Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Human disturbance causes widespread disruption of animal movement


Disturbance and habitat modification by humans can alter animal movement, leading to negative impacts on fitness, survival and population viability. However, the ubiquity and nature of these impacts across diverse taxa has not been quantified. We compiled 208 studies on 167 species from terrestrial and aquatic ecosystems across the globe to assess how human disturbance influences animal movement. We show that disturbance by humans has widespread impacts on the movements of birds, mammals, reptiles, amphibians, fish and arthropods. More than two-thirds of 719 cases represented a change in movement of 20% or more, with increases in movement averaging 70% and decreases −37%. Disturbance from human activities, such as recreation and hunting, had stronger impacts on animal movement than habitat modification, such as logging and agriculture. Our results point to a global restructuring of animal movement and emphasize the need to reduce the negative impacts of humans on animal movement.


Humans have transformed the world’s ecosystems through agriculture, resource extraction, urbanization, pollution, dredging and many other disturbances1,2. Although some relatively intact sea- and landscapes remain, they are not immune to anthropogenic impacts. Roads, tourism, recreation, hunting, shipping, fishing and other disturbances can permeate into even the most remote protected areas3. These disturbances alter resource availability and habitat connectivity and can exacerbate other threats. In response, many animals in human-modified environments either have their normal movement behaviour constrained or must move farther to survive4,5. The consequences of altered movement can be profound, leading to reduced fitness and survival, lower reproductive rates, genetic isolation and local extinction6,7. However, the scale and prevalence of altered animal movement across taxonomic groups is not clear.

Whether animals increase, decrease or show no change in movement may depend on the characteristics of the species and the nature of the disturbance. For instance, movement can be reduced by persistent structural changes to a landscape, such as barriers caused by urbanization or agriculture4,8, or by resource subsidies or reduced competition and predation that benefit animals9,10. In contrast, immediate threats within remaining habitat, such as hunting, may cause animals to increase movement as they flee an unpredictable threat11,12. Movement also scales positively with body size, and carnivores move farther than herbivores13,14, but we do not know how disturbances interact with these traits to shape animal movement. In this Article, we compiled studies from across the globe to assess how animals from a range of taxonomic groups change their movement in response to habitat modification or disturbance from humans (hereafter ‘disturbance’) and tested how these responses vary according to taxonomic group, trophic level, body mass and disturbance type. In this way, we provide a global assessment of the ubiquity of wildlife movement responses to human activities.

Results and discussion

We compiled a database of studies of animal movement from terrestrial, marine and freshwater ecosystems (Methods and Extended Data Fig. 1). The database includes birds (37 species), mammals (77), reptiles (17), amphibians (11), fish (13) and arthropods (12) spanning seven orders of magnitude in animal body size, from the 0.05 g sleepy orange butterfly (Eurema nicippe) to the >2 t great white shark (Carcharodon carcharias; Fig. 1). We extracted data on home range size and movement distances in disturbed and undisturbed treatments in before–after (BA), control–impact (CI), or BACI designs that compared, for example, fragmented with continuous habitat, hunted with non-hunted areas or polluted with non-polluted habitat. We used the means from each treatment to calculate log–response ratios (lnRR), where values of zero represent no effect, positive values represent increased movement in response to disturbance and negative values represent reduced movement (Methods and Extended Data Fig. 2). Overall, our database comprised 719 effect sizes from 208 studies, 167 species and 6 continents (Fig. 1). We recorded the disturbance type for each effect size and the body mass (g) and trophic level (herbivore, omnivore or carnivore) of each species (Methods and Extended Data Fig. 3). We used Bayesian mixed-effects models to determine how movement responses varied according to taxonomic group, body size, trophic level and disturbance type (Methods).

Fig. 1: The global extent and magnitude of the impacts of disturbance on animal movement.
figure 1

a, Study locations, with each point representing an individual case (N = 208 studies and 719 effect sizes). b, Histograms of effect sizes (percentage change in movement in response to disturbance). Bin size is 25%. Five movement distance and six home range effect sizes >500% are not shown. c, Photos of representative taxa included in our database, left to right: sleepy orange butterfly (Eurema nicippe), southern leopard frog (Rana sphenocephala), tawny owl (Strix aluco), red-eared slider turtle (Trachemys scripta), diademed sifaka (Propithecus diadema) and great white shark (Carcharodon carcharias). Photos adapted from Flickr under Creative Commons license CC BY 2.0.

Animals were more likely to increase than decrease their movement distances in response to disturbance (N = 231 positive, 141 negative and 4 neutral effect sizes), whereas there were similar numbers of positive (177) and negative (163) home range responses (and 3 neutral). Across both movement types, the mean positive effect was a 70% increase in movement in response to disturbance (back-transformed 95% credible interval (CI): 57–84%) and the mean negative effect was a 37% decrease (−42 to −33%; Extended Data Fig. 4). Of the 719 cases, 67% comprised an increase or decrease in movement of 20% or more and 37% a change of 50% or more, indicating that substantial changes in animal movement are very common. For movement distances, the overall effect of disturbance was positive (mean lnRR = 0.15, 95% CI = 0.06–0.24, N = 376), representing a mean 16% increase in movement distances, whereas the overall effect of disturbance on home range size was neutral (0.02, −0.11–0.15, 343). Changes in movement represent a range of behavioural mechanisms, including fleeing or avoiding humans15, travelling farther to find food or mates16, sheltering to avoid humans or predators17, reduced movement due to anthropogenic resource subsidies10, increased or decreased efficiency of movement through modified habitat18 and restriction of movement by physical barriers19, among others.

The effects of disturbance differed among taxonomic groups and movement types, with mean movement distances increasing by 38% for arthropods (back-transformed 95% CI: 6–79%), 27% for birds (−2–67%) and 19% for mammals (4–34%; Fig. 2 and Extended Data Fig. 5). For example, moose (Alces alces) increased their movement rates in response to skiers, aircraft, off-road vehicles, military activity and when crossing roads. There were weak negative and positive effects for amphibian and reptile movement distances, respectively, and no clear directional effect for fish (Fig. 2). Birds increased their home range size by 43% on average (11–84%; Fig. 2) and reptiles decreased their home range size by 21% on average, although there was high uncertainty in this effect (−47–17%). In some cases, bird home range sizes doubled in response to logging20, whereas reptile home range size decreased by up to 75% in response to urbanization21. There was no clear directional effect of disturbance on home range size for fish or mammals (Fig. 2 and Extended Data Fig. 5).

Fig. 2: Effects of disturbance on animal movement distances and home range size according to taxonomic groups.
figure 2

Positive effects represent increased movement in response to disturbance, and the opposite for negative effects. Symbols represent posterior means and coloured bands represent 95%, 80% and 50% CIs. Silhouettes adapted from PhyloPic under Creative Commons licenses CC0 1.0 and CC BY 3.0.

All trophic levels increased their mean movement distances in response to disturbance, with herbivores responding slightly more positively than omnivores and carnivores, but the CIs overlapped for all groups (Extended Data Figs. 6 and 7). There were no clear differences between trophic levels for changes in home range size (Extended Data Figs. 6 and 7). Given that animal movements scale allometrically with body size across a range of taxa13,14, it might be expected that larger species more commonly increase their movement due to an ability to move at a scale that exceeds that of the disturbance. Indeed, this was the case with birds and there was some evidence for this in mammals. Birds with higher body mass (~0.1–5.8 kg) typically increased their home range (slope of relationship between body size and lnRR: 0.12, 0–0.25) and moved farther (0.07, −0.05–0.19) in response to disturbance, whereas smaller birds showed a mixture of increasing and decreasing responses (Fig. 3). Movement distances of mammals became increasingly positive with increased body size (0.02, −0.01–0.06), particularly for species >10 kg, including ungulates such as mule deer, mountain sheep, ibex, caribou, elk and moose (Fig. 3 and Extended Data Fig. 8). However, arthropods and reptiles showed the opposite response, with changes in movement distances becoming increasingly negative as body mass increased (arthropods: −0.22, −0.41 to −0.03; reptiles: −0.14, −0.31–0.05; Fig. 3). This suggests that thermoregulatory capacity may interact with body size to affect movement responses22,23, although it requires further study because there were no clear effects for reptile or mammal home ranges, or for the other ectotherms (amphibians and fish), which had the least amount of data.

Fig. 3: Effects of body mass on animal movement distances (MD) and home range size (HR) in response to disturbance.
figure 3

Positive effects represent increased movement in response to disturbance, and the opposite for negative effects. Green circles represent individual effect sizes, black lines represent mean relationships and grey bands represent 95% CIs. Darker circles represent areas where points overlap.

Human activities, such as hunting, aircraft and recreation, caused much stronger increases in movement distances (35%, 15–60%) than habitat modification, such as logging, agriculture and habitat fragmentation (12%, 1–25%; Extended Data Fig. 9) did. This difference can also be seen in the analysis of individual disturbance types for mammals, where average movement distances increased by 28% in response to agriculture, 65% in response to aircraft and 68% in response to roads (Fig. 4 and Extended Data Fig. 9). There were no clear directional effects of either of the two broad disturbance types on home range size (human activities: 4%, −20–35%; habitat modification: 8%, −8–26%; Extended Data Fig. 9). However, mammal home range size decreased by 49% and 27%, on average, in response to urbanization and agriculture, respectively (Fig. 4). These negative responses support other recent studies that have found smaller home ranges of caribou24 and red foxes9 and smaller movement distances of mammals4 in areas of high human footprint or disturbance. In contrast, the overall increase in mammal movement distances that we recorded differs from the previous studies, possibly because our analyses included a range of disturbances caused by human activity, which caused stronger increases in movement. Other key results for individual taxa include increases in bird movement distance (mean 52%) and home range (51%) in response to fragmentation and increases in arthropod, reptile and fish movement distances of 35%, 79% and 63% in response to fragmentation, grazing and pollution, respectively (Fig. 4 and Extended Data Fig. 9).

Fig. 4: Effects of different disturbance types on animal movement distances and home range size according to taxonomic groups.
figure 4

Positive effects represent increased movement in response to disturbance, and the opposite for negative effects. Symbols represent posterior means and coloured bands represent 95%, 80% and 50% CIs. The black box encloses the results for mammals. Silhouettes adapted from PhyloPic under Creative Commons licenses CC0 1.0 and CC BY 3.0.

Our results point to a global restructuring of animal movement, with potentially profound impacts on populations, species and ecosystem processes. Increased or decreased movement in response to disturbance represents a trade-off between the energetic costs of moving and the fitness consequences of acquiring resources and avoiding threats25,26. In some cases, animals decrease their movements because exploitation of anthropogenic resources means they need to travel less to find food9,10. However, in many other cases, changes in movements will be costly and ultimately impact fitness and reproductive success, through either increased energy expenditure when avoiding threats or searching for resources15,16,25, or reduced access to resources or mates when movement is constrained17,27. Even a small change in movement can have big impacts on an individual and when these costs accumulate across an entire population, reproductive rates and population viability may be compromised7,16,27. Further, because animal movement is linked to many functional network processes, such as pollination, seed dispersal, soil turnover and predator–prey dynamics, disrupted movement can have cascading ecosystem impacts17,28,29. For instance, in the United States, experimental playback of humans talking caused a mean 34% decrease in mountain lion movement speed and decreases in the activity of medium-sized carnivores, resulting in cascading benefits for small rodents that increased their space use and foraging efficiency by 45% and 17%, respectively17. In New Zealand, flightless rails (Gallirallus australis) provide important seed dispersal services, but birds in areas of high human activity (campgrounds) moved 35–41% shorter distances than birds away from campgrounds, thus impairing their seed dispersal potential29. Such evidence demonstrates substantive, ecosystem-wide consequences from changes in animal movement due to human disturbances.

The stronger impacts of human activities on animal movement compared with habitat modification may reflect the acute and unpredictable nature of the former (for example, hunting and recreation), whereas the latter generally manifests as persistent structural changes to landscapes (for example, logging and habitat fragmentation). For disturbances that are episodic in nature, it may in fact be easier to reduce their impacts, such as by banning certain activities in wilderness areas30 or avoiding animals’ breeding periods or key activity times31. Regarding habitat modification, our results support calls for avoiding the transformation or degradation of sea- and landscapes that currently remain relatively unmodified. Historical landscape modifications will be more difficult to address and may rely on improving connectivity or resource availability32,33,34. Where habitat modification is unavoidable, we recommend the integration of movement ecology principles into landscape design and management to facilitate animal movement, resource acquisition, reproduction and dispersal35. Reducing the negative impacts of humans on animal movement will be of key importance for biodiversity conservation in an increasingly human-dominated world.


Literature search

We searched Scopus and Web of Science (Core Collection, BIOSIS, SciELO and Zoological Record) on 1 May 2019 using the following search string: (‘home range’ OR ‘home-range’ OR ‘space use’ OR ‘movement rate’ OR ‘movement pattern’ OR ‘movement ecology’ OR ‘animal movement’) AND (fragmented OR ‘habitat fragmentation’ OR ‘habitat loss’ OR ‘habitat modification’ OR ‘habitat degradation’ OR ‘land use change’ OR agricultur* OR farmland OR urban* OR forestry OR logging OR plantation OR mining OR road* OR dams OR damming OR fishing OR hunting OR poaching OR dredg* OR pollut* OR aquaculture OR disturbance OR anthropogenic). We restricted the Scopus search to the subject areas of environmental science and agricultural and biological sciences and the Web of Science search to the research areas of environmental sciences, ecology, zoology, biodiversity conservation, marine freshwater biology, forestry and fisheries. This returned 10,267 results from Web of Science and 5,300 from Scopus, with 12,021 unique records remaining after removing duplicates (Extended Data Fig. 1).

We first read paper titles and abstracts to exclude papers that were clearly not relevant on the basis of the inclusion criteria below. We then inspected 975 full texts to determine their suitability for the study. For a study to be included in the database, it had to be a peer-reviewed journal article that met the following criteria: (1) study of animal home range size or movement distances in either: disturbed and undisturbed treatments (for example, grazed versus ungrazed) or before and during/after a disturbance (for example, pre- versus post-logging, or before versus during road crossing); (2) data were collected in a standardized way across treatments; (3) estimates were available for >1 individual animal in each treatment and recorded at the species level and (4) sufficient information was available to calculate mean movement parameters, and, where possible, sample sizes and standard deviations. We excluded from the database and all analyses the single data point for a non-arthropod invertebrate (the sea urchin Paracentrotus lividus). Our final database comprised data from 208 studies reporting 719 cases.

Data collation

We extracted home range and movement distance data from the text, tables, figures and appendices of papers. We used WebPlotDigitiser to extract data from figures36. We recorded separate estimates where data were recorded for discrete groups, such as age classes, sexes, years, seasons or study regions. We only extracted one set of values where a study reported home range size based on multiple estimators (for example, 95% minimum convex polygon and 95% fixed kernel home range). Where available, we chose kernel estimates over minimum convex polygons, because the former is generally considered to provide a more accurate measure of space use37. There were seven studies that reported data from multiple time points before and/or after disturbance. In these cases, we only extracted one set of values that was closest in time to the disturbance, with those time periods falling within the range used by the remaining studies. If standard deviations were not reported, we derived these from the sample size and standard error or confidence intervals, where possible. If this was not possible (for example, due to sample sizes also not being reported), we contacted authors to request missing values. In cases where we were ultimately unable to obtain missing values, we only included those studies in the unweighted analyses (described below).

We categorized response variables into two broad groups. The ‘distance’ group included linear measures of animal movement, such as daily movement distance, movement rate and step length. We conducted a supplementary analysis to confirm that the pooling of these different measures was appropriate (described below). The second group (‘home range’) included any areal measures of home range, core area or territory size, typically estimated using minimum convex polygon or kernel density methods. Home range metrics generally related to animal space use measured over weeks to months, while movement distances represented shorter range movements within the home range, measured over seconds to days. Some tracking methods and home range estimators may either over- or underestimate movement more than others. However, any such bias is constrained within each study and does not propagate across effect sizes because each effect size is derived from animals tracked in either control and impact areas or before and after disturbance using the same methods within the same ecosystem and at the same time. As such, the effect size (lnRR, described below) represents the proportional change in movement in response to disturbance and is directly comparable across studies.

We classified disturbance types as one of agriculture, grazing, logging, hunting, tourism and recreation, aircraft, boats, roads, military, mining, pollution, urbanization, habitat fragmentation and loss or other (Extended Data Fig. 3). We also used a higher-level classification that distinguished habitat-modification effects from disturbance by human activity. The habitat modification group included any disturbance that altered landscape structure, resource availability or habitat quality, including agriculture, fragmentation, grazing, logging, mining, pollution and any relevant studies from the ‘other’ category. The human activity group included any disturbance that related to the presence of humans, their activities or vehicles (including roads, aircraft, boats, hunting, military activities, tourism and recreation) and any relevant studies from the ‘other’ category. There were some exceptions to these rules. We placed two of the mining effect sizes and four of the pollution effect sizes in the human activities category because they involved either simulated noise disturbance from seismic activity, radar disturbance or artificial light at night. We also excluded from the classification of broad disturbance type 75 cases where it was not possible to clearly distinguish the effects of human activities from those of habitat modification (for example, urbanization that caused changes in both habitat availability and the presence of humans, or logging where changes in forest structure were confounded with the actual logging process).

We used the metafor package in R38 to calculate lnRR39 according to the formula:

$${\mathrm{lnRR}} = {\mathrm{ln}}\left( {\frac{{Y_{\mathrm{A}}}}{{Y_{\mathrm{B}}}}} \right)$$

where Y is the mean home range size or movement distance in disturbed (A) and undisturbed (B) treatments. Using this approach, values of zero represent no effect, positive values represent greater movement in disturbed treatments and negative values represent reduced movement in disturbed treatments. The exponent of lnRR can be interpreted as the percentage difference between the two treatments. This index provides a measure of the relative change in movement and so facilitates comparison of impacts across a range of body sizes and response variables (for example, different home range metrics). Sampling variance was calculated as:

$$v_{{\mathrm{lnRR}}} = \frac{{{\mathrm{s.d.}}_{\mathrm{A}}^2}}{{N_{\mathrm{A}}Y_{\mathrm{A}}^2}} + \frac{{{\mathrm{s.d.}}_{\mathrm{B}}^2}}{{N_{\mathrm{B}}Y_{\mathrm{B}}^2}}$$

where N represents the sample size (number of animals) for treatments A and B. We converted any studies with a BACI design to CI by dividing ‘after’ values by ‘before’ values for each treatment. We excluded one effect size (lnRR = 5.64) that was a clear outlier compared with the remaining data (mean = 0.11, range = −3.51–3.53). This effect size related to the change in moose movement rate in response to skiing disturbance40 and when back-transformed was an order of magnitude higher than the next highest.

Collation of body mass and trophic level data

We obtained body mass data for most birds, mammals, reptiles, amphibians and fish from published databases41,42,43,44. The data available for birds and mammals were mean adult weights whereas the data available for reptiles, amphibians and fish were maximum adult weights, but this does not present a problem for the analyses because the taxonomic groups were analysed separately. For all arthropods and any other species without mass data in the former databases, we sourced values from journals, books, reports, the Animal Diversity website45 and AmphibiaWeb46. For some fish species, we estimated maximum weight using allometric equations and maximum reported lengths47. For two species (Pomacentrus wardi and Hypseleotris compressa), the formulas were based on data at the sub-family or family level. For one amphibian species that was a hybrid (Plethodon shermani x chattahoochee), values were only available for one of the true species (P. shermani), which is what we used. We were unable to obtain a mass value for one arthropod species (Prokelisia crocea).

We classified the trophic level of each study species as either herbivore, omnivore or carnivore. For birds and mammals, we based the classification on dietary information contained in the EltonTraits database41. That database records the percentage of a species’ diet (in intervals of 10) made up of 10 categories: four related to plants, one related to invertebrates and five related to vertebrates and scavenging. We classified herbivores as species whose diet is ≥90% plants, carnivores as species whose diet is ≥90% vertebrates and/or invertebrates and all other species as omnivores. For lizards, we used the classifications of Meiri48. We sourced information for arthropods, fish, snakes, turtles and amphibians from journals, books, reports, FishBase44, the Animal Diversity website45 and AmphibiaWeb46. Due to the descriptive nature of the information available, we classified herbivores as species only reported to feed on plants/algae, carnivores as species feeding only on vertebrates and/or invertebrates and all other species as omnivores.

Statistical analyses

We used Bayesian mixed-effects models implemented in the brms package in R49 to analyse variation in effect sizes. We fitted models assuming a normal distribution and included random effects for species and study ID to account for non-independence between effect sizes from the same study or species. We also tested models with additional random effects of taxonomic order and family but found that the model with only species was superior. We specified weakly informative half Cauchy priors (mean = 0, s.d. = 1) for the random effects50,51. We analysed the home range and movement distance data separately and classified each data point according to taxonomic group (birds, mammals, reptiles, amphibians, fish and arthropods), disturbance type, body mass and trophic level. We fitted six models for each of the movement distance and home range size data:

  1. (1)

    Null model to estimate overall effect sizes. We fitted separate models on the full dataset for each movement type and then individual models for effect sizes either ≥0 or ≤0 to estimate mean positive and negative effects for both movement types combined.

  2. (2)

    Taxonomic group (birds, mammals, reptiles, amphibians, fish and arthropods). Amphibians were excluded from the home range model because there were only two effect sizes.

  3. (3)

    Trophic level (herbivores, omnivores and carnivores).

  4. (4)

    Body mass (by taxonomic group). We log transformed body mass to improve normality.

  5. (5)

    Disturbance type (by taxonomic group). This analysis was restricted to combinations of taxonomic group and disturbance that had a minimum of 10 effect sizes from a minimum of 3 studies (combined). We excluded the ‘other’ category of disturbance type.

  6. (6)

    Broad disturbance type (habitat modification versus human activity). We excluded 75 cases where it was not possible to clearly distinguish the effects of human activities from those of habitat modification.

For models with categorical predictor variables (2, 3, 5 and 6), we excluded the intercept from the model formulation to allow the estimation of coefficients for every factor level, rather than one level being used as the reference level. To aid model convergence of the disturbance type model, we used a combined taxonomic group–disturbance type variable (for example, fragmentation–birds), rather than an interaction between the two. This also allowed us to exclude the intercept. We ran 4 chains of 10,000 iterations each for each model, with a burn-in of 1,000 iterations, resulting in 36,000 samples. We assessed convergence by inspecting trace plots and ensuring that the Gelman–Rubin statistic52 was <1.1. We present posterior means and 95%, 80% and 50% CIs.

We performed unweighted analyses on the full dataset (N = 719 effect sizes) and weighted analyses where we were able to calculate sampling variances (521). Effect sizes were weighted by their inverse standard errors. In the large majority of cases, the direction and magnitude of effects was very similar between the weighted and unweighted analyses. However, there were some exceptions. The clearest example is the mean effect of human activities on movement distances, which was 0.30 for the unweighted analysis and 0.47 for the weighted analysis. There was a range of other cases where the weighted results were either stronger or weaker than the unweighted results (Extended Data Figs. 46, 8 and 9). However, given that there is no clear bias for one set of results being stronger than the other, we present the unweighted analysis in the main text and the weighted analysis as Extended Data, because the unweighted analysis uses the full dataset. For ease of interpretation, we present some results as back-transformed model coefficients that represent the percentage change in movement. All original model results are provided as Extended Data. To provide a high-level indicator of how common changes in animal movement are, we also report the percentage of studies where the change in movement exceeded arbitrary cut-offs of ±20% and ±50%.

Publication bias and supplementary analyses

We used funnel plots, Egger’s test53 and fail-safe numbers54 to test for potential publication bias in our database. There was no indication of publication bias for the home range data (Extended Data Fig. 2; Egger’s test: z = 1.38, P = 0.1675). The funnel plot for the movement distance data was mildly asymmetrical (more positive values; Extended Data Fig. 2) and Egger’s test was significant (z = 2.39, P = 0.017), but the fail-safe analysis indicated that 6,062 additional studies with null results would need to be added to the dataset to reduce the significance level to P= 0.05. As such, we do not consider publication bias to be an issue for interpretation of the results.

We conducted an analysis to confirm that the pooling of various linear measures of movement in the distance group was appropriate. Specifically, we refitted the taxonomic class model by only including distance data that were clearly some form of movement rate (distance moved per unit time). Mean effects were very similar for mammals (all data: 0.17; subset: 0.19), birds (0.24; 0.23), arthropods (0.32; 0.35) and reptiles (0.12; 0.13). The amphibian effect size decreased (−0.15; −0.29) and the fish effect size increased (0.05; 0.20), although the CIs were wide in both cases, indicating high uncertainty across both datasets. Given these results, we are confident that our results based on the full movement distance dataset are robust.

There was some geographic bias in the available data, with most studies coming from North America or Europe. To investigate potential geographic differences in the results, we fitted an unweighted model to each of the movement distance and home range data with ‘region’ as a predictor (Africa, Asia, Europe, North America, Oceania and South America). There were only two data points for Central America, which we pooled with South America.

The movement distance model showed that the regions were broadly similar, with Africa having the largest mean effect (0.30), followed by Europe (0.20), North America (0.15), Asia (0.11), South America (0.09) and Oceania (−0.01; Extended Data Fig. 10). The CIs of Africa, Asia, Oceania and South America almost completely spanned those of Europe and North America (Extended Data Fig. 10). The home range results were more variable, with Africa, Europe and South America all having very similar results (0.10–0.19), whereas North America was lower (0.01), and Asia and Oceania lower still (−0.28 and −0.32; Extended Data Fig. 10). Asia and Oceania also had the lowest sample sizes (N = 14 and 18 effect sizes, respectively) and widest CIs. The CIs of Europe and North America were again almost completely encapsulated by the other regions (Extended Data Fig. 10). Given the broad similarity of these results and the low level of replication for some regions, we proceeded with pooling data across regions and did not make any further comparisons between regions.

Reporting Summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

Data is available at Figshare

Code availability

Code is available at Figshare


  1. Newbold, T. et al. Global effects of land use on local terrestrial biodiversity. Nature 520, 45–50 (2015).

    CAS  PubMed  Article  Google Scholar 

  2. Halpern, B. S. et al. A global map of human impact on marine ecosystems. Science 319, 948–952 (2008).

    CAS  PubMed  Article  Google Scholar 

  3. Queiroz, N. et al. Global spatial risk assessment of sharks under the footprint of fisheries. Nature 572, 461–466 (2019).

    CAS  PubMed  Article  Google Scholar 

  4. Tucker, M. A. et al. Moving in the anthropocene: global reductions in terrestrial mammalian movements. Science 359, 466–469 (2018).

    CAS  PubMed  Article  Google Scholar 

  5. Wang, X. et al. Stochastic simulations reveal few green wave surfing populations among spring migrating herbivorous waterfowl. Nat. Commun. 10, 2187 (2019).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  6. Fahrig, L. Non-optimal animal movement in human-altered landscapes. Funct. Ecol. 21, 1003–1015 (2007).

    Article  Google Scholar 

  7. Cosgrove, A. J., McWhorter, T. J. & Maron, M. Consequences of impediments to animal movements at different scales: a conceptual framework and review. Divers. Distrib. 24, 448–459 (2018).

    Article  Google Scholar 

  8. Mergey, M., Helder, R. & Roeder, J.-J. Effect of forest fragmentation on space-use patterns in the European pine marten (Martes martes). J. Mammal. 92, 328–335 (2011).

    Article  Google Scholar 

  9. Main, M. T., Davis, R. A., Blake, D., Mills, H. & Doherty, T. S. Human impact overrides bioclimatic drivers of red fox home range size globally. Divers. Distrib. (2020).

  10. Laver, P. N. & Alexander, K. A. Association with humans and seasonality interact to reverse predictions for animal space use. Mov. Ecol. 6, 5 (2018).

    PubMed  PubMed Central  Article  Google Scholar 

  11. Riotte-Lambert, L. & Matthiopoulos, J. Environmental predictability as a cause and consequence of animal movement. Trends Ecol. Evol. 35, 163–174 (2020).

    PubMed  Article  Google Scholar 

  12. Laurian, C., Ouellet, J.-P., Courtois, R., Breton, L. & St-Onge, S. Effects of intensive harvesting on moose reproduction. J. Appl. Ecol. 37, 515–531 (2000).

    Article  Google Scholar 

  13. Tamburello, N., Côté, I. M. & Dulvy, N. K. Energy and the scaling of animal space use. Am. Nat. 186, 196–211 (2015).

    PubMed  Article  Google Scholar 

  14. Jetz, W., Carbone, C., Fulford, J. & Brown, J. H. The scaling of animal space use. Science 306, 266–268 (2004).

    CAS  PubMed  Article  Google Scholar 

  15. Perona, A. M., Urios, V. & López-López, P. Holidays? Not for all. Eagles have larger home ranges on holidays as a consequence of human disturbance. Biol. Conserv. 231, 59–66 (2019).

    Article  Google Scholar 

  16. Staggenborg, J., Schaefer, H. M., Stange, C., Naef-Daenzer, B. & Grüebler, M. U. Time and travelling costs during chick-rearing in relation to habitat quality in little owls Athene noctua. Ibis (Lond. 1859) 159, 519–531 (2017).

    Article  Google Scholar 

  17. Suraci, J. P., Clinchy, M., Zanette, L. Y. & Wilmers, C. C. Fear of humans as apex predators has landscape-scale impacts from mountain lions to mice. Ecol. Lett. 22, 1578–1586 (2019).

    PubMed  Article  Google Scholar 

  18. Blomquist, S. & Hunter, M. L. Jr A multi-scale assessment of habitat selection and movement patterns by northern leopard frog (Lithobates [Rana] pipiens) in a managed forest. Herpetol. Conserv. Biol. 4, 142–160 (2009).

    Google Scholar 

  19. Peaden, J. M., Nowakowski, A. J., Tuberville, T. D., Buhlmann, K. A. & Todd, B. D. Effects of roads and roadside fencing on movements, space use, and carapace temperatures of a threatened tortoise. Biol. Conserv. 214, 13–22 (2017).

    Article  Google Scholar 

  20. Siffczyk, C., Brotons, L., Kangas, K. & Orell, M. Home range size of willow tits: a response to winter habitat loss. Oecologia 136, 635–642 (2003).

    PubMed  Article  Google Scholar 

  21. Breininger, D. R., Bolt, M. R., Legare, M. L., Drese, J. H. & Stolen, E. D. Factors influencing home-range sizes of eastern indigo snakes in central Florida. J. Herpetol. 45, 484–490 (2011).

    Article  Google Scholar 

  22. Hirt, M. R., Jetz, W., Rall, B. C. & Brose, U. A general scaling law reveals why the largest animals are not the fastest. Nat. Ecol. Evol. 1, 1116–1122 (2017).

    PubMed  Article  Google Scholar 

  23. Garland, T. & Albuquerque, R. L. Locomotion, energetics, performance, and behavior: a mammalian perspective on lizards, and vice versa. Integr. Comp. Biol. 57, 252–266 (2017).

    PubMed  Article  Google Scholar 

  24. Wilson, K. S., Pond, B. A., Brown, G. S. & Schaefer, J. A. The biogeography of home range size of woodland caribou Rangifer tarandus caribou. Divers. Distrib. 25, 205–216 (2019).

    Article  Google Scholar 

  25. Wang, Y., Smith, J. A. & Wilmers, C. C. Residential development alters behavior, movement, and energetics in a top carnivore. PLoS ONE 12, e0184687 (2017).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  26. Vangestel, C., Braeckman, B. P., Matheve, H. & Lens, L. Constraints on home range behaviour affect nutritional condition in urban house sparrows (Passer domesticus). Biol. J. Linn. Soc. Lond. 101, 41–50 (2010).

    Article  Google Scholar 

  27. Hinam, H. L. & St. Clair, C. C. High levels of habitat loss and fragmentation limit reproductive success by reducing home range size and provisioning rates of northern saw-whet owls. Biol. Conserv. 141, 524–535 (2008).

    Article  Google Scholar 

  28. Herrera, J. M., de Sá Teixeira, I., Rodríguez-Pérez, J. & Mira, A. Landscape structure shapes carnivore-mediated seed dispersal kernels. Landsc. Ecol. 31, 731–743 (2016).

    Article  Google Scholar 

  29. Carpenter, J. K., O’Donnell, C. F. J., Moltchanova, E. & Kelly, D. Long seed dispersal distances by an inquisitive flightless rail (Gallirallus australis) are reduced by interaction with humans. R. Soc. Open Sci. 6, 190397 (2019).

    PubMed  PubMed Central  Article  Google Scholar 

  30. Januchowski-Hartley, F. A., Graham, N. A. J., Feary, D. A., Morove, T. & Cinner, J. E. Fear of fishers: human predation explains behavioral changes in coral reef fishes. PLoS ONE 6, e22761 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  31. Whittington, J., Low, P. & Hunt, B. Temporal road closures improve habitat quality for wildlife. Sci. Rep. 9, 3772 (2019).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  32. Soanes, K. et al. Movement re-established but not restored: inferring the effectiveness of road-crossing mitigation for a gliding mammal by monitoring use. Biol. Conserv. 159, 434–441 (2013).

    Article  Google Scholar 

  33. Jacobsen, L. B., Chrenková, M., Sunde, P. & Salek, M. Effects of food provisioning and habitat management on spatial behaviour of little owls during the breeding season. Ornis Fenn. 93, 121–129 (2016).

    Google Scholar 

  34. Zeller, K. A., Lewsion, R., Fletcher, R. J., Tulbure, M. G. & Jennings, M. K. Understanding the importance of dynamic landscape connectivity. Land (Basel) 9, 303 (2020).

    Google Scholar 

  35. Doherty, T. S. & Driscoll, D. A. Coupling movement and landscape ecology for animal conservation in production landscapes. Proc. R. Soc. Lond. B 285, 20172272 (2018).

    Google Scholar 

  36. Rohatgi, A. WebPlotDigitizer, version 4.2 (2019);

  37. Börger, L. et al. Effects of sampling regime on the mean and variance of home range size estimates. J. Anim. Ecol. 75, 1393–1405 (2006).

    PubMed  Article  Google Scholar 

  38. Viechtbauer, W. Conducting meta-analyses in R with the metafor package. J. Stat. Softw. 36, 1–48 (2010).

    Article  Google Scholar 

  39. Hedges, L. V., Gurevitch, J. & Curtis, P. S. The meta-analysis of response ratios in experimental ecology. Ecology 80, 1150–1156 (1999).

    Article  Google Scholar 

  40. Neumann, W., Ericsson, G. & Dettki, H. Does off-trail backcountry skiing disturb moose? Eur. J. Wildl. Res. 56, 513–518 (2010).

    Article  Google Scholar 

  41. Wilman, H. et al. EltonTraits 1.0: species-level foraging attributes of the world’s birds and mammals. Ecology 95, 2027 (2014).

    Article  Google Scholar 

  42. Feldman, A., Sabath, N., Pyron, R. A., Mayrose, I. & Meiri, S. Body sizes and diversification rates of lizards, snakes, amphisbaenians and the tuatara. Glob. Ecol. Biogeogr. 25, 187–197 (2016).

    Article  Google Scholar 

  43. Oliveira, B. F., São-Pedro, V. A., Santos-Barrera, G., Penone, C. & Costa, G. C. AmphiBIO, a global database for amphibian ecological traits. Sci. Data 4, 170123 (2017).

    PubMed  PubMed Central  Article  Google Scholar 

  44. Froese, R. & Pauly, D. (eds) Fishbase (2019);

  45. Myers, P. et al. The Animal Diversity Web (Univ. Michigan, 2020);

  46. AmphibiaWeb (Univ. California Berkeley, 2020);

  47. Froese, R., Thorson, J. T. & Reyes, R. B. A Bayesian approach for estimating length–weight relationships in fishes. J. Appl. Ichthyol. 30, 78–85 (2014).

    Article  Google Scholar 

  48. Meiri, S. Traits of lizards of the world: variation around a successful evolutionary design. Glob. Ecol. Biogeogr. 27, 1168–1172 (2018).

    Article  Google Scholar 

  49. Bürkner, P. C. Advanced Bayesian multilevel modeling with the R package brms. R J. 10, 395–411 (2018).

    Article  Google Scholar 

  50. Bürkner, P. C. brms: an R package for Bayesian multilevel models using Stan. J. Stat. Softw. 80, 1–28 (2017).

    Article  Google Scholar 

  51. Williams, D. R., Rast, P. & Bürkner, P.-C. Bayesian meta-analysis with weakly informative prior distributions. Preprint at PsyArXiv (2018).

  52. Gelman, A. & Rubin, D. B. Inference from iterative simulation using multiple sequences. Stat. Sci. 7, 457–511 (1992).

    Google Scholar 

  53. Egger, M., Davey Smith, G., Schneider, M. & Minder, C. Bias in meta-analysis detected by a simple, graphical test. BMJ 315, 629–634 (1997).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  54. Rosenberg, M. S. The file-drawer problem revisted: a general weighted method for calculating fail-safe numbers in meta-analysis. Evolution 59, 464–468 (2005).

    PubMed  Article  Google Scholar 

Download references


We thank the following people for providing additional information about their studies: L. Amo, J. Beasley, K. Borkin, H. Ling Chen, T. Crist, S. Dale, C. Dussault, C. Gómez Posada, T. Gehring, N. Haddad, C. Huveneers, C. Lanctôt, K. Mabry, L. Powell, A. Trochet, C. Vangestel, K. VerCauteren and D. Zeller. We acknowledge the Wurundjeri people of the Kulin nations as the traditional custodians of the land on which this review was conducted. We acknowledge the technical assistance of H. Lydecker from the Sydney Informatics Hub, a Core Research Facility of the University of Sydney, and the use of the University of Sydney’s high performance computing cluster, Artemis. T.S.D. was supported by an Alfred Deakin Postdoctoral Research Fellowship from Deakin University and a Discovery Early Career Researcher Award from the Australian Research Council.

Author information

Authors and Affiliations



T.S.D. conceived the study, collated and analysed the data, and led the writing of the manuscript. G.C.H. and D.A.D. helped write the manuscript.

Corresponding author

Correspondence to Tim S. Doherty.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Ecology & Evolution thanks Laura Prugh and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Database collation summary.

PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) diagram detailing the procedure for identifying and including relevant publications.

Extended Data Fig. 2 Funnel plots for the movement distance data (left) and home range data (right).

The white area bordered by dashed lines represents the region of 95% pseudo confidence intervals where 95% of studies are expected to fall in the absence of bias and heterogeneity.

Extended Data Fig. 3 Definition of disturbance types.

Full reference details for study ID numbers can be found in the data file at Figshare.

Extended Data Fig. 4 Modelling results to estimate (a) the mean effects of disturbance on each movement type and (b) the mean positive and negative effects.

Posterior means (‘estimate’) and 95% credible intervals are presented for weighted and unweighted analyses of movement distance and home range size.

Extended Data Fig. 5 Modelling results for the effect of taxonomic group on animal movement responses to disturbance.

Posterior means (‘estimate’) and 95% credible intervals are presented for weighted and unweighted analyses of movement distance and home range size. – indicates that insufficient data was available to fit that particular model.

Extended Data Fig. 6 Modelling results for the effect of trophic level on animal movement responses to disturbance.

Posterior means (‘estimate’) and 95% credible intervals are presented for weighted and unweighted analyses of movement distance and home range size.

Extended Data Fig. 7 Effects of disturbance on animal movement distances and home range size according to trophic level (herbivore, omnivore or carnivore).

Positive effects represent increased movement in response to disturbance, and the opposite for negative effects. Symbols represent posterior means and coloured bands represent 95%, 80% and 50% credible intervals.

Extended Data Fig. 8 Modelling results for the effect of body mass on animal movement responses to disturbance.

Posterior means (‘estimate’) and 95% credible intervals are presented for weighted and unweighted analyses of movement distance and home range size. Sample size for each model is given in parentheses below taxonomic group. – indicates that insufficient data was available to fit that particular model.

Extended Data Fig. 9 Modelling results for the effect of (a) broad disturbance type (human activities or habitat modification) and (b) individual disturbance type on animal movement responses to disturbance.

Posterior means (‘estimate’) and 95% credible intervals are presented for weighted and unweighted analyses of movement distance and home range size. The analysis of individual disturbance types was restricted to combinations of taxonomic group and disturbance that had a minimum of 10 effect sizes from a minimum of three studies (combined). – indicates that insufficient data was available to fit that particular model.

Extended Data Fig. 10 Modelling results for the effect of geographic region on animal movement responses to disturbance.

Posterior means (‘estimate’) and 95% credible intervals are presented for unweighted analyses of movement distance and home range size.

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Doherty, T.S., Hays, G.C. & Driscoll, D.A. Human disturbance causes widespread disruption of animal movement. Nat Ecol Evol 5, 513–519 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing