Dataset Information

Sentiment Analysis of Conservation Studies Captures Successes of Species Reintroductions.

ABSTRACT: Learning from the rapidly growing body of scientific articles is constrained by human bandwidth. Existing methods in machine learning have been developed to extract knowledge from human language and may automate this process. Here, we apply sentiment analysis, a type of natural language processing, to facilitate a literature review in reintroduction biology. We analyzed 1,030,558 words from 4,313 scientific abstracts published over four decades using four previously trained lexicon-based models and one recursive neural tensor network model. We find frequently used terms share both a general and a domain-specific value, with either positive (success, protect, growth) or negative (threaten, loss, risk) sentiment. Sentiment trends suggest that reintroduction studies have become less variable and increasingly successful over time and seem to capture known successes and challenges for conservation biology. This approach offers promise for rapidly extracting explicit and latent information from a large corpus of scientific texts.

SUBMITTER: Van Houtan KS

PROVIDER: S-EPMC7660424 | biostudies-literature | 2020 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Sentiment Analysis of Conservation Studies Captures Successes of Species Reintroductions.

Van Houtan Kyle S KS Gagne Tyler T Jenkins Clinton N CN Joppa Lucas L

Patterns (New York, N.Y.) 20200320 1

Learning from the rapidly growing body of scientific articles is constrained by human bandwidth. Existing methods in machine learning have been developed to extract knowledge from human language and may automate this process. Here, we apply sentiment analysis, a type of natural language processing, to facilitate a literature review in reintroduction biology. We analyzed 1,030,558 words from 4,313 scientific abstracts published over four decades using four previously trained lexicon-based models ...[more]

PMID: 33205082

Similar Datasets

Project description:Reintroductions are a powerful tool for the recovery of endangered species. However, their long-term success is strongly influenced by the genetic diversity of the reintroduced population. The chances of population persistence can be improved by enhancing the population's adaptive ability through the mixing of individuals from different sources. However, where source populations are too diverse the reintroduced population could also suffer from outbreeding depression or unsuccessful admixture due to behavioural or genetic barriers. For the reintroduction of Asiatic wild ass Equus hemionus ssp. in Israel, a breeding core was created from individuals of two different subspecies (E. h. onager & E. h. kulan). Today the population comprises approximately 300 individuals and displays no signs of outbreeding depression. The aim of this study was a population genomic evaluation of this conservation reintroduction protocol. We used maximum likelihood methods and genetic clustering analyses to investigate subspecies admixture and test for spatial autocorrelation based on subspecies ancestry. Further, we analysed heterozygosity and effective population sizes in the breeding core prior to release and the current wild population. We discovered high levels of subspecies admixture in the breeding core and wild population, consistent with a significant heterozygote excess in the breeding core. Furthermore, we found no signs of spatial autocorrelation associated with subspecies ancestry in the wild population. Inbreeding and variance effective population size estimates were low. Our results indicate no genetic or behavioural barriers to admixture between the subspecies and suggest that their hybridization has led to greater genetic diversity in the reintroduced population. The study provides rare empirical evidence of the successful application of subspecies hybridization in a reintroduction. It supports use of intraspecific hybridization as a tool to increase genetic diversity in conservation translocations.

Project description:Our study was undertaken to better understand how to increase the success rates of recovery plantings of a rare hemiparasite, golden paintbrush (Castilleja levisecta-Orobanchaceae). This species is endemic to western Washington and Oregon, USA, and southwestern British Columbia, Canada. Over 5000 golden paintbrush plants were outplanted as plugs in 2007 at six different native prairie sites that were considered to be suitable habitat, based on general evaluations of vegetation and soil conditions. Outplantings were installed at regular intervals along transects up to 1 km long to include a range of conditions occurring at each site. All plantings were re-examined five years later. The patchy distribution of surviving plugs and new recruits within each reintroduction site suggested success is strongly influenced by microsite characteristics. Indicator species analysis of taxa growing in microsites around outplanted golden paintbrush identified species that were positively or negatively associated with paintbrush survival. Species such as Festuca roemeri, Eriophyllum lanatum, and Viola adunca were strong indicators at some sites; non-natives such as Hypochaeris radicata and Teesdalia nudicaulis tended to be frequent negative indicators. Overall, higher richness of native perennial forbs was strongly correlated with both survival and flowering of golden paintbrush, a pattern that may reflect interactions of this hemiparasite with the immediately surrounding plant community. Topographic position also influenced outcomes, with greater survival occurring on mounds and in swales, where soils generally were deeper. Our findings suggest that assessments of site suitability based on vegetation alone, and coarser, site-level assessments that do not characterize heterogeneity at the microsite scale, may not be strong predictors of restoration success over the longer term and in sites with variability in vegetation and soils. By identifying suitable microsites to focus rare species plantings, survival and efficiency may be significantly enhanced.

Project description:Successful conservation plans are conditioned by our ability to detect anthropogenic change in space and time and various statistical analyses have been developed to handle this critical issue. The main objective of this paper is to illustrate a new approach for spatial analysis in conservation biology. Here, we propose a two-step protocol. First, we introduce a new disturbance metric which provides a continuous measure of disturbance for any focal communities on the basis of the surrounding landscape matrix. Second, we use this new gradient to estimate species and community disturbance thresholds by implementing a recently developed method called Threshold Indicator Taxa ANalysis (TITAN). TITAN detects changes in species distributions along environmental gradients using indicators species analysis and assesses synchrony among species change points as evidence for community thresholds. We demonstrate our method with soil arthropod assemblages along a disturbance gradient in Terceira Island (Azores, Portugal). We show that our new disturbance metric realistically reflects disturbance patterns, especially in buffer zones (ecotones) between land use categories. By estimating species disturbance thresholds with TITAN along the disturbance gradient in Terceira, we show that species significantly associated with low disturbance differ from those associated with high disturbance in their biogeographical origin (endemics, non-endemic natives and exotics) and taxonomy (order). Finally, we suggest that mapping the disturbance community thresholds may reveal areas of primary interest for conservation, since these may host indigenous species sensitive to high disturbance levels. This new framework may be useful when: (1) both local and regional processes are to be reflected on single disturbance measures; (2) these are better quantified in a continuous gradient; (3) mapping disturbance of large regions using fine scales is necessary; (4) indicator species for disturbance are searched for and; (5) community thresholds are useful to understand the global dynamics of habitats.

Dataset Information

Sentiment Analysis of Conservation Studies Captures Successes of Species Reintroductions.

Publications

Sentiment Analysis of Conservation Studies Captures Successes of Species Reintroductions.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets