumption become contractions once dispersal limitation is imposed. Primates and eulipotyphlans (shrews, moles and relatives) emerged as the most dispersal-limited orders. A sensitivity analysis showed the geographic and taxonomic patterns were robust to large (over 200 km per year) variation in dispersal-velocity estimates, but sensitive to dispersal frequency: substituting a breeding-dispersal ceiling for natal dispersal reduced the share of primate ranges unable to keep pace by 52%. The paper's Discussion explicitly enumerated what the analysis could not do: the dispersal estimates ignored landscape permeability — the degree to which the landscape facilitates animal movement — including human land use, topography, rivers and roads; incorporating species-specific habitat preferences and habitat change through time was judged infeasible because the required data are unavailable for most species and least-cost path modeling (identifying movement routes that balance travel through unsuitable terrain against detour length) was too computationally expensive for hundreds of species; dispersal mortality and the more tortuous paths of small-bodied species were ignored, so the reported percentages of species unable to keep pace are probably underestimates, while ignored adaptive evolution of dispersal distance or local adaptation could cut the other way; and the bioclimatic models underlying the suitable-climate maps cannot project into novel climates, account for evolution, or capture fine-scale local adaptation, so adding vegetation, soil, or alternative modeling frameworks could change the vulnerability estimates. As management responses, the authors proposed increasing landscape connectivity, assisted colonization (the intentional translocation of species to newly suitable areas beyond their dispersal reach, which they noted demands species-level cost-benefit assessment), and, above all, emissions reduction. Later work built a distinct research program on these foundations: climate connectivity — the capacity of a landscape to allow organisms to track analogous future climates, mapped with circuit-theory methods that model movement as flow through a resistance surface rather than along a single least-cost path — is now computable at continental and global scale (McGuire et al. 2016; Senior et al. 2019; Parks et al. 2020; Wu et al. 2026), and species-specific dispersal constraints have been imposed for particular regional faunas such as Amazon primates (Sales et al. 2019). Yet systematic reviews through 2024 confirm that dispersal data are still missing or nonrepresentative for most taxa, that most forecasting models still omit dispersal or collapse it into allometric estimates or binary scenarios, and that observed range shifts still do not match model expectations in ways that remain unexplained (Seaborn et al. 2020; Santini et al. 2024; Lawlor et al. 2024).

Problem Statement

The task is to make range-shift projections for mammals that incorporate realistic dispersal feasible for large numbers of species, and to determine the direction in which the simplifications used in Schloss et al. 2012-style analyses bias the resulting vulnerability estimates. Concretely, the problem is to construct projections of mammal responses to climate change in which (i) dispersal is species-specific, distinguishing natal from breeding dispersal and, where sex-biased, between sexes; (ii) movement is routed through a landscape whose permeability reflects habitat preferences, human land use, topography, rivers, roads, and habitat change through time, rather than along straight lines to the nearest suitable cell; and (iii) demographic realism includes dispersal mortality and realistic path sinuosity — and then to determine whether the earlier allometric, straight-line, mortality-free dispersal assumptions under- or over-estimate the ability of mammals to keep pace with climate change, treating the separate uncertainties the source flagged in the bioclimatic envelope (novel climates, evolution, and local adaptation) as error sources to be controlled for rather than resolved. These deliverables share one scientific objective: a reliable, species-specific assessment of the extent to which dispersal limitation will prevent mammals from tracking climate change, as needed to judge the magnitude of projected vulnerability (the 9.2% average, locally up to 39%, and the roughly 20% of projected range reductions attributed to dispersal limitation in the source analysis) and to ground the connectivity and assisted-colonization responses the source proposed. Meeting this objective requires all three conditions (i)–(iii) together with the direction determination, demonstrated at the scale of the source's benchmark — a fauna of the order of several hundred mammal species across a continental extent, not a single species or a small regional subset. Because measured species-specific dispersal values are unavailable for most mammals, satisfying (i) additionally requires a documented, auditable route from the data that do exist — measured values where available, otherwise proxy or derived estimates — to the natal, breeding and, where dispersal is sex-biased, sex-specific values used, with that route's species coverage and uncertainty reported.

References

  1. Schloss CA, Nuñez TA, Lawler JJ (2012) Dispersal will limit ability of mammals to track climate change in the Western Hemisphere. Proceedings of the National Academy of Sciences USA 109(22):8606-8611. doi:10.1073/pnas.1116791109, https://doi.org/10.1073/pnas.1116791109
  2. Loarie SR, Duffy PB, Hamilton H, Asner GP, Field CB, Ackerly DD (2009) The velocity of climate change. Nature 462:1052-1055. doi:10.1038/nature08649, https://doi.org/10.1038/nature08649
  3. McGuire JL, Lawler JJ, McRae BH, Nuñez TA, Theobald DM (2016) Achieving climate connectivity in a fragmented landscape. Proceedings of the National Academy of Sciences USA 113(26):7195-7200. doi:10.1073/pnas.1602817113, https://doi.org/10.1073/pnas.1602817113
  4. Senior RA, Hill JK, Edwards DP (2019) Global loss of climate connectivity in tropical forests. Nature Climate Change 9(8):623-626. doi:10.1038/s41558-019-0529-2, https://doi.org/10.1038/s41558-019-0529-2
  5. Parks SA, Carroll C, Dobrowski SZ, Allred BW (2020) Human land uses reduce climate connectivity across North America. Global Change Biology 26(5):2944-2955. doi:10.1111/gcb.15009, https://doi.org/10.1111/gcb.15009
  6. Littlefield CE, Krosby M, Michalak JL, Lawler JJ (2019) Connectivity for species on the move: supporting climate-driven range shifts. Frontiers in Ecology and the Environment 17(5):270-278. doi:10.1002/fee.2043, https://doi.org/10.1002/fee.2043
  7. Wu Z, Wang J, Wu H, Dai W, Wang Z (2026) Warming erodes climate connectivity for terrestrial vertebrates. Nature Climate Change 16(6):728-736. doi:10.1038/s41558-026-02658-1, https://doi.org/10.1038/s41558-026-02658-1
  8. Sales LP, Ribeiro BR, Pires MM, Chapman CA, Loyola R (2019) Recalculating route: dispersal constraints will drive the redistribution of Amazon primates in the Anthropocene. Ecography 42(10):1789-1801. doi:10.1111/ecog.04499, https://doi.org/10.1111/ecog.04499
  9. Seaborn TJ, Goldberg CS, Crespi EJ (2020) Integration of dispersal data into distribution modeling: what have we done and what have we learned? Frontiers of Biogeography 12(4). doi:10.21425/f5fbg43130, https://doi.org/10.21425/f5fbg43130
  10. Williams JE, Blois JL (2018) Range shifts in response to past and future climate change: Can climate velocities and species' dispersal capabilities explain variation in mammalian range shifts? Journal of Biogeography 45(9):2175-2189. doi:10.1111/jbi.13395, https://doi.org/10.1111/jbi.13395
  11. Santini L, Fernando M, Mancini G, Di Marco M (2024) The Neglected Role of Sex-Biased Dispersal in Range-Shift Prediction Under Climate Change. Diversity and Distributions 31(2):e13942. doi:10.1111/ddi.13942, https://doi.org/10.1111/ddi.13942
  12. Day CC, Zollner PA, Gilbert JH, McCann NP (2020) Individual-based modeling highlights the importance of mortality and landscape structure in measures of functional connectivity. Landscape Ecology 35(10):2191-2208. doi:10.1007/s10980-020-01095-5, https://doi.org/10.1007/s10980-020-01095-5
  13. Lawlor JA, Comte L, Grenouillet G, Lenoir J, Baecher JA, Bandara RMWJ, Bertrand R, Chen I-C, Diamond SE, Lancaster LT, Moore N, Murienne J, Oliveira BF, Pecl GT, Pinsky ML, Rolland J, Rubenstein M, Scheffers BR, Thompson LM, van Amerom B, Villalobos F, Weiskopf SR, Sunday J (2024) Mechanisms, detection and impacts of species redistributions under climate change. Nature Reviews Earth & Environment 5(5):351-368. doi:10.1038/s43017-024-00527-z, https://doi.org/10.1038/s43017-024-00527-z
  14. Kamal P, Thompson PL, Lewis N, Fronhofer EA (2025) Dispersal evolution can only rescue a limited set of species from climate change. Proceedings of the Royal Society B 292(2051):20250116. doi:10.1098/rspb.2025.0116, https://doi.org/10.1098/rspb.2025.0116
  15. Van der Meersch V, Armstrong E, Mouillot F, Duputié A, Davi H, Saltré F, Chuine I (2025) Paleorecords Reveal Biological Mechanisms Crucial for Reliable Species Range Shift Projections Amid Rapid Climate Change. Ecology Letters 28(2):e70080. doi:10.1111/ele.70080, https://doi.org/10.1111/ele.70080
  16. Morrison L, Estrada A, Early R (2018) Species traits suggest European mammals facing the greatest climate change are also least able to colonize new locations. Diversity and Distributions 24(9):1321-1332. doi:10.1111/ddi.12769, https://doi.org/10.1111/ddi.12769
  17. Pacifici M, Rondinini C, Rhodes JR, Burbidge AA, Cristiano A, Watson JEM, Woinarski JCZ, Di Marco M (2020) Global correlates of range contractions and expansions in terrestrial mammals. Nature Communications 11:2840. doi:10.1038/s41467-020-16684-w, https://doi.org/10.1038/s41467-020-16684-w
  18. Rangel-Rivera CE, Rojas-Soto OR, Guevara R, López-Barrera F, Arias-Alzate A, Prieto-Torres DA (2026) Will seed dispersers keep pace? Vulnerability of seed-dispersing mammals to climate change in the Mesoamerican seasonally dry forest. Regional Environmental Change 26(2):106. doi:10.1007/s10113-026-02595-x, https://doi.org/10.1007/s10113-026-02595-x