The natural habitats of terrestrial mammals are being lost due to the expansion of human activities (Crooks and Sanjayan, 2006), and the resulting habitat fragmentation and loss of connectivity threaten mammal populations worldwide (Crooks et al., 2011; Crooks and Sanjayan, 2006; Tucker et al., 2018; Woodroffe and Ginsberg, 1998). Large carnivores are particularly susceptible to habitat loss and fragmentation given their large body sizes and extensive home ranges required to find sufficient food and shelter (Crooks, 2002; Thompson et al., 2021), their low densities, and their slow growth rates (Stier et al., 2016; Woodroffe, 2000). Moreover, the carnivores have a much higher chance of becoming extinct regionally in fragmented habitats due to human encounters or persecution compared to other species groups (Woodroffe and Ginsberg, 1998; Woodroffe, 2000; Csermak et al., 2022). Thus, it is essential to identify how patch sizes affects the movement patterns of carnivores to understand the species' adaptability and provide improved evidence to manage and conserve large carnivores in human-dominated landscapes.

Movement studies of large carnivores indicate how vital habitat connectivity is (Kramer-Schadt et al., 2011), as these species cross great distances and have extensive home ranges (Thompson et al., 2021). In continuous habitats, large carnivores are able to find food and reproduce. In contrast, they are often forced to explore anthropic areas for food in landscapes with small and scattered habitat patches (Dobson et al., 2006; Mortelliti and Boitani, 2008; Péron et al., 2017), which often leads to carnivore-human conflict and consequently, persecution by humans (Woodroffe, 2000; Csermak et al., 2022).

The jaguar (Panthera onca) is one of the large carnivores in the Neotropics that faces challenges due to habitat fragmentation and conversion of natural lands to agriculture, which also contribute to the reduction of its current distribution (de la Torre et al., 2018; Cullen et al., 2016; Haag et al., 2010; Zimmermann et al., 2021). Although jaguars prefer natural vegetation areas (Alvarenga et al., 2021; Conde et al., 2010; Morato et al., 2018a), they also use agriculture, peri-urban areas and roads (Alegre et al., 2023; Colchero et al., 2011; Cerqueira et al., 2021; Morato et al., 2018a) which increase their risk of mortality from road collisions and encounters with humans (Alegre et al., 2023; Cerqueira et al., 2021; Cullen et al., 2016). Additionally, jaguars' natural tendency to stay near forest edges (Dos Santos et al., 2022; Alegre et al., 2024) may lead to interactions with human activities, particularly in modified landscapes (Zimmermann et al., 2021). However, quantitative studies exploring jaguar movement patterns in landscapes modified by agriculture are scarce. By modeling jaguar movement responses to habitat modification, we can identify the landscape variables that impact the species and determine the degree of landscape modification they can tolerate (Bastille‐Rousseau and Wittemyer, 2021; Kays et al., 2015). This knowledge is essential for developing effective conservation strategies and management measures that promote coexistence between jaguars and human activities and ensure the long-term survival of this iconic species.

Here, we deepen the knowledge about the movement behavior in human-dominated landscapes where the jaguar lives, a top predator in Central and South America with a near-threatened category by the IUCN (Quigley et al., 2017). Our primary aim is to determine the movement patterns of jaguars concerning the structure of forest patches and agriculture areas, as well as the distances within and outside these structures. Additionally, we analyze how the proximity to drainage and roads influences these movement patterns. We considered movement patterns such as revisit (number of visits), time inside (time that the jaguar spent in one of these landscape structures), time since the last visit (of the structures), and speed as response variables within our models, translating the frequency of use of areas into more detailed return metrics. In addition, we determined at what depth jaguars enter agriculture and forest areas and how long they remain within them. We expected that, as a carnivore that depends on forested spaces for its survival (De Angelo et al., 2013, 2011), the jaguar tends to spend more time inside the forest and avoids spending long time in agriculture or near roads. Revisits are frequently carried out in small patches of forests and agriculture, used in the movement of the jaguar and search for resources and expansion of its home range (Conde et al., 2010; Thompson et al., 2021). Finally, we expected the jaguar movement pattern to respond negatively to anthropic structures (such as agriculture and roads).

Given the exploratory nature of this study and the limited peer-reviewed literature, we did not provide specific expectations for every variable combination. Instead, our approach was to generate a comprehensive understanding of jaguar movement patterns, which can inform more specific hypotheses in future research. If there are no specific expectations for some relationships, we aimed to present results regarding movement patterns relative to natural and anthropogenic variables, which provides clearer and more concise insights.

Our descriptive results indicated that the mean distance traveled by the jaguar inside agriculture was 534.5 meters, while in forest, it was higher at 986.5 meters. However, the mean distance the jaguar traveled outside agriculture was 3709.1 meters, whereas it was lower - 888.3 meters - outside forests. In contrast, some variables showed similar means; the speed in agriculture was 272 (m/4 h), whereas, in forest, it was slightly lower at 251(m/4 h). The revisitation rate was also similar, 7.02 in agriculture and 6.71 in forest. In terms of time inside agriculture, the mean was 11.6 h, while in forest, it was higher at 14.2 h. However, the mean time since the last visit in agriculture was 477.4 h, whereas in forest, it was 442.3 h (Table 2 and Figure A in Supplementary Material).

• -

  Forest

In our models, we observed that the size of the forest patch does not have a significant effect on the movement speed of jaguars (β = 0.016, p > 0.05). In contrast, we observed that increasing the size of forest patches significantly reduces the number of revisits (β = −0.076, p < 0.01), which may be due to larger forest patches providing more extensive resources, such as prey and shelter, reducing the need for frequent returns. Additionally, the time jaguars spend within these larger forest patches increases significantly (β = 0.166, p < 0.001; Figure A in Supplementary Material), suggesting that larger patches offer a more suitable or rewarding environment, encouraging jaguars to remain within them longer. We registered that increasing the size of forest patches significantly reduces the time since the last visit (β = −0.527, p < 0.001). This implies that larger patches are likely revisited sooner, possibly because they serve as key habitats that jaguars rely on regularly for foraging, resting, or other essential activities.

Distance from the forest does not significantly affect jaguar speed (β = −0.065, p > 0.05). However, increasing the distance from the forest significantly increases the number of revisits (β = 0.057, p < 0.01); in fact, we observed that all the revisits are taking place at the forest edges (Fig. 3). Distance from the forest does not significantly affect the time inside that the jaguars spend in these areas (β = −0.01, p > 0.05). However, increasing distance from forests significantly affects increasing time since the last visit (β = 0.232, p < 0.05; Fig. 4, Table B in Supplementary Material).

• -

  Agriculture

Fig. 3.

The number of visits across different distances of four land cover variables. Negative values indicate the distance within the variable, 0 is the edge, and positive values indicate the distance outside the variables. We provide figures for these variables separately in Figure C of the supplementary material.

(0.29MB).

Fig. 4.

Estimates of the four developed models for the Jaguar (Panthera onca) revisitation metrics (revisits, time inside, and time since last visit) and speed, with the different environmental variables (Values in Table B—Supplementary material).

(0.21MB).

In our models, we observed that the size of the agriculture patch tends to reduce the movement speed of jaguars, although this effect was not statistically significant (β = −0.062, p ≈ 0.08). However, larger agriculture patches significantly reduce the number of revisits (β = −0.096, p < 0.001). Additionally, we found that larger agriculture patches significantly reduce the time inside that jaguars spend in these areas (β = −0.303, p < 0.001; Figure B in Supplementary Material). This result suggests that jaguars tend to avoid areas with larger agriculture patches. Furthermore, increasing the size of the agriculture patches significantly increases the time since the last visit (β = 0.316, p < 0.001), indicating that jaguars are not likely to return to these areas sooner.

Regarding distance to agriculture, we found that this distance does not significantly affect the movement speed of jaguars (β = 0.027, p > 0.05). However, increasing the distance from agriculture significantly reduces the number of revisits (β = −0.049, p < 0.05), with the jaguar's most frequent revisits occurring at the edge of agricultural areas (Fig. 3). Increasing the distance from agriculture significantly reduces the time inside of these areas (β = −0.47, p < 0.001). Finally, we found that increasing the distance from the agriculture significantly decreases the time since the last visit (β = −0.15, p < 0.05), indicating that jaguars are more likely to return sooner to areas further away from agriculture (Fig. 4, Table B in Supplementary Material).

• -

  Drainage distance

In our models, we observed that increasing the distance to the drainage significantly positively affects the movement speed of jaguars (β = 0.246, p < 0.001). This result indicates that jaguars tend to move slowly near drainage areas. Furthermore, increasing the distance to the drainage significantly reduces the number of revisits (β = −0.27, p < 0.001). Still, like the other structures, we observed that revisits are evident at the edge of this structure (Fig. 3). This suggests that jaguars are less likely to revisit areas further away from the drainage.

Additionally, increasing distance to drainage significantly increased the time jaguars spent within these areas (β = 0.08, p < 0.001). This counterintuitive result could indicate that when jaguars venture further from drainage areas, they tend to stay longer. Finally, increasing the distance to drainage significantly reduces the time since the last visit (β = −0.91, p < 0.001), suggesting that jaguars return sooner to areas that are farther away from drainage (Fig. 4, Table B in Supplementary Material).

• -

  Road distance

Regarding the roads, we found that increasing the distance to the roads significantly negatively affects the movement speed of the jaguars (β = −0.465, p < 0.01). This suggests that jaguars move faster near roads. Furthermore, increasing the distance to roads significantly reduces the number of revisits (β = −0.202, p < 0.05), indicating that jaguars revisit the roadside frequently (Fig. 3).

Furthermore, increasing the distance to roads significantly reduces the time inside that jaguars spend in these areas (β = −2.519, p < 0.001). This suggests that jaguars do not prefer to spend time in areas away from roads. However, distance to roads does not significantly affect time since the last visit (β = −0.441, p > 0.05; Fig. 4, Table B in Supplementary Material).

Our results reveal that jaguars frequently visit the interface between forests, agricultural areas, roads, and drainages. The time between revisits and how long they stay inside these structures varies according to the size of the agriculture and forest patches, offering new insights into jaguar behavior in fragmented habitats. Furthermore, distance to these structures, such as forests, agriculture, drainage areas, and roads, significantly and variably affects jaguar behavior. Jaguar movement speed is differentially influenced by proximity to various landscape features: jaguars tend to move faster near roads and slower near drainage areas. At the same time, distance to forests and agricultural lands seems to have a less pronounced effect. These results contribute to a better understanding of what structural aspects within the habitat should be considered to manage natural habitats throughout the species distribution.

Jaguars are considered a crucial species for conserving natural habitats because they are closely associated with forested areas (Cullen Junior et al., 2013; Thornton et al., 2016; Morato et al., 2018a). Large carnivores require extensive areas with abundant natural prey (Cavalcanti and Gese, 2009; de la Torre et al., 2017; Thompson et al., 2021). Our results showed that the jaguar makes many revisits in small patches of forest and at the edge of the forest, which could suggest that it uses these areas for sustenance. This behavior might be driven by the variability in resource quality, where jaguars exploit lower-quality prey or easier hunting opportunities at these patches (Norris et al., 2010), as rodents (e.g., Agouti paca) are abundant in smaller forest patches (Norris et al., 2008). These smaller forest patches, also named “resource islands”, are created by trees and shrubs, which support higher biodiversity and greater ecological interactions compared to the surrounding grassland (Vetaas, 1992). Other larger prey, such as ungulates, frequently visit forest edges, influenced by landscape configuration (Lima et al., 2019; Norris et al., 2008; Ries et al., 2004), which can attract constant revisits by the jaguar. However, we also recorded that jaguars spend more time in larger forest patches than smaller ones, indicating that forests remain a safe area. This preference for larger patches likely reflects a strategy to optimize access to high-quality resources and avoid the risks associated with more anthropized areas (Cavalcanti and Gese, 2009; Conde et al., 2010; Morato et al., 2018a). Similar behavior is observed in pumas, which spend more time in larger patches and less in concentrated prey areas (Laundré, 2010). These findings support our expectation that jaguars will spend more time inside the forest, underscoring the critical importance of conserving large, contiguous forest patches. These areas not only serve as safe havens but are also essential for maintaining the ecological dynamics of jaguar populations, providing the necessary resources that are less available in anthropized environments.

We expected that jaguars would avoid agriculture; our results indicated a reduction in visits to larger agricultural areas and that it does not spend much time in them, which is an avoidant behavior in the species in proportion to the size of the anthropic structure. However, we recorded that most of the revisits were carried out on the edge of this structure, which might suggest that these edges provide easier access to opportunistic prey (Alegre et al., 2024; Esparza-Carlos et al., 2018) that may feed on crops or waste (Fitzgibbon, 1997; Dijak and Thompson, 2000; Bateman and Fleming, 2012; Ditmer et al., 2021). For example, the white-lipped peccaries, a vital prey of the jaguar, can use agriculture areas to feed, and due to this, conflict with humans have been registered in specific areas of Brazil (e.g. near the Emas National Park) where land use was mostly coverted for corn plantations (Lima et al., 2019; Csermak et al., 2022). This highlights how human structures and landscape modifications directly and indirectly alter predator-prey interactions (Patten et al., 2019; Berger, 2007; Murphy et al., 2021), sometimes leading to "empty forests" (Pires and Galetti, 2022), causing jaguars to leave the forest for search of resources.

Jaguars also frequently revisit natural drainage edges, which provide hydration and potential prey like capybaras and caimans (Azevedo and Murray, 2007; Azevedo and Verdade, 2012) in regions such as Pantanal wetlands. Furthermore, we found that they move slowly near the drainage, but the time spent near them is less, which delays the next visit. The drainage areas in many jaguar habitats are exploited and intervened by man (Ikeda-Castrillon et al., 2022), which could cause this transitory use.

Our results also indicated that jaguars frequently revisit the edges of the roads. Recent studies show that roads did not directly affect jaguar space use (Cerqueira et al., 2021). However, adverse effects of roads on jaguars' space use occurred indirectly in the same study because paved roads were associated with a low proportion of forest. Jaguars may also use the road positively as a resource for displacement (Alegre et al., 2023). Similar studies with large carnivores demonstrated road use according to traffic intensity (Kautz et al., 2022). However, roads, especially paved ones, pose significant risks due to high probability of wildlife collisions, leading to high mortality (Espinosa et al., 2018). These results underscore the importance of considering landscape structure in conservation strategies and highlight the complex interactions between jaguars and their fragmented habitats.

On the other hand, our study has some limitations that must be considered. One limitation is the heterogeneity in GPS data collection methods for different jaguars, which can introduce variability in the precision and resolution of the movement data. Additionally, our analysis did not account for the sex of the jaguars because our data did not include a uniform distribution of males and females across different environmental types. Only males were present in some areas, while in others, only females or a mix of both, making it challenging to analyze sex-specific differences in movement patterns. Another limitation is the reliance on auxiliary road data from OpenStreetMap (OSM), as we were unable to verify the accuracy of all mapped roads due to resource constraints. Future studies should consider validating these datasets, as this could enhance the precision of movement analysis. Finally, future research should develop movement patterns that differentiate between sexes, as this could provide more detailed information on how these landscape structures affect the jaguar population. Addressing these limitations in future research will enhance our understanding of the jaguar movement and contribute to more effective conservation strategies.