Direct counts underestimate mountain ungulate population size

doi:10.21203/rs.3.rs-4009600/v1

Download PDF

Research Article

Direct counts underestimate mountain ungulate population size

https://doi.org/10.21203/rs.3.rs-4009600/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The topography of mountain habitats limits the accuracy of methods to assess the population size of mountain ungulates. This fact hampers decision-making for monitoring and conservation purposes and thus any attempt to evaluate the precision in known-size populations is more than welcome. In this work, we tested the accuracy of direct counts and distance sampling to assess the size of an Iberian ibex (Capra pyrenaica) flock of known size. We evaluated the influence of observer expertise (beginners and experts) on the detection error of female and male ibexes and whether the training of observers contributes to boosting the accuracy of density estimates. The ibex flock was comprised of 18 individuals (9 males, 8 females, and a male yearling) living in a 17 ha enclosure with natural Mediterranean vegetation in the National Game Reserve of Els Ports de Tortosa i Beseit, northeast Spain. After 27 surveys, experts detected 16% more ibexes than beginners. Male ibexes were ~ 13% easier to detect than females, and experts were more accurate than beginners in sexing. Additionally, the detection error in absolute counts was quite similar among beginners, but different among experts (> 10%). Despite the reduction in detection error over increasing effort scenarios, under-detection was greater than 50% in all events (> 85% for beginners and > 67% for experts). This study suggests the systematic underestimation of direct counts and density estimates of mountain ungulate populations in Mediterranean landscapes and the contribution of expertise to the improvement of the direct observation method. Our results show that wildlife managers assessing mountain ungulate populations for managing purposes should consider completing direct counts with alternative methods to minimize this systematic underestimation. Furthermore, surveys of the impact of infectious diseases on ungulate populations by direct observations may also result in the underestimation of the disease's impact on the host population.

Absolute counts

disease surveillance

distance sampling

Capra pyrenaica

trophy hunting

population size

Population size assessment is crucial for population management purposes (Williams et al., 2002). Obtaining accurate estimates, however, is difficult and costly for most mammal species (Overton, 1971), but in particular for those elusive species inhabiting forested (Hewison et al., 2007) and mountainous habitats (Kavčić al., 2021). Most European ungulate populations are managed for hunting purposes (Putman et al. 2011), and their hunting quotas rely on population size estimations through annual counts consisting of the direct observation of individuals (i.e., direct counts) of different sex and age categories (Imperio et al., 2010). Since hunting statistics, and thus hunting quotas depend on the accuracy of such counts (Baskin, 2016; Månsson et al., 2011), errors or biases in determining the number age and sex of animals to harvest will determine hunting sustainability in the short and long terms. Along the same lines, counts for population monitoring for conservation purposes are also affected by the same bias limiting the extent and applications of management decisions.

Mountain ungulates (e.g., wild goats, sheep, and their relatives), are the most appreciated species among European ungulates because of their economic ecological and aesthetic relevance. These mammals inhabit highly heterogeneous habitats limiting the accuracy of such population estimates (Pérez et al. 2002b). Even though mountain ungulates inhabit places where visual contact is limited, their population size assessments rely on the direct observation of individuals in line transects (Escos and Alados. 1988, Focardi et al. 2007, Torres et al. 2014), block counts (Herrero et al. 2011, Buzzard et al. 2012, Corlatti et al. 2015), or aerial surveys (Franke et al., 2012; Peterson et al., 2020). These methodologies usually require a considerable sampling effort to increase detection probability (Focardi et al., 2007) and to reduce the variation associated with the estimated densities (Pérez et al., 2017). Such potential sources of variation in population counts range from the choice of the best time of the day and the habitat type for spotting the target species to the inter-observer differences in the ability to detect individuals (Garel et al. 2005). However, detection of error (DE) in direct counts has barely been assessed on populations of known size, and it is hard to properly determine how to minimise biases or errors when estimating wild ungulate population sizes (but see, Vincent et al., 1996). Minimising systematic errors during population counts to avoid bias is already the focus of modern wildlife management research (Schmidt et al., 2019).

In this study, we conducted a field experiment to assess the systematic error of observers with different expertise on a captive Iberian ibex population of known size. Populations of this iconic Iberian ungulate are systematically monitored for conservation (Perez et al. 2002a) or hunting purposes (Carvalho et al. 2020). Iberian ibex are often affected by sarcoptic mange caused by Sarcoptes scabiei with deleterious effects on individuals (López-Olvera et al., 2015) and populations (Pérez et al., 2002). Accurately assessing mange prevalence depends on the ability to detect healthy and sick animals (Valldeperes et al., 2019, Espinosa et al. 2020), and thus is directly affected by the aforementioned limitations of direct counts. Likewise, hunting quotas of ibexes rely on direct counts (Carvalho et al. 2020).

Thus, is crucial to determine the limitations of direct observations to support further management plans, not only for this endemic ungulate but also for other caprine counterparts sharing similar habitats. In this work, taking advantage of a flock of Iberian ibex kept in captivity for conservation purposes in an enclosure with mountainous orography and natural Mediterranean vegetation, we aimed to (1) evaluate the effects of observer expertise (beginners vs experts) and sampling effort on the DE of male and female ibexes, and (2) to assess the differences in the detection errors obtained between absolute and relative counts using distance sampling. We hypothesise that (1) DE of expert observers would be considerably lower than that of beginners for absolute counts and relative abundance, (2) because the sexual dimorphism of this ungulate DE would rely on the sex of ibexes and (3) increasing effort would reduce DE rate in both expertise groups.

The study was conducted in a double-fenced 17 ha enclosure (Fig. 1.) located in a mountainous area of the National Game Reserve “Els Ports de Tortosa i Beseit” (NGR hereafter), in northeastern Spain (40º46’08’’N, 0º20’04’’E, 450 m. a.s.l.). Vegetation in the enclosure is similar to the surrounding environment. The climate at the NGR is typically Mediterranean with hot, dry summers and cool, wet winters (Kottek et al. 2006). Natural vegetation in the enclosure is similar to vegetation in the NGR and provides enough food resources for the flock. The dominant land cover is a coniferous forest of Pinus pinaster, P. nigra (57%) with occasional Holly oak patches of Quercus ilex, whereas the shrub (Pistacia lentiscus, Erica multiflora, Q. coccifera, Rosmarinus officinalis, Genista scorpius, and Ulex parviflorus) and grassland stratum represent 22% and 20% respectively. A detailed description of the vegetation of NGR can be found in Casas et al. (1985).

From March to July 2017, we conducted 55 direct counts during seven field sessions and four vantage points repeated twice in the morning and the afternoon. Due to sudden bad weather conditions, one vantage point was excluded from our analysis.

These vantage points were placed outside the enclosure, and visibility from each point covered between 65 to 78% of the enclosure (Fig. 1). Following Garel et al. (2005), we established two observer categories with different expertise. The first category (experts) consisted of four wildlife biologists (JLM, JRLO, SL, and ES) with prior experience observing Iberian ibex in mountain terrain. The second category (beginners) consisted of four veterinary and biology master students (GP, CC, MV, ME), with experience in wildlife watching but without experience in mountain field counts or with Iberian ibex sightings. For the two expertise categories, two observers performed multiple observer point counts with an independent double-observer approach (Riddle et al., 2010). Both beginners and experts simultaneously observed ibexes from the same vantage points and there was no communication between them. Moreover, none of the observers knew the true number of animals in the enclosed survey area before the counts began. Both beginner and expert pairs simultaneously surveyed ibexes from the same vantage points. Each group was equipped with spotting scopes (ATS Swarovsky 20 x 60), and binoculars (Swarovsky 8 x 42). Observers performed direct observations for 20 minutes from 07:00 to 10:00, and from 17:00 to 21:00 coinciding with the peak of Ibex activity. Observers recorded the total number of individuals, group size, and sex-age class of ibexes, as well as the time of the day. The radial distance of the individual ibexes or clusters to the vantage point was also recorded using a 6x42 rangefinder (Bushnell Prime 1300) to inform distance sampling (DS hereafter) analysis. Once the experiment was finished, we determined the size of the flock after the physical capture of all animals using vertical nets (more details in López-Olvera et al. 2009 and Colom-Cadena et al. 2014).

Data analysis

We used the number of ibexes in the enclosure according to their sex and age class to calculate differences in DE defined as follows:

$$Detection eror \left(DE\right)=100*\frac{Real number of ibexes-Observed ibexes }{Real number of ibexes}$$

Then, we run a set of linear mixed models (LMMs) to determine the effect of observer expertise on DE including the observer expertise as a fixed term, and the observation point as a random factor. Since DS followed a normal distribution in terms of skew and kurtosis, we used the Identity function link and Gaussian errors in our LMM. For the assessment of the sex-specific DE, we used the same LMM approach after omitting the single yearling individual in the flock and hence 17 (9 males and 8 females) was the flock size for this analysis. In this second model, we included the observer's expertise and the ibex sex as fixed factors, and the observation point as a random term. Sex-specific DE did not follow a normal distribution and thus we conducted a Box-Cox transformation to minimise the residual pattern. Moreover, we evaluated the use of DS to minimise the impact of systematic error in the counts to provide a final population density (Marques et al., 2011), comparing the DS and direct observation density estimates between beginner and expert observers. In this analysis, we did not compare our sex-specific densities since counts were below the recommended number of observations to obtain reliable density estimates (Barraclough, 2000). To select a detection function for DS analysis, we tested uniform, half-normal, and hazard rate, and a maximum of two adjustment terms were used to balance between bias and precision (Pérez et al. 2017). Model selection was performed according to the lowest AIC (Akaike’s Information Criterion, Burnham and Anderson 2002) value and the coefficient of variation (CV) at a 95% confidence interval (CI) for population density was also calculated. In addition, we evaluated the evolution of DE over the sampling sessions to evaluate the potential training effect on beginner and expert observers.

Mixed models were performed using the packages “lme4” 1.1–23 version (Bates et al. 2015), and “MuMIn” 1.43.17 version for pseudo-R² estimations (Bartoṅ, 2023). Box-Cox transformation was conducted in the “MASS” library 7.3–60 version (Venables and Ripley 2022). Statistical analyses were performed in the statistical software R 4.3.3 (R Development Core Team 2024), and the Distance Sampling software (version 7.3, Thomas et al. 2010).

For the experts, the minimum DE for global counts ranged from 5.6% (17 ibexes sighted out of 18) to 94.4% (one individual sighted) and to 22.2% (12 ibexes sighted) to 94.4% (1 individual sighted) for beginners (Table 1, Fig. 2). Our linear mixed models revealed a significant effect of the expertise over the DE across the vantage points (F = 4.4, df = 45, P > 0.04). In fact, the observer expertise contributed more to explaining the observed variability in the DE (R² = 11.2%), than the differences among count sessions (R² = 8.2%).

Table 1

Descriptive statistics (mean ± SD, min-max) of the detection error in direct counts [(real number of ibexes – observed number of ibexes) / real number of ibexes] in %, of beginner and expert observers from four vantage points to determine the real number of ibexes in a 17 ha enclosure in the National Game Reserve Els Ports de Tortosa i Beseit, northeastern Spain.
Expertise	Vantage points
Expertise	1st VP	2nd VP	3rd VP	4th VP
Beginner	66.7 ± 12.2 (55.6–88.9)	77.8 ± 12.8 (66.7–88.9)	88.8 ± 5.5 (83.3–94.4)	68.5 ± 40.2 (22.2–94.4)
Expert	55.5 ± 22.1 (11.1–88.9)	62.8 ± 24.8 (5.6–88.9)	64.3 ± 25.23 (11.1–88.9)	54.8 ± 30.9 (11.1–94.4)

Similarly, the sex-specific DE was independently affected by the observer expertise (F = 14.5, df = 1, 584, P = 2e-04), and the sex of animals (F = 13.2, df = 1, 584; P = 3e-04). Mean DE was slightly higher in females than for males (4%), suggesting that experts and beginners sighted more males than females. Finally, the trend in the gradual decrease in the DE over the sampling sessions (10%) did not differ between expert categories (F = 0.3, P = 0.59) stabilising after 11 observation sessions (Fig. 3).

Regarding our DS population density assessment, the experts recorded 141 ibex observations whereas beginners recorded 68. The best detection function for the expert category was the hazard-rate function with cosine series-term adjustment (AIC = 1759.3). For the beginners, however, it was the hazard-rate function without series-term adjustments (AIC = 851.9).

The true population density in the enclosure was 106 ibexes/ km² (18 individuals in 0.17 km²). In this way, density estimation by experts for the hazard-rate function with cosine expansion was 22 ibexes/km² (equivalent to nearly four individuals in 17 ha) with a coefficient of variation (CV) of 18.4% [95% CI (15.5–31.8)], and thus, the systematic error underestimation of experts was around 78% (Table 2). The density estimation by beginners instead, for hazard-rate function without adjustments was 10 ibexes/ km² (equivalent to nearly two ibexes in 17 ha) with a CV estimate of 23% [95% CI (6.4–15.8)], resulting in an underestimation bias of 89%.

Table 2

Summary of Distance Sampling fit (HR, hazard rate; c, cosine adjustment) with respective adjustment terms and detection probability obtained. Besides, it is shown actual density (D) and estimated density ($\widehat{\text{D}}$) by different observer categories (Obs. Cat.) as well as coefficient of variation (CV) and confidence interval (95%).
Expertise	Model	Adj. Terms	Detection probability	D	$\widehat{\text{D}}$	CV (%)	95% CI
Expert	HRc	2	0.54	105.9	22.2	18.1	15.5–31.2
Beginner	HR	2	0.60	105.9	10.1	22.5	6.4–15.7

In this work, we have compared and evaluated the DE from observers with different expertise through direct observations using two different methods of density estimation in a population of known size. Our results are in line with previous work in other mountain habitats that hamper the ability of observers to detect ungulates but in particular for untrained observers (Corlatti et al., 2015). Detection error also depended on the sex of ibexes being females harder to detect than their larger male counterparts. The remarkable disparities between the current known number of Iberian ibexes in the enclosure and the estimates produced by the two observer categories underline the importance of observer skills. Previous studies have already evaluated the effect of observer training ( i.e., Evans et al. 2007; Forcey et al. 2006; Elphick 2008), but few have compared and assessed the differences in the results according to the observer experience when increasing sampling efforts. Many study conservation organizations worldwide rely heavily upon volunteers (e.g., British Deer Society), for their monitoring purposes (Newman et al., 2003). It is unusual, however, in programmes based on volunteers or on people with little experience in mammal monitoring, validating their effectiveness in conducting the monitoring (but see Farr et al. 2023). It is important, therefore, to identify tasks and methods that are both useful for mammal monitoring and suitable for volunteers as well as to quantify the potential bias of beginners compared to experts as well as other factors affecting their competence. In our case, and even though observer expertise influenced the ability to detect ibexes, landscape characteristics appear to drive the detection of this mountain ungulate. Even though training improves population monitoring (Jeffress et al., 2011), our systematic population size underestimation was only slightly corrected through sessions of training but not enough to get accurate population size estimates. In fact, after nine sampling sessions, both observer categories underestimated the real population size by more than 50%. In addition, ibex sex also affected detectability being females harder to detect than males independently of the observer expertise. According to Focardi et al. (2007), these sexual differences in the detection of gregarious dimorphic ungulates may arise from morphological (females are smaller than males), behavioral (females are more cautious than males, particularly when breeding) and spatial segregation differences between males and females. In our case study, we did not observe spatial segregation in the enclosure and thus the most plausible explanation for the higher detectability of males is their larger body sizes. But even if our counts had been focused only on males, more than half of the individuals remained undetected. This is surely due to the habitat structure in this diverse environment and the ibexes' position within that structure has influenced visibility. Recent research focused on evaluating how habitat structure and animal position influence the observer's capability to detect animals (Stein et al., 2022), concludes that landscapes with dense and diverse vegetation strongly influence the potential viewshed and consequently the visual cues accessible to individuals and in turn DE.

Obtaining absolute counts of the wildlife population is an arduous task and is sometimes not possible. However, estimates of the population size using methodological approaches accounting for the detection probability can bring more accurate results (Collier et al., 2008; Pollock et al., 2002). Nevertheless, in our study, the use of DS did not significantly improve the DE of both observer categories (Table 2). Increasing sampling effort is another way to increase the precision and accuracy of population estimates (Torres et al., 2014). Typically, the increase in sampling efforts allows the achievement of precise estimates in DS procedures (e.g., < 20% CV, Pérez et al. 2017). However, there exists a systematic error that cannot be reduced by increasing survey effort (Chen, 1998), as appears to occur in our study, surely due to the high vegetation cover and the rough topography (Fig. 4). Another alternative is the use of an independent double-observer approach to reduce observational bias and variation in direct observations (Forcey et al., 2006).

Management implications

Iberian ibex population assessments based on direct counts would be persistently affected by a systematic underestimation in the Iberian Peninsula. Such underestimation has practical consequences affecting annual hunting quotas for ibex populations (Carvalho et al. 2020), population management actions to get accurate Ibex numbers to avoid the impact of ibex overabundance (Perea et al. 2015), and population monitoring for wildlife health surveillance (Barroso et al. 2023, 2024). Along the same lines, basic research exploring links between population density and the natural history of ibex populations (e.g., Serrano et al. 2011, Carvalho et al. 2015), would also be affected by this inaccuracy.

To our understanding, this population underestimation is difficult to solve. According to our own experience, infrared thermography technology set in binoculars would improve the detection of individuals in dense vegetation (Cilulko et al. 2013).

Thermal imaging binoculars are fast being used to observe and detect wild animals in their habitats since animals appear as warm spots against a dark, cool background in the thermogram, which is sufficient to confirm their presence. An example of the improvement due to thermal imaging incorporation into wildlife monitoring can be seen in Ditchkoff et al (2005), who demonstrated that thermal imaging was more efficient than other reported methods for capturing white-tailed deer fawns. However, even though thermography is creating new opportunities for wildlife research has certain limitations, particularly in the Mediterranean environment. In fact, the high summer solar radiation heating the ground and rocks blinds the observer due to the excessive brightness (personal observations and also reported in Cilulko et al. 2013). Another approach is the use of dung counts and DS to assess ungulate density in dense vegetation areas (Marques et al. 2001) which in turn are related to herbivore pressure on the vegetation (Limpens et al. 2020). There is no easy discrimination among pellets from different ungulate species nor direct contact with individuals to assess their age and sex categories limiting its use for target population control including game management or diseased individual’s alternative would be the use of camera traps to fuel random encounter models (REM). This approach has been widely applied due to its practical advantages such as no need for species-specific study design. Cameras would be set in areas with low visibility or dense vegetation obtaining precise estimates for a wide range of ungulate species as shown by Palencia et al. 2022.

Wildlifers in charge of the sustainable management of Iberian ibex populations must take into account the systematic underestimation of real population numbers in their management plans. Since there is no sole solution to this underestimation, managers should conduct population-specific designs to improve not only Iberian ibex management but also other ungulate species living in areas with low visibility.

Author contribution G.P.C., J.M.L.M., S.L., and E.S., contributed to conceptualization and methodology. G.P.C., C.C., M.V., and E.S., conducted field work., G.P.C., M.E.G., M.T.D. and J.M.P.J., contributed to investigation and analysis. G.P.C., and E.S., contributed to writing the manuscript’s original draft. J.R.L.O., and M.T.D., contributed to the supervision, review, and editing of the manuscript. All authors reviewed the manuscript.

Funding Research activities of J.M. Pérez are partially supported by the Junta de Andalucía (RNM-118 group) and by the Universidad de Jaén (1b action). This project benefitted from the research grants CGL2012-40043-C02-01, CGL2012-40043-C02-02 and CGL2016- 80543-P of the Spanish Ministerio de Economía y Competitividad. M. Valldeperes was supported by a FI-GENCAT Fellowship (2020_FI_B2_00049) and M. Escobar by an FPU grant (FPU19/ 04651).

Availability of data and materials datasets analyzed during the current study are available to readers upon request.

Ethics approval and consent to participate No approval of research ethics committees was required to accomplish the goals of this study.

Competing interests. R. López-Olvera is an associated editor of the European Journal of Wildlife Research.

Acknowledgements Thanks to Juanjo Estrada, Johan Espunyes, Cristina Recasens, and Federica Pizzato for their assistance during fieldwork. We greatly appreciate the support provided by the Reserva Nacional de Caça dels Ports de Tortosa i Beseit to conduct this work.

Bartoń K (2023) MuMIn: Multi-model inference_. R package version 1.47.5, <https://CRAN.R-project.org/package=MuMIn>.
Barraclough RK (2000) Distance sampling: A discussion document produced for the Department of conservation. Sci Res Intern Rep 175.
Barroso P, López-Olvera JR, Kiluba wa Kiluba T, Gortázar C (2024) Overcoming the limitations of wildlife disease monitoring. Research Directions: One Health. Published online 2024:1-37. doi:10.1017/one.2023.16
Barroso P, Relimpio D, Zearra JA, Cerón JJ, Palencia P, Cardoso B, Ferreras E, Escobar M, Cáceres G, López-Olvera JR, Gortázar C (2023) Using integrated wildlife monitoring to prevent future pandemics through One Health approach. One Health 16: 100479. https://doi.org/10.1016/j.onehlt.2022.100479
Baskin L (2016) Hunting as sustainable wildlife management. Mammal Study 41: 173–180 https://doi.org/10.3106/041.041.0402
Bates D, Maechler M, Bolker B, Walker S (2015) Fitting linear mixed-effects models using lme4. J Stat Softw 67(1): 1-48. doi:10.18637/jss.v067.i01.
Burnham, K.P., Anderson, D.R., 2002. Model selection and multimodel inference: a practical information-theoretic approach. Springer, New York
Buzzard PJ, Wong HM, Zhang H (2012) Population increase at a calving ground of the endangered Tibetan antelope Pantholops hodgsonii in Xinjiang, China. Oryx 46: 266–268. https://doi.org/10.1017/S0030605311001657
Carvalho J., Granados JE, López-Olvera J, Cano-Manuel FJ, Pérez JM, Fandos P, Soriguer RC, Velarde R, Fonseca C, Ráez A, Espinosa J, Pettorelli N, Serrano E (2015) Sarcoptic mange breaks up bottom-up regulation of body condition in a large herbivore population. Par Vectors. 8:572
Carvalho J, Büntgen U, Pettorelli N, Mentaberre G, Olivé-boix X, Eizaguirre O, Pérez JM, Fandos P, Torres RT, Lavín S, Fonseca C, Serrano E (2020) Habitat and harvesting practices influence horn growth of male Iibex. J Wild Manage 84: 651-665
Casas i Sicart ., Cros i Matas R, Brugués M, Sérgio C, Sim M (1985) Estudi de la brioflora dels Ports de Beseit. Orsis Org i Sist Rev Botànica Zool i Ecol 13–31.
Cilulko J, Janiszewski P, Bogdaszewski M, Szcygielska E (2013) Infrared thermal imaging in studies of wild animals. Eur J Wildl Res 59: 17-23.
Chen SX (1998) Measurement errors in line transect surveys. Biometrics 54: 899. https://doi.org/10.2307/2533844
Colom-Cadena A, Velarde R, Salinas J, Borge C, García-Bocanegra I, Serrano E, Gassó D, Bach E, Casas-Díaz E, López-Olvera JR, Lavín S, León-Vizcaíno L, Mentaberre G (2014) Management of a caseous lymphadenitis outbreak in a new Iberian ibex (Capra pyrenaica) stock reservoir. Acta Vet Scand 56: 83.
Collier N, Mackay DA, Benkendorff K (2008) Is relative abundance a good indicator of population size? Evidence from fragmented populations of a specialist butterfly (Lepidoptera: Lycaenidae). Pop Ecol 50: 17–23. https://doi.org/10.1007/s10144-007-0056-2
Corlatti L, Fattorini L, Nelli L (2015) The use of block counts, mark-resight and distance sampling to estimate population size of a mountain-dwelling ungulate. Pop Ecol 57: 409–419. https://doi.org/10.1007/s10144-015-0481-6
Ditchkoff SS, Raglin JB, Smith JM, Collier BA (2005) From the field: capture of white-tailed deer fawns using thermal imaging technology. Wildl Soc Bull 33(3):1164–1168
Elphick CS (2008) How you count counts: The importance of methods research in applied ecology. J Appl Ecol 45: 1313–1320. https://doi.org/10.1111/j.1365-2664.2008.01545.x
Escos J, Alados CL (1988) Estimating mountain ungulate density in Sierras de Cazorla y Segura. Mammalia 52: 425–428. https://doi.org/10.1515/mamm-1988-0315
Espinosa J, Pérez JM, Raéz–Bravo A, Fandos P, Cano–Manuel FJ, Soriguer RC, López–Olvera JR, Granados JE (2020) Recommendations for the management of sarcoptic mange infree–ranging Iberian ibex populations. Anim Biodiv Cons 43: 137–149. https://doi.org/10.32800/abc.2020.43.0137
Evans JW, Evans CA, Packard JM, Calkins G, Elbroch M (2007) Determining observer reliability in counts of river otter tracks. J Wildl Manage 73: 426–432. https://doi.org/10.2193/2007-514
Farr CM, Ngo F, Olsen B (2023) Evaluating data quality and changes in species identification in a citizen science bird monitoring project. Citiz Sci Theory Pract 8:24.DOI: https://doi.org/10.5334/cstp.604
Focardi S, Isotti R, Tinelli A (2007) Line transect estimates of ungulate populations in a Mediterranean forest J Wildl Manage 66: 48-58. https://doi.org/10.2307/3802870
Forcey GM, Anderson JT, Ammer FK, Whitmore RC (2006) Comparison of two double-observer point-count approaches for estimating breeding bird abundance. J Wildl Manage 70: 1674–1681. https://doi.org/10.2193/0022-541x(2006)70[1674:cotdpa]2.0.co;2
Franke U, Goll B, Hohmann U, Heurich M (2012) Aerial ungulate surveys with a combination of infrared and high-resolution natural colour images. Anim Biodiv Conserv 35: 285–293. https://doi.org/10.32800/abc.2012.35.0285
Garel M, Cugnasse JM, Gaillard JM, Loison A, Santosa Y, Maublanc ML (2005) Effect of observer experience on the monitoring of a mouflon population. Acta Theriol 50;109–114. https://doi.org/10.1007/BF03192623
Herrero J, García Serrano A, Prada C, Fernández Arberas O (2011) Using block counts and distance sampling to estimate populations of chamois. Pirineos 166: 123–133. https://doi.org/10.3989/pirineos.2011.166006
Hewison A, Angibault JM, Bruno C, Coulon A, Jean-Luc R, Serrano E, Verheyden E, Morellet N (2007) Using radio-tracking and direct observation to estimate roe deer Capreolus capreolus density in a fragmented landscape: a pilot study. Wildl Biol 13: 313-320. 10.2981/0909-6396(2007)13[313:URADOT]2.0.CO;2.
Imperio S, Ferrante M, Grignetti A, Santini G, Focardi S (2010) Investigating population dynamics in ungulates: do hunting statistics make up a good index of population abundance? Wildl Biol 16: 205–214.
Jeffress MR, Paukert CP, Sandercock BK, Gipson PS (2011) Factors affecting detectability of river otters during sign surveys. J Wildl Manage 75: 144-150. https://doi.org/10.1002/jwmg.12
Kamil B (2020) MuMIn: Multi-model Inference. R package version 1.43.17. https://CRAN.R-project.org/package=MuMIn
Kavčić K, Palencia P, Apollonio M, Vicente J, Sprem N (2021) Random encounters model to estimate the density of mountain-dwelling ungulate. Eur J Wildl Res 67:87
Kirk J H, Sharp E J (2016) Thermal imaging techniques to survey and monitor animals in the wild : A methodology. Amsterdam, Netherlands: Academic Press,. Print. 341 pp
Kottek M, Grieser J, Beck C, Rudolf B, Rubel F (2006) World map of the Köppen. Geiger climate classification updated. Adv For Sci 15: 259–263. https://doi.org/10.1127/0941-2948/2006/0130
Limpens A, Serrano E, Rivera-Sánchez L, Bartolomé J, Baraza E (2020) Pellet accumulation as a proxy for herbivore pressure in a Mediterranean ecosystem. Range Ecol Manage 73: 636-641.
López-Olvera JR, Marco I, Montané J, Casas-Díaz E, Mentaberre G, Lavín S (2009) Comparative evaluation of effort, capture and handling effects of drive-nets to capture roe deer (Capreolus capreolus), Southern chamois (Rupicapra pyrenaica) and Spanish ibex (Capra pyrenaica). Eur J Wildl Res 55(3); 193-202.
López-Olvera JR, Serrano E, Armenteros A, Pérez JM, Fandos P, Carvalho P, Velarde R, Cano-Manuel FJ, Ráez A, Espinosa J, Soriguer RC, Granados JE (2015) Sex-biased severity of sarcoptic mange at the same biological cost in a sexually dimorphic ungulate. Parasit Vectors 8: 583.
Marques FC, Buckland ST, Coffin D, Dixon CE, Borches DL, Mayle BA, Peace AJ (2001) Estimating deer abundance from line transect surveys of dung: sika deer in southern Scotland. J Appl Ecol 38:349-363
Marques TA, Borchers STB, Borchers DL, Rexstad EG, Thomas L (2011) Distance Sampling. Int Encycl Stat Sci 1: 398–400. https://doi.org/10.1007/978-3-642-04898-2_214
Newman C, Buesching C D, Macdonald DW (2003)Validating mammal monitoring methods and assessing the performance of volunteers in wildlife conservation—“Sed quis custodiet ipsos custodies ?” Biol Conserv 113:189-197
Overton WS (1971) Estimating the numbers of animals in wildlife populations, in: Robert H Giles (Ed.), Wildlife Management Techniques Manual. The Wildlife Society, Washington, pp. 403–455
Palencia P, Barroso P, Vicente J, Hofmeester TR, Ferreres J, Acevedo P (2022), Random encounter model is a reliable method for estimating population density of multiple species using camera traps. Remote Sens Ecol Conserv 8: 670-682.
Perea R, Perea-García-Calvo R, Díaz-Ambrona, C, San Miguel A (2015) The reintroduction of a flagship ungulate Capra pyrenaica: Assessing sustainability by surveying woody vegetation. Biol Conserv 181: 9-17
Pérez JM, Granados JE, Soriguer RC, Fandos P, Márquez FJ, Crampe JP (2002) Distribution, status and conservation problems of the Spanish Ibex, Capra pyrenaica (Mammalia: Artiodactyla). Mammal Rev 32:26–39
Pérez J M, López-Martín JM, Espunyes J, Colom-Cadena A, Fernández-Aguilar X, Gassó D, Mentaberre G, Marco I, Corominas J, Lavín S, Serrano E (2017) Temporal pooling of point transect data increases precision in density estimates of southern chamois. Mamm Biol 86: 75-78
Peterson MK, Foley AM, Tri AN, Hewitt DG, DeYoung RW, DeYoung CA, Campbell TA (2020) Mark-recapture distance sampling for aerial surveys of ungulates on rangelands. Wildl Soc Bull 44: 713–723. https://doi.org/10.1002/wsb.1144
Pollock KH, Nichols JD, Simons TR, Farnsworth GL, Bailey LL, Sauer JR (2002) Large scale wildlife monitoring studies: Statistical methods for design and analysis. Environmetrics 13: 105–119. https://doi.org/10.1002/env.514
Putman R, Apollonio M, Andersen R, Eds (2011) Ungulate management in Europe. Cambridge University Press, Cambridge. 398 pp. https://doi.org/10.1017/CBO9780511974137
R Core Team (2024) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
Riddle JD, Pollock KH, Simons TR (2010) An unreconciled double-observer method for estimating detection probability and abundance. Auk 127: 841–849. https://doi.org/10.1525/auk.2010.09118
Schmidt JH, Reynolds JH, Rattenbury KL, Phillips LM, White KS, Schertz D, Morton JM, Kim HS (2019) Integrating distance sampling with minimum counts to improve monitoring. J Wildl Manage 83: 1454–1465. https://doi.org/10.1002/jwmg.21691
Serrano E, Granados JE, Sarasa M, González FJ, Fandos P, Soriguer RC, Pérez JM (2011) The effects of winter severity and population density on body stores in the
Iberian wild goat (Capra pyrenaica) in a highly seasonal mountain
environment. Eur J Wildl Res 57:45–55.
Stein RM, Lecigne B, Eitel JUH, Johnson TR, McGowan C, Rachlow JL (2022) Vegetation and vantage point influence visibility across diverse ecosystems: Implications for animal ecology. Front Ecol Evol 10:911051
Thomas L, Buckland ST, Rexstad EA, Laake JL, Strindberg S, Hedley SL, Bishop JRB, Marques TA, Burham KP (2010) Distance software: design and analysis of distance sampling surveys for estimating population size. J Appl Ecol 47: 5–14.
Torres RT, Herrero J, Prada C, Garcia-Serrano A, Fernandez-Arberas O, Garcia-Post R (2014) Estimating the population density of Iberan wild goat Capra pyrenaica and mouflon Ovis aries in a Mediterranean forest environment. For Syst 23: 36–43. https://doi.org/10.5424/fs/2014231-03322
Valldeperes M, Granados JE, Pérez JM, Castro I, Ráez-Bravo A, Fandos P, López-Olvera JR, Serrano E, Mentaberre G (2019) How sensitive and specific is the visual diagnosis of sarcoptic mange in free-ranging Iberian ibexes? Parasit Vectors 12: 1–7. https://doi.org/10.1186/s13071-019-3665-7
Venables W N, Ripley B D (2002) Modern applied Statistics with S. Fourth Edition. Springer, New York, 497 pp. ISBN 0-387-95457-0
Vincent J-P, Hewison A J M, Angibault J-M, Cargnelutti B (1996) Testing density estimators on a fallow deer population of known size. J Wildl Manage 60: 18-28.
Williams B, Nichols J, Conroy M (2002) Analysis and management of animal populations. Academic Press, USA, 817 pp.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
07 May, 2024
Reviews received at journal
07 May, 2024
Reviews received at journal
02 Apr, 2024
Reviewers agreed at journal
28 Mar, 2024
Reviewers agreed at journal
18 Mar, 2024
Reviewers invited by journal
13 Mar, 2024
Editor assigned by journal
13 Mar, 2024
Submission checks completed at journal
04 Mar, 2024
First submitted to journal
03 Mar, 2024

You are reading this latest preprint version

Direct counts underestimate mountain ungulate population size

Status:

Version 1

Abstract

Figures

Introduction

Material and methods

Data analysis

Results

Discussion

Declarations

References

Additional Declarations

Status:

Version 1