Past, present and future of chamois science
Data files
May 26, 2022 version files 1.89 KB
-
analysis_trend.csv
-
README.rtf
Abstract
The chamois Rupicapra spp. is the most abundant mountain ungulate of Europe and the Near East, where it occurs as two species, the Northern chamois R. rupicapra and the Southern chamois R. pyrenaica. Here, we provide a state-of-the-art overview of research trends and the most challenging issues in chamois research and conservation, focusing on taxonomy and systematics, genetics, life history, ecology and behavior, physiology and disease, management, and conservation. Research on Rupicapra has a longstanding history and has contributed substantially to the biological and ecological knowledge of mountain ungulates. Although the number of publications on this genus has markedly increased over the past two decades, major differences persist with respect to knowledge of species and subspecies, with research mostly focusing on the Alpine chamois R. r. rupicapra and, to a lesser extent, the Pyrenean chamois R. p. pyrenaica. In addition, a scarcity of replicate studies of populations of different subspecies and/or geographic areas limits the advancement of chamois science. Since environmental heterogeneity impacts behavioral, physiological and life history traits, understanding the underlying processes would be of great value from both an evolutionary and conservation/management standpoint, especially in the light of ongoing climatic change. Substantial contributions to this challenge may derive from a quantitative assessment of reproductive success, investigation of fine-scale foraging patterns, and a mechanistic understanding of disease outbreak and resilience. Improving conservation status, resolving taxonomic disputes, identifying subspecies hybridization, assessing the impact of hunting and establishing reliable methods of abundance estimation are of primary concern. Despite being one of the most well-known mountain ungulates, substantial field efforts to collect paleontological, behavioral, ecological, morphological, physiological and genetic data on different populations and subspecies are still needed to ensure a successful future for chamois conservation and research.
Methods
We investigated the trend of scientific publications on chamois species and subspecies between 1980 and 2020, using the Scopus search engine and searching for titles, abstracts, and keywords using the statement "rupicapra". Genus name was preferred over common names to avoid inclusion of articles related to other fields (e.g., chamois leather manufacture). To investigate temporal trends, the yearly number of chamois publications was regressed against publication year (numerical explanatory variable) by fitting generalized linear models in R (R Core Team 2021), assuming a Poisson conditional distribution. The predictor included year as either a linear or a polynomial term (from the 2nd to the 6th degree); the best model was selected using AIC values corrected for small samples (AICc: Hurvich and Tsai 1989) and inspected for quantile residual distribution (Dunn and Smyth 2018). In all Poisson models, a first-order autoregressive term was included to account for temporal correlation. The best model describing this pattern was a 3rd-degree polynomial (β = -0.648, z-value = -2.075, p-value = 0.038), which suggests a steady increase in the number of publications until 2010, with substantial stability since then.
Next, the trend of scientific publications on chamois species and subspecies in relation to other Caprinae publications, between 1980 and 2020, was investigated using the Scopus search engine and searching for titles, abstracts, and keywords using various statements and Boolean logic, while keeping all other default settings. First, the yearly number of publications that directly or indirectly involved chamois, was assessed using the statement "rupicapra" (see above). Then, the yearly number of publications on all Caprinae was assessed using the following search string: "rupicapra" OR "ammotragus" OR "arabitragus" OR "budorcas" OR "capra aegagrus" OR "capra caucasica" OR "capra cylindricornis" OR "capra falconeri" OR "capra ibex" OR "capra nubiana" OR "capra pyrenaica" OR "capra sibirica" OR "capra walie" OR "capricornis" OR "hemitragus" OR "naemorhedus" OR "nilgiritragus" OR "oreamnos" OR "ovibos" OR "ovis ammon" OR "ovis canadensis" OR "ovis dalli" OR "ovis nivicola" OR "ovis orientalis" OR "pantholops" OR "pseudois". For all statements, scientific names were preferred over common names to avoid the inclusion of articles related to other fields. For wild sheep and goats, full scientific names were preferred over genus names to avoid the inclusion of articles on the domestic forms. We are aware that these search criteria may overlook some literature, but this conservative approach leads to noise reduction in the outcome, which is desirable when comparing publication numbers over time. To investigate temporal trends, the yearly proportion of chamois publications within Caprinae publications (response variable) was regressed against publication year (numerical explanatory variable) by fitting linear models in R, assuming a Gaussian conditional distribution. For each response variable, the predictor included year as either a linear or a polynomial term (from the 2nd to the 6th degree); the best model was selected using AICc values and inspected for residual distribution. Gaussian models did not indicate issues of temporal autocorrelation. The best model describing this pattern was a simple linear model (Fig. 2), which revealed a slightly increasing trend over time (β = 0.001, t-value = 2.079, p-value = 0.044), meaning the number of publications increased more rapidly for chamois than for the Caprinae as a whole.
The statements "rupicapra rupicapra" and "rupicapra pyrenaica" vs. the statement "rupicapra" were used in Scopus to assess the proportion of species-specific chamois literature, pooling data between 1980 and 2020. Similarly, the proportion of subspecies-specific literature was investigated using the following statements: "rupicapra rupicapra cartusiana", "rupicapra rupicapra tatrica", "rupicapra rupicapra carpatica", "rupicapra rupicapra balcanica", "rupicapra rupicapra asiatica", "rupicapra rupicapra caucasica" vs. "rupicapra rupicapra", and "rupicapra pyrenaica parva", "rupicapra pyrenaica ornata" vs. "rupicapra pyrenaica". The two main subspecies, the Alpine chamois and the Pyrenean chamois, were not included in this latter search because in the literature they are often indicated as the species R. rupicapra and R. pyrenaica. Finally, the ratios between topic-specific and complete chamois literature were assessed using statements reflecting different research areas, including taxonomy and systematics, genetics, life history, ecology and behavior, physiology and diseases, conservation and management (cf. search statements in Table 1) vs. the statement "rupicapra". It should be noted that up to 1985 Rupicapra pyrenaica ssp. were all considered subspecies of Rupicapra rupicapra: although the articles published within this timeframe represent only about 3% of the entire literature analyzed (26 articles out of 756), we manually adjusted for the bias by inspecting the species and subspecies in these earlier articles.
References
Dunn, P. K., Smyth, G. K. 2018. Generalized linear models with examples in R. – Springer.
Hurvich, C.M., Tsai, C.L. 1989. Regression and time series model selection in small samples. – Biometrika 76: 297–307.
R Core Team 2021. R: A language and environment for statistical computing. – Vienna, Austria: R Foundation for Statistical Computing. – Retrieved from https://www.R-project.org/