First records on habitat use of semi-feral cattle in southern forests: topography reveals more than vegetation

8/20/2019 First records on habitat use of semi-feral cattle in southern forests: topography reveals more than vegetation

1/34

NICOLÁS SEOANEHabitat use of semi-feral cattle1

2

First records on habitat use of semi-feral cattle in southern forests: topography reveals3

more than vegetation 4

5

Nicolás Seoane6

7

Laboratorio Ecotono, Universidad Nacional del Comahue, Quintral 1250, (8400) Bariloche,8

Argentina. (Email: [email protected]) Tel. +54 (02944) 4233749

10

11

12

13

14

15

16

17

18

19

20

21

22

23


2/34

NICOLÁS SEOANE ABSTRACT 24

25

While the presence of cattle in forests is quite common, how they use the habitat is often26

overlooked. When this is examined, most studies focus on measurements of the vegetation27

variables influencing habitat selection. This current study provides a suitable model to study28

habitat use by livestock in forested areas. The model was applied to data from semi-feral cattle in29

order to obtain the first description of their habitat use in southern forests. Furthermore, the30

model accounted for individual variability, and hinted at population patterns of habitat use. The31

positions of four individual animals with GPS collars were recorded during four months in a32

Nothofagus (southern beech) forest in Patagonia (Argentina). By projecting these GPS location33

data into a geographical information system (GIS), a resource selection probability function34

(RSPF) that considers topographic and vegetation variables was built. The habitat use of semi-35

feral cattle in southern beech forests showed a large interindividual variability, but also some36

similar characteristics which enable a proper description of patterns of habitat use. It was found37

that habitat selection by cattle was mainly affected by topographic variables such as altitude and38

the combination of slope and aspect. In both cases the variables were selected below average39

relative to availability, suggesting a preferred habitat range. Livestock also tended to avoid areas40

of closed shrublands and showed a slight preference for pastures (meadows). This result suggests41

that cattle are showing adaptations to the major features of these mountainous forests due to42

ferality. Furthermore, these results are the basis for management applications such as predictive43

maps of use by semi-feral livestock.44

45

46


3/34

NICOLÁS SEOANEKEYWORDS 47

48

Nothofagus forests; beef cattle; semi feral cattle; resource selection probability functions (RSPF);49

GPS tracking; habitat distribution modeling. 50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69


4/34

NICOLÁS SEOANEINTRODUCTION 70

71

Cattle ( Bos taurus) are an important species for forest dynamics globally, and the originally72

domesticated animals have transgressed to become semi-feral to feral again in many areas73

(Decker et al. 2015, Berteaux and Micol 1992, Hernandez et al. 1999, Baker 1990, Atkinson74

2001, Lazo 1994). Livestock can affect the diversity and structure of plant communities75

(Coughenour 1991, Cingolani et al. 2008) mainly through diet selection (Rook et al. 2004), but76

also due to indirect actions such as changing the distribution of nutrients and compacting the soil77

by trampling (Milchunas et al. 1998, Cingolani et al. 2008). Thus, it is possibly one of the main78

ecosystem engineers for these environments, as has already been demonstrated in grasslands79

(Jones et al. 1997). Although the impacts of cattle are often studied, their habitat use, which80

would allow a comprehensive understanding of the livestock-forest ecological system, has81

generally been overlooked.82

Southern beech forests (i.e., Nothofagus forests) in Patagonia have been used to raise83

cattle since the late 19th century. The traditional management involves extensive cattle ranching84

in the forest. But occasionally and for different reasons, some animals escape the herd,85

establishing feral populations (Atkinson 2001, Veblen et al. 1996). Although feral cattle can be86

found in many areas of the Nothofagus forests in Patagonia (Novillo and Ojeda 2008) and New87

Zealand (Howard 1966, Veblen and Stewart 1982), no studies have been performed on their88

behavior or habitat selection in southern forests. Additionally, many levels of ferality already89

exist depending on contact with human settlements, hunting pressure, isolation time, etc.90

The proportion of time an animal spends in a location to meet their biological demands91

defines its habitat use (Beyer et al. 2010). As resources are often heterogeneously distributed,92


5/34

NICOLÁS SEOANEindividuals need to explore multiple environments to satisfy their needs (Law and Dickman93

1998). Therefore, the selection of habitats is a key feature of behavior and population dynamics94

(Morris 1987). This is especially relevant for large herbivores, because their distribution on the95

landscape can decisively influence the productivity and biodiversity of fields and pastures96

(Bailey and Provenza 2008).97

There are multiple relationships between livestock and their environment but one of the98

major ones is the change in the heterogeneity of vegetation produced by foraging (Adler et al.99

2001). In the case of forests, despite having high heterogeneity and frequent cattle presence,100

there is no systematic review of the resulting ecological processes due to the introduction of this101

herbivore worldwide. Furthermore, it is common to have little or no information about the102

activity patterns of semi-wild cattle in general, which makes their management even more103

difficult. Some studies that have been conducted in different forests account for diet selection104

along seasons (Marquardt et al. 2010) or in relation to anthropogenic changes, usually due to105

forest management practices (Roath and Krueger 1982, Kaufmann et al. 2013). In these studies,106

vegetation type and topography were the most common variables to explain habitat selection by107

cattle in forests.108

A variety of technological and statistical tools can be applied to study habitat use. Among109

the first, radiotelemetry and GPS devices allow for accurate and frequent data collection on110

animal location, which facilitates the development of mechanistic models of habitat use and111

movement (Beyer et al. 2010). Therefore, their use is widespread (Johnson et al. 2004, Ungar et112

al. 2005, Cagnacci et al. 2010, Peinetti et al. 2011). In addition to these data collection methods,113

habitat use can be modeled by resource selection functions (RSF) and resource selection114

probability functions (RSPF) that estimate the probability that a given environment has been115


6/34

NICOLÁS SEOANEused by an animal (Boyce and McDonald 1999, Beyer et al. 2010). Thus, the RSFs provide a116

framework for developing an ecological theory of habitat use and allow linking landscape117

ecology and population biology (Boyce and McDonald 1999, Boyce 2006). One way to build a118

RSPF with GPS data is to model the intensity of use in sampling units, relating the relative119

frequency of records per unit to the environmental features of such sampling units (Nielson and120

Sawyer 2013).121

The objective of this study was to evaluate the habitat use of semi-feral cattle in southern122

beech forests as no current data or a common methodology exists for this purpose. Furthermore,123

the use of vegetation variables is usual in studies on habitat use by cattle, but the role of124

topography could be of importance especially in mountainous areas, yet is often overlooked.125

Specifically, this study address the following questions: (i) What is the habitat use of semi-feral126

cattle in Nothofagus forests?; and (ii) Are the vegetation variables of the landscape more127

important than topography for habitat use by cattle?. These research questions were addressed by128

using a utilization distribution approach and the construction of a resource selection probability129

function (RSPF) which included topographic and vegetation variables.130

131

132

METHODS 133

134

- Study Area 135

The study was carried out in the Llodconto Valley, Northwest Patagonia (Rio Negro, Argentina).136

The valley is within the Nahuel Huapí National Park, between 41º21' and 41º27' south latitude,137

and 71º31' and 71º41' west longitude, at an elevation ranging from 920 to 2000 m.a.s.l.. It is a138


7/34

NICOLÁS SEOANEmountainous area with cold-temperate to moist-cold climate with prevailing westerly winds. The139

summers are dry and winters rainy, with a mean monthly temperature ranging from 2.4ºC in July140

to 12.9ºC in January. The area is part of the Andean-Patagonian forest and includes a wide141

variety of habitats such as forests, shrublands, meadows, burned forest, riparian and high142

mountain areas. The dominant species is Nothofagus antarctica which is one of the main143

components of both forests and shrublands. Other species such as the Nothofagus pumilio tree,144

the Schinus patagonicus shrub, and the Chusquea coleou bamboo are also abundant. An145

uncertain number of free-ranging cattle inhabit the valley and are traditionally and minimally146

managed by local people only during the summer months.147

148

- Data collection and explanatory variables149

In order to capture the animals, an enclosure located at the center of the valley was used to herd150

the animals into. Four female adults caught in 2012 and one more caught in 2013 were fitted151

with custom-made GPS collars (GPS Module ZX4120 and Amicus GPS Shiled Antenna, ©152

Crownhill Associates Ltd.). These GPS collars store data on board and recapture was necessary153

to recover the data. Each collar was set to record location every ten minutes and these records154

were later screened by hour in order to standardize the records and eliminate the unacquired data.155

All capture and handling methods used in this study met the standards of animal welfare156

practices recommended for scientific research (Rollin and Kessel 1998, Gannon and Sikes 2007).157

All research was approved by the proper authorities of the protected area.158

Topographic mapping and spatial statistical analysis were conducted using Quantum159

GIS™ software (QGIS™) version 1.8.0 using a digital elevation model (ASTER GDEM) as the160

data source. Animal locations were overlayed on the digital elevation model and a minimum161


8/34

NICOLÁS SEOANEconvex polygon (MCP) with a 200 meter buffer was used around the complete set of GPS162

positions to define our study area. One thousand sampling units of 100 meters radii each (i.e.,163

sampling unit area= 0.03km2) were allocated randomly inside this area.164

Habitat types were assigned for each sampling unit by means of a visual classification on165

Google Earth (© Google Inc.), identifying the following vegetation types: forests, shrubland,166

burned areas, meadows, open areas (vegetation cover 0.6) were not used together in the same model. Squared177

elevation and slope were included in the models for the purpose of evaluating preferred ranges of178

use by animals. In order to account for a strong regional climatic pattern, the interaction between179

slope and NWness was also considered since steep and northwest exposed slopes are generally180

arid, with short vegetation and low coverage.181

182

- Data analysis 183

There are several ways to model habitat use. The most widely used include the GLMs and related184


9/34

NICOLÁS SEOANEmodels (Guisan and Zimmermann 2000), but multivariate approaches (Clark et al. 1993, Burke185

et al. 2013), GIS-based habitat suitability models (Osborne et al. 2001, Store and Jokimäki186

2003), individual based models (Railsback and Harvey 2002), artificial neural networks187

approaches (Özesmi and Özesmi 1999), and maximum entropy modeling (Baldwin 2009) also188

exist. Here I applied a utilization distribution approach described by Nielson and Sawyer (2013)189

to develop the RSPF because it accounts for intensity of use among habitats and is unbiased in190

the face of temporally correlated animal location data.191

The utilization distribution approach allows the modeling of habitat use by using the192

relative frequency of positions within each sampling unit as an empirical estimator of the use193

distribution (Millspaugh et al. 2006). This relative frequency of positions was used as a response194

variable in a generalized linear model (GLM) with negative binomial distribution that accounts195

for over-dispersed count data. In order to relate the observed counts in a sampling unit with the196

total number of positions recorded for each animal, an offset term was included, thus allowing for197

estimating the probability of use by individual ( [i]/total , references below).198

199

Thus, the RSPF was set as follows: 200

θ , μnomial NegativeBi y ii ~ 201

...orest.Wness.uggedness.lope.levation.lnln 54321 iiiii0i f β N β r β s β e β + β +total = μ 202

203

where:204

yi = number of observations at the i-th sampling unit; = relative frequency in each sampling205

unit by animal; = overdispersion parameter; total = number of recorded positions in the whole206

study area by animal; i = sampling unit with associated habitat covariates.207


10/34

NICOLÁS SEOANESeveral models exploring different combinations of predictor variables were fitted for208

each animal and compared using the Akaike information criterion (AIC). The overdispersion209

parameter of the negative binomial distribution was estimated simultaneously with the models210

using the glm.nb function in R (Venables and Ripley 2002, R Core Team 2013 version 3.0.2). A211

set of ecologically plausible models was chosen to work with all animals based on a stepwise212

regression. The package glmulti (Calcagno and Mazancourt 2010) was used as a tool for quickly213

identifying the best models for each animal. The deviance (Guisan et al. 2000) was used as a214

goodness of fit indicator.215

A population-level model was then developed considering each animal as an216

experimental unit, which reflects the individual nature of resource selection (Marzluff et al.217

2004, Millspaugh et al. 2006). A single set of covariates was selected considering the comparison218

among models and the biological relevance of the variables. The estimated coefficients, variance219

and confidence intervals were calculated for each parameter of this general proposed model.220

The coefficients of the population model were calculated according to: 221

̂ ̄βk =1

n ∑ β̂kj

222

where ̂ ̄βkj is the estimated coefficient k for each individual j ( j=1,2,3,4) and n is the total number223

of animals (n=4).224

The variance of the estimated coefficients for the population model was computed as:225

Var ( ̂ ̄βk )= 1n− 1∑ j= 1

n

( ̂βij− ̂̄βij)2

226

227

228

229


11/34

NICOLÁS SEOANERESULTS 230

231

A total of 2575 position records (an 87% fix rate for the GPS collars) was obtained from 4 of the232

animals. One of the collared animals could not be relocated during the study and therefore was233

not recaptured.234

Based on the fitted models for each animal, the general pattern observed that topographic235

variables were always more important than vegetation. Given the full additive model (not shown)236

as an example, the only meaningful variables for all animals were elevation, aspect and the237

interaction between slope and NWness. In all cases, topographic variables had negative238

coefficients because the animals preferred elevations below the available mean in the landscape239

and avoided hillsides with northwestern exposure and steep slopes. The vegetation variables240

showed different tendencies for each individual and in most cases were not significant (p > 0.05).241

The burned areas and shrubs had a tendency of being avoided by cows 1 and 2 but were242

preferred by cow 3, which also preferred wet meadows. Forest habitat had low coefficients for all243

the animals with values near zero, which showed no tendency to be selected, either in favor or244

against.245

By means of a step-by-step selection procedure over all the models, five plausible246

ecological models were selected for each animal in order to use them as a RSPF for semi-wild247

cattle (Table 1). These models were among the best ten for each animal. It can be seen that all248

models have as variables elevation, some aspect indicator, its interaction with slope, and in some249

cases the vegetation types with greater selection: shrubs and meadows. The coefficients of these250

five models for all the animals are summarized in Appendix 1.251

The most parsimonious models for each animal (Table 2) were significantly different252


12/34

NICOLÁS SEOANEfrom the null model (p


13/34

NICOLÁS SEOANEDISCUSSION 276

277

Using GPS logs and an approach that takes into account the intensity of use, a RSPF that278

accounts for topography and vegetation was built for semi-wild cattle in the Andean-Patagonian279

forest. This study shows that habitat use by livestock in these forests has a large inter-individual280

variation but also common features that would allow a description of a pattern of use.281

Topographic variables affected the probability of habitat use by cattle to a greater extent than282

vegetation types. In the general proposed model, for instance, a change in a single standard283

deviation unit of elevation (~ 300 meters) resulted in a threefold increased in the probability of284

use (Table 3) and the squared slope has a similar role in some of the alternative models (Table 1).285

This suggests that terrain features have a leading role in defining the use of space by livestock,286

perhaps as an adaptation due to the long history of use and because topography is important in287

these mountainous forests.288

Some common features such as the preferred elevation range and the tendency to prefer289

meadows suggests that it would be possible to describe a population-level home-range for a more290

comprehensive description of habitat use. In this regard, the squared elevation would play an291

important role in defining a preferred elevation range, although there seems to be a high292

variability among individuals. This variability could be modeled by a RSPF with hierarchical293

formulation. Cattle tend to slightly avoid shrubland (Table 3). This may be due to a lower294

proportion of trails in this type of thick vegetation, in contrast to other more open areas or those295

that offer better protection from snow, such as forest. Moreover, these paths are unusable in the296

winter. In this scenario, the availability and network of trails would play an important role in297

defining the population-level home-range of cattle, and therefore, in determining habitat use. At298


14/34

NICOLÁS SEOANEthe same time, use by livestock has created the paths over time, so there is a reciprocal process299

that shows a type of population memory process.300

The response variable used in this study proved to be appropriate and useful as it301

describes the habitat use of cattle in probabilistic terms. However, the experimental design has302

limitations that must be made explicit. This includes the low sample size of animals. However,303

the different trajectories and spaces used by the sampled individuals suggest that the study was304

able to capture considerable variability in resource use by cattle (Figures 1 and 2) and even305

studies with one GPS-collared animal have accomplished some general features in the behavior306

of a herd (Moritz et al. 2010). Another limitation is the temporal span of the data collection. As307

the data collection period occurred between the months of April and July, it can be assumed that308

this covers a typically cold season (corresponding to the austral autumn and winter). Therefore,309

all models and conclusions of this work should be most reflective of that season.310

A disadvantage of using logistic regressions such as RSPFs is that the model coefficients311

decrease as the availability of a resource increases, and may even change the sign (Beyer et al.312

2010). This could explain some of the inter-individual variation observed in the case of meadow313

use (Figure 2), with animals 1 and 2 both negatively selecting this vegetation, and moving into314

an area where the abundance of meadows is higher in relation to the areas where individuals 3315

and 4 moved (unreported movement data). A similar case is that of the forest variable, which was316

not significant in any model (Table 1) and yet it is a very abundant vegetation type and widely317

used by cattle. To overcome this limitation it would be necessary to develop models that also318

account for animal movement, as habitat selection and movement are not independent processes.319

Methods for estimating movement patterns and habitat selection simultaneously have recently320

been developed (Moorcroft and Barnett 2008, Smouse et al. 2010).321


15/34

NICOLÁS SEOANEComparing the results of this study with others conducted on cattle in forests we find322

interesting relationships in the applied variables and scales. Studies on seasonal diet selection323

(Marquardt et al. 2010) in Bolivian forests for example, show that different functional groups of324

plants are preferred in different seasons. It would be possible to infer species selected by325

livestock throughout the year in the Andean-patagonian forest if direct observations of consumed326

plants was added to this study. Roath and Krueger (1982) also used topographic and vegetation327

variables to study the behavior of cattle in a forested mountainous environment, finding that the328

type of vegetation and water availability determined the degree of use by livestock. These329

apparently opposite results are consistent if we consider the differences between the study areas.330

While Roath and Krueger (1982) worked in a dry, arid vegetation and elevational range of 300331

meters, Llodconto Valley is an area with high water availability, abundant streams, and altitudes332

ranging from 800 to 1900 masl. Kaufmann et al. (2013) found that in silvopastoral systems cattle333

prefer places with forest cover, avoiding places with high slope and low biomass, particularly334

logged sites. A comparison with this study would make an analogy between logged and burned335

areas, both without forest cover. Nevertheless, with the available data we cannot make inferences336

about the type of selection on burned areas so far (Tables 1- 3). It may be noted, however, that337

other studies in Patagonia (Blackhall et al. 2008, De Paz and Raffaele 2013) found a significant338

influence of livestock on vegetation in burned environments, making it clear that cattle forage in339

these areas. Interestingly, Johnson et al. (2004) conducted a study on resource selection by340

caribou at two spatial scales, finding that topographic variables such as elevation and slope are341

more important at a landscape scale, while vegetation variables were more important at the patch342

scale. Something similar could be occurring in this present study, which is closer to the343

landscape scale.344


16/34

NICOLÁS SEOANEAs mentioned, cattle can increase landscape heterogeneity, especially when the spatial345

pattern of vegetation is already heterogeneous (Adler et al. 2001) as in the case of the currently346

studied forests. This has implications for forests dynamics because although there are studies on347

cattle inhabiting forest (Bartolomé et al. 2011, Kauffmann et al. 2013, etc.), there still remains a348

lack of knowledge regarding how forest dynamics are affected by the distribution of wild cattle349

worldwide. Recently, a study conducted in Patagonia developed a model that postulates distance350

to meadows as an estimator of forest use by livestock (Quinteros et al. 2012). While meadows351

are of crucial importance to the Andean-Patagonian forests and Quinteros et al. (2012) make a352

practical contribution in relation to forest management, their model has a spatial limitation, being353

restricted to meadows surroundings. Thus, it is not possible to make inferences at a landscape354

scale and the model fails to consider other relevant variables such as topography, which proved355

to have greater importance in resource selection by cattle (Tables 2-3, Figure 2). Nonetheless, the356

approaches are consistent, because as the distance to meadows increases, the intensity of use by357

livestock decreases (Quinteros et al. 2012), although altitude is also higher, which is predicted by358

the model in the current study.359

There remains a need for tools that enable animal management in agreement with360

conservation management practices of native forests. In Patagonia, for instance, extensive361

livestock farming has been identified as an activity with potential to be carried out in the forest.362

Yet, there have been only some pioneer experiences of silvopastoral handling (Fertig and Guitart363

2006), and current livestock handling is still scarce (Ormaechea et al. 2009). The main364

limitations for management worldwide are a lack of understanding of animal foraging decisions365

and the spatial scale of diet selection (Rook et al. 2004). Although effective management366

depends heavily upon knowledge about the interaction of the animal with its environment, it is367


17/34

NICOLÁS SEOANEcommon to have poor information on the activity patterns of semi-wild cattle (Moyo et al. 2012).368

This current study provides a direct contribution to this knowledge-gap, and direction for future369

research in these systems. As an example, a practical application of this work is the production of370

maps showing intensity of land use by livestock. Similar maps have been developed for other371

regions using equivalent methodologies (Nielsen et al. 2003, Sawyer et al. 2009), and therefore372

offer a management and research tool within protected areas or productive forests.373

In summary, the present study demonstrated that the habitat use by semi-feral cattle in374

southern beech forests has a large inter-individual variability but also similar characteristics375

which allow the description of a pattern of use. Specifically, the results indicate that vegetation376

variables are no more important than topography features. Conversely, habitat selection by cattle377

is mainly affected by topographic variables such as altitude and the combination of slope and378

aspect. These results suggest an adaptation by free-ranging cattle to the major features of the379

landscape of these mountainous forests maybe due to ferality.380

381

382

383

ACKNOWLEDGEMENTS 384

385

Funding for this research was provided by a PhD grant from the CONICET (Argentina). The386

author thanks to Juan Manuel Morales for helpful comments on the manuscript.387

388

389

390


18/34

NICOLÁS SEOANELITERATURE CITED 391

392

Adler, P., D. Raff, and W. Lauenroth. 2001. The effect of grazing on the spatial heterogeneity of393

vegetation. Oecologia 128(4):465–479. doi:10.1007/s004420100737394

Atkinson, I. E. 2001. Introduced mammals and models for restoration. Biological Conservation 395

99(1):81–96. doi:10.1016/S0006-32070000189-0396

Bailey, D. W., and F. D. Provenza. 2008. Mechanisms determining large-herbivore distribution.397

In Resource Ecology pp. 7–28. Springer.398

Baker, S. J. 1990. Escaped exotic mammals in Britain. Mammal Review 20(2-3):75–96.399

Baldwin, R. A. 2009. Use of maximum entropy modeling in wildlife research. Entropy 11(4): 400

854–866. doi:10.3390/e11040854401

Bartolomé, J., J. Plaixats, J. Piedrafita, M. Fina, E. Adrobau, A. Aixàs, … L. Polo. 2011.402

Foraging Behavior of Alberes Cattle in a Mediterranean Forest Ecosystem. Rangeland403

Ecology and Management 64( 3):319–324. doi:10.2111/REM-D-09-00160.1404

Berteaux, D., and T. Micol. 1992. Population studies and reproduction of the feral cattle Bos405

taurus of Amsterdam Island, Indian Ocean. Journal of Zoology 228(2):265–276.406

Beyer, H. L., D. T. Haydon, J. M. Morales, J. L. Frair, M. Hebblewhite, M. Mitchell, and J.407

Matthiopoulos. 2010. The interpretation of habitat preference metrics under use-availability408

designs. Philosophical Transactions of the Royal Society of London. Series B, Biological409

Sciences 365(1550):2245–2254. doi:10.1098/rstb.2010.0083410

Blackhall, M., E. Raffaele, and T. T. Veblen. 2008. Cattle affect early post-fire regeneration in a411


19/34

NICOLÁS SEOANE Nothofagus dombeyi-Austrocedrus chilensis mixed forest in northern Patagonia, Argentina.412

Biological Conservation 141(9):2251–2261. doi:10.1016/j.biocon.2008.06.016413

Boyce, M. S. 2006. Scale for resource selection functions. Diversity and Distributions 414

12(3):269–276. doi:10.1111/j.1366-9516.2006.00243.x415

Boyce, M. S., and L. L. McDonald. 1999. Relating populations to habitats using resource416

selection functions. Trends in Ecology and Evolution 14(7):268–272. doi:10.1016/S0169-417

53479901593-1418

Burke, B. J., W. T. Peterson, B. R. Beckman, C. Morgan, E. A. Daly, and M. Litz. 2013.419

Multivariate models of adult Pacific salmon returns. PloS One 8(1): e54134.420

doi:10.1371/journal.pone.0054134421

Cagnacci, F., L. Boitani, R. A. Powell, and M. S. Boyce. 2010. Animal ecology meets GPS-based422

radiotelemetry: a perfect storm of opportunities and challenges. Philosophical Transactions423

of the Royal Society of London. Series B, Biological Sciences 365(1550):2157–62.424

doi:10.1098/rstb.2010.0107425

Calcagno, V., and C. De Mazancourt. 2010. glmulti : An R Package for Easy Automated Model426

Selection with Generalized Linear Models. Journal of Statistical Software 34(12):1–29.427

Cingolani, A. M., I. Noy-Meir, D. D. Renison, and M. Cabido. 2008. La ganadería extensiva: es428

compatible con la conservación de la biodiversidad y de los suelos? Ecología Austral 429

18(3):253–271.430

Clark, J. D., J. E. Dunn, and K. G. Smith. 1993. A Multivariate Model of Female Black Bear431

Habitat Use for a Geographic Information System. The Journal of Wildlife Management 432

57( 3):519–526.433


20/34

NICOLÁS SEOANEDe Paz, M., and E. Raffaele. 2013. Cattle change plant reproductive phenology, promoting434

community changes in a post-fire nothofagus forest in northern Patagonia, Argentina.435

Journal of Plant Ecology 6(6):459–467. doi:10.1093/jpe/rtt004436

Decker, J. E., J. F. Taylor, M. A. Cronin, L. J. Alexander, J. Kantanen, A. Millbrooke, … M. D.437

MacNeil. 2015. Origins of cattle on Chirikof Island, Alaska. bioRxiv p. 014415. Cold438

Spring Harbor Labs Journals. doi:10.1101/014415439

Fertig, M., and E. Guitart. 2006. Cuaderno de Campo para Vacunos p. 112. Esquel: INTA EEA440

Esquel.441

Gannon, W. L., and R. S. Sikes. 2007. Guidelines of the American Society of Mammalogists for442

the Use of Wild Mammals in Research. Journal of Mammalogy 88(3):809–823.443

Guisan, A. and N. E. Zimmermann. 2000. Predictive habitat distribution models in ecology.444

Ecological Modelling 135(2):147–186. doi:10.1016/S0304-38000000354-9.445

Hernandez, L., H. Barral, G. Halffter, and S. S. Colon. 1999. A note on the behavior of feral446

cattle in the Chihuahuan Desert of Mexico. Applied Animal Behaviour Science 63(4):259– 447

267. doi:10.1016/S0168-15919900015-5448

Howard, W. E. 1966. Ecological changes in New Zealand due to introduced mammals. Towards449

a New Relationship of Man and Nature in Temperate Lands, 219-240, UICN.450

Johnson, C. J., D. R. Seip, and M. S. Boyce. 2004. A quantitative approach to conservation451

planning: Using resource selection functions to identify important habitats for mountain452

caribou. Journal of Applied Ecology 4(1):238–251.453

Jones, C. G., J. H. Lawton, and M. Shachak. 1997. Positive and negative effects of organisms as454


21/34

NICOLÁS SEOANE physical ecosystem engineers. Ecology 78(7):1946–1957.455

Kaufmann, J., E. W. Bork, P. V. Blenis, and M. J. Alexander. 2013. Cattle habitat selection and456

associated habitat characteristics under free-range grazing within heterogeneous Montane457

rangelands of Alberta. Applied Animal Behaviour Science 146(1):1–10.458

doi:10.1016/j.applanim.2013.03.014459

Law, B. S., and C. R. Dickman. 1998. The use of habitat mosaics by terrestrial vertebrate fauna:460

Implications for conservation and management. Biodiversity and Conservation 7(3):323– 461

333. doi:10.1023/A:1008877611726462

Lazo, A. 1994. Social segregation and the maintenance of social stability in a feral cattle463

population. Animal Behaviour 48(5):1133-1141. doi:10.1006/anbe.1994.1346464

Marquardt, S., H. Alzérreca, E. Hillmann, F. D. Encinas, A. C. Mayer, and M. Kreuzer. 2010.465

Seasonal variation in activity and habitat use of free-ranging cattle in southern Bolivian466

subtropical mountain forests. Cuban Journal of Agricultural Science 44(4).467

Marzluff, J. M., J. J. Millspaugh, P. Hurvitz, and M. S. Handcock. 2004. Relating resources to a468

probabilistic measure of space use: Forest fragments and Steller’s Jays. Ecology 85( 5):469

1411–1427. doi:10.1890/03-0114470

Milchunas, D. G., W. K. Lauenroth, and I. C. Burke. 1998. Livestock grazing: animal and plant471

biodiversity of shortgrass steppe and the relationship to ecosystem function. Oikos:65–74.472

Millspaugh, J., R. M. Nielson, L. McDonald, J. M. Marzluff, R. A. Gitzen, C. D. Rittenhouse, …473

S. L. Sheriff. 2006. Analysis of Resource Selection Using Utilization Distributions. Journal474

of Wildlife Management 70(2):384–395.475


22/34

NICOLÁS SEOANEMoorcroft, P. R., and A. Barnett. 2008. Mechanistic home range models and resource selection476

analysis: a reconciliation and unification. Ecology 89(4):1112–1119. doi:10.1890/06-1985.1477

Moritz, M., E. Soma, P. Scholte, N. Xiao, L. Taylor, T. Juran, and S. Kari. 2010. An Integrated478

Approach to Modeling Grazing Pressure in Pastoral Systems: The Case of the Logone479

Floodplain Cameroon. Human Ecology 38(6):775–789. doi:10.1007/s10745-010-9361-z480

Morris, D. W. 1987. Ecological Scale and Habitat Use. Ecology 68(2):362-369.481

doi:10.2307/1939267482

Moyo, B., S. Dube, M. Lesoli, and P. Masika. 2012. Behavioural patterns of cattle in the483

communal areas of the Eastern Cape Province, South Africa. African Journal of484

Agricultural Research 7(18):2824–2834. doi:10.5897/AJAR11.930485

Nielsen, S. E., M. S. Boyce, G. B. Stenhouse, and R. H. M. Munro. 2003. Development and486

testing of phenologically driven grizzly bear habitat models. Ecoscience 10(1):1-10.487

Nielson, R. M., and H. Sawyer. 2013. Estimating resource selection with count data. Ecology and488

Evolution 3(7):2233–2240. doi:10.1002/ece3.617489

Novillo, A., and R. A. Ojeda. 2008. The exotic mammals of Argentina. Biological Invasions 490

10(8):1333–1344. doi:10.1007/s10530-007-9208-8491

Ormaechea, S. G., P. L. Peri, R. Molina, and J. P. Mayo. 2009. Current situation and livestock492

management on landholdings in south Patagonia with ñire Nothofagus antarctica forest. In:493

Actas 1er Congreso Nacional de Sistemas Silvopastoriles, Posadas, Misiones, Argentina.494

Osborne, P. E., J. C. Alonso, and R. G. Bryant. 2001. Modelling landscape-scale habitat use using495

GIS and remote sensing: a case study with great bustards. Journal of Applied Ecology496


23/34

NICOLÁS SEOANE38(2):458–471. doi:10.1046/j.1365-2664.2001.00604.x497

Özesmi, S. L., and U. Özesmi. 1999. An artificial neural network approach to spatial habitat498

modelling with interspecific interaction. Ecological Modelling 116(1):15–31.499

doi:10.1016/S0304-38009800149-5500

Peinetti, H. R., E. L. Fredrickson, D. P. Peters, A. F. Cibils, J. O. Roacho-Estrada, and A. S.501

Laliberte. 2011. Foraging behavior of heritage versus recently introduced herbivores on502

desert landscapes of the American Southwest. Ecosphere 2(5): art57. doi:10.1890/ES11-503

00021.1504

Quinteros, C. P., P. M. Bernal, M. E. Gobbi, and J. O. Bava. 2012. Distance to flood meadows as505

a predictor of use of Nothofagus pumilio forest by livestock and resulting impact in506

Patagonia, Argentina. Agroforestry Systems 84(2):261-272. doi:10.1007/s10457-011-9461-9507

R Core Team. 2013. R: A Language and Environment for Statistical Computing. Vienna, Austria.508

Retrieved from http://www.r-project.org/509

Railsback, S. F., and B. C. Harvey. 2002. Analysis of habitat-selection rules using an individual-510

based model. Ecology 83(7):1817–1830.511

Roath, L. R., and W. C. Krueger. 1982. Cattle grazing and behavior on a forested range. Journal512

of Range Management 35(3):332–338.513

Rollin, B. E., and M. L. Kessel. 1998. Guidelines for the treatment of animals in behavioural514

research and teaching. Anim. Behav 55:251–257.515

Rook, A. J., B. Dumont, J. Isselstein, K. Osoro, M. F. WallisDeVries, G. Parente, and J. Mills.516

2004. Matching type of livestock to desired biodiversity outcomes in pastures – a review.517


24/34

NICOLÁS SEOANE Biological Conservation 119(2):137–150. doi:10.1016/j.biocon.2003.11.010518

Sawyer, H., M. J. Kauffman, and R. M. Nielson. 2009. Influence of Well Pad Activity on Winter519

Habitat Selection Patterns of Mule Deer. Journal of Wildlife Management 73(7):1052–1061.520

Smouse, P. E., S. Focardi, P. R. Moorcroft, J. G. Kie, J. D. Forester, and J. M. Morales. 2010.521

Stochastic modelling of animal movement. Philosophical Transactions of the Royal Society522

of London. Series B, Biological Sciences, 365(1550):2201–11. doi:10.1098/rstb.2010.0078523

Store, R., and J. Jokimäki. 2003. A GIS-based multi-scale approach to habitat suitability524

modeling. Ecological Modelling 169(1):1-15. doi:10.1016/S0304-38000300203-5525

Ungar, E. D., Z. Henkin, M. Gutman, A. Dolev, A. Genizi, and D. Ganskopp. 2005. Inference of526

Animal Activity From GPS Collar Data on Free-Ranging Cattle. Rangeland Ecology and527

Management 58(3):256–266.528

Veblen, T. T., C. Donoso, T. Kitzberger, and A. J. Rebertus. 1996. Ecology of southern Chilean529

and Argentinean Nothofagus forests. The Ecology and Biogeography of Nothofagus Forests. 530

pp. 293–353.531

Veblen, T. T., and G. H. Stewart. 1982. The effects of introduced wild animals on New Zealand532

forests. Annals of the Association of American Geographers 72(3):372–397.533

Venables, W. N., and B. D. Ripley. 2002. Modern Applied Statistics with S . New York: Springer.534

Retrieved from http://www.stats.ox.ac.uk/pub/MASS4535

536

537

538


25/34

NICOLÁS SEOANETable 1. Alternative models for free-range cattle habitat use in the study area; the table shows539

Akaike information criterion value (AIC), number of predictor variables (K ) and ranking of540

models relative to each individual (R).541

542

Models Animal 1 Animal 2 Animal 3 Animal 4

K AIC ΔAIC R AIC ΔAIC R AIC ΔAIC R AIC ΔAIC R

A

Elevation + elevation2 +

slope + slope2 + NWness 5 187.5 0.0 1 619.9 2.4 3 516.0 5.8 2 1813.5 18.1 4

B

Model A + shrubland +

meadow 7 188.9 1.4 2 622.7 5.2 5 510.2 0.0 1 1809.2 13.8 3

C

Elevation +

NWness*slope 4 189.5 2.0 3 617.5 0.0 1 528.9 18.7 5 1795.4 0.0 1

D

Model C + shrubland +

meadow 6 192.4 4.9 4 619.8 2.3 2 520.5 10.3 3 1795.5 0.1 2

E

Elevation + slope +

Nwness + forest +

shrubland + meadow 6 196.3 8.8 5 620.8 3.3 4 521.7 11.5 4 1832.1 36.7 5

543

544

545

546

547


26/34

NICOLÁS SEOANE548

Table 2. Coefficients and 95% confidence intervals of the most parsimonious RSF models for549

selection of topographic and vegetation resources during autumn by free-range cattle in the study550

area.551

552

Variables Animal 1 Animal 2 Animal 3 Animal 4

Coef. 95% CIs Coef. 95% CIs Coef. 95% CIs Coef. 95% CIs

Intercept -39.91 -67.66 -12.16 -2.73 -4.05 -1.61 -2.22 -3.44 -1.07 -1.59 -2.38 -0.83

Elevation -56.92 -110.64 -3.20 -1.54 -2.90 -0.46 -3.54 -4.63 -2.56

Elevation² -1.68 -2.82 -0.73

Slope -2.65 -4.21 -1.11 -1.43 -2.79 -0.33 1.45 0.68 2.38 0.73 0.20 1.30

Slope² -0.26 -1.20 0.48 1.08 0.47 1.76

NWness -2.64 -5.27 -0.00 0.77 0.12 1.46 1.15 -0.01 2.54 2.01 1.29 2.73

Shrubland -0.58 -1.63 0.53

Meadow -1.21 -3.46 1.04 4.01 1.61 7.54

NW*Slope -1.42 -2.90 -0.26 1.99 1.38 2.63

553

554

555

556

557

558

559


27/34

NICOLÁS SEOANE560

Table 3. Population model of resource selection by cattle, with estimate coefficients561

(standardized variables), SEs and 90% percentile confidence intervals.562

563

90% Confidence intervals

Coefficient SE Lower limit Upper limit

Intercept -9.368 1.136 -12.409 -6.327

Elevation -3.016 0.864 -4.776 -1.255

Slope:Nwness -4.772 1.319 -8.867 -0.677

Slope -5.958 1.799 -13.576 1.660

NWness -8.165 1.598 -14.180 -2.151

Shrubland -0.060 0.795 -1.432 1.545

Meadow 1.521 1.361 -2.839 5.882

564

565

566

567

568

569

570

571


28/34

NICOLÁS SEOANE572

573

Figure 1. Use vs. Availability of topographic resources elevation and slope.574

575

Figure 2. General proposed model estimates (black) and individual model estimates (grey) with576

90 % confidence intervals.577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594


29/34

NICOLÁS SEOANE595

596

Figure 1597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617


30/34

NICOLÁS SEOANE618

619

Figure 2620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640


31/34

NICOLÁS SEOANE641

Appendix: The models and their coefficients for all animals.642

############################################################ 643

1) Description of the models644

2) Coefficients of all tested models and animals645

############################################################646

1- Description of the models647

θ , μnomial NegativeBi y ii ~ 648

i j0i variable β + β +total = μ .lnln 649

where:650

y = number of observations per sampling unit; = relative frequency in each sampling unit by animal; =651

overdispersion parameter; total = number of recorded positions in whole study area by animal; i = sampling unit652

with associated habitat covariates; j = number of variables in each model.653

654

Model Description of variables involved in each model

A elevation + sqr.elev + slope + sqr.slope + NWness

B elevation + sqr.elev + slope + sqr.slope + NWness + shrubland + meadow

C elevation + slope + NWness * slope

D elevation + NWness * slope + shrubland + meadow

E elevation + slope + NWness + forest + shrubland + meadow

655

656

657

658

659


32/34

NICOLÁS SEOANE660

661

2- Coefficients of all tested models and animals* 662

* Coefficient with this symbol are significant at a level of 95% 663

# Animal 1664

Model A Model B Model C Model D Model E

Intercept -72.577* -75.991* -11.633* -11.422* -10.516*

Elevation -130.564* -137.462* -5.748* -5.718* -5.394*

Elevation2 -62.145* -65.576*

Slope -2.343 -2.657 -3.063* -3.051* -2.536*

Slope2 -0.101 -0.236

NWness -0.060 -0.043 -2.852* -2.621* -0.010

Shrubland -0.259 -0.284 -0.252

Meadow -1.681 -1.127 -1.218

Forest 0.174

NWness*Slope -2.279* -2.117*

Theta 0.212 0.238 0.164 0.168 0.160

Theta SE 0.098 0.113 0.067 0.069 0.067

665

# Animal 2666


Intercept -3.174* -2.686* -3.614* -3.272* -3.126*

Elevation -1.577* -1.585* -1.967* -1.937* -1.572*


33/34

NICOLÁS SEOANE

Elevation2 0.127 -0.078

Slope -1.515* -1.442* -1.334* -1.227* -1.167*

Slope2

-0.231 -0.252

NWness 0.650* 0.783* 1.039* 1.312* 0.784

Shrubland -0.627 -0.685 -0.275

Meadow -0.203 -0.232 0.105

Forest 0.387

NWness*Slope 0.403 0.516

Theta 0.062 0.063 0.063 0.065 0.064

Theta SE 0.011 0.011 0.011 0.012 0.012

667

# Animal 3668


Intercept -2.834* -2.683* -1.943* -2.264* -2.939*

Elevation -1.512* -1.636* -1.713* -1.698* -1.711*

Elevation2 -1.288* -1.597*

Slope 1.652* 1.969* 2.344* 3.010* 2.927*

Slope2 1.196* 1.043*

NWness 0.874* 1.444* 0.233 1.034* 0.994*

Shrubland -0.672 -0.712 -0.202

Meadow 2.929* 3.937* 4.166*

Forest 1.340


34/34

NICOLÁS SEOANE

NWness*Slope -0.161 -0.873*

Theta 0.040 0.044 0.031 0.037 0.036

Theta SE 0.008 0.008 0.006 0.007 0.007

669

# Animal 4670


Intercept 0.073 -1.003* -1.594* -1.817* 0.230

Elevation -3.173* -3.131* -3.545* -3.420* -2.639*

Elevation2 -1.273* -0.782*

Slope 1.471* 1.591* 0.739* 0.783* 1.364*

Slope2 0.714* 0.830*

NWness 0.214 0.132 2.012* 1.740* -0.177

Shrubland 1.120* 0.634 -0.053

Meadow 0.212 0.794 0.764

Forest -0.647

NWness*Slope 1.996* 1.872*

Theta 0.072 0.075 0.078 0.079 0.068

Theta SE 0.007 0.007 0.007 0.008 0.006

671

672

673

Documents

First records on habitat use of semi-feral cattle in southern forests: topography reveals more than vegetation