Past and future uses of text mining in ecology and evolution

Unsupervised literature mining approaches for extracting relationships pertaining to habitats and reproductive conditions of plant species

Introduction

Fine-grained, descriptive information on habitats and reproductive conditions of plant species are crucial in forest restoration and rehabilitation efforts. Precise timing of fruit collection and knowledge of species' habitat preferences and reproductive status are necessary especially for tropical plant species that have short-lived recalcitrant seeds, and those that exhibit complex reproductive patterns, e.g., species with supra-annual mass flowering events that may occur in irregular intervals. Understanding plant regeneration in the way of planning for effective reforestation can be aided by providing access to structured information, e.g., in knowledge bases, that spans years if not decades as well as covering a wide range of geographic locations. The content of such a resource can be enriched with literature-derived information on species' time-sensitive reproductive conditions and location-specific habitats.

Methods

We sought to develop unsupervised approaches to extract relationships pertaining to habitats and their locations, and reproductive conditions of plant species and corresponding temporal information. Firstly, we handcrafted rules for a traditional rule-based pattern matching approach. We then developed a relation extraction approach building upon transformer models, i.e., the Text-to-Text Transfer Transformer (T5), casting the relation extraction problem as a question answering and natural language inference task. We then propose a novel unsupervised hybrid approach that combines our rule-based and transformer-based approaches.

Results

Evaluation of our hybrid approach on an annotated corpus of biodiversity-focused documents demonstrated an improvement of up to 15 percentage points in recall and best performance over solely rule-based and transformer-based methods with F1-scores ranging from 89.61 to 96.75% for reproductive condition - temporal expression relations, and ranging from 85.39% to 89.90% for habitat - geographic location relations. Our work shows that even without training models on any domain-specific labeled dataset, we are able to extract relationships between biodiversity concepts from literature with satisfactory performance.

Overcoming confusion and stigma in habitat fragmentation research

Anthropogenic habitat loss is widely recognized as a primary environmental concern. By contrast, debates on the effects of habitat fragmentation persist. To facilitate overcoming these debates, here we: (i) review the state of the literature on habitat fragmentation, finding widespread confusion and stigma; (ii) identify consequences of this for biodiversity conservation and ecosystem management; and (iii) suggest ways in which research can move forward to resolve these problems. Confusion is evident from the 25 most-cited fragmentation articles published between 2017 and 2021. These articles use five distinct concepts of habitat fragmentation, only one of which clearly distinguishes habitat fragmentation from habitat area and other factors ('fragmentation per se'). Stigmatization is evident from our new findings that fragmentation papers are more charged with negative sentiments when compared to papers from other subfields in the environmental sciences, and that fragmentation papers with more negative sentiments are cited more. While most empirical studies of habitat fragmentation per se find neutral or positive effects on species and biodiversity outcomes, which implies that small habitat patches have a high cumulative value, confusion and stigma in reporting and discussing such results have led to suboptimal habitat protection policy. For example, government agencies, conservation organizations, and land trusts impose minimum habitat patch sizes on habitat protection. Given the high cumulative value of small patches, such policies mean that many opportunities for conservation are being missed. Our review highlights the importance of reducing confusion and stigma in habitat fragmentation research. To this end, we propose implementing study designs in which multiple sample landscapes are selected across independent gradients of habitat amount and fragmentation, measured as patch density. We show that such designs are possible for forest habitat across Earth's biomes. As such study designs are adopted, and as language becomes more precise, we expect that confusion and stigma in habitat fragmentation research will dissipate. We also expect important breakthroughs in understanding the situations where effects of habitat fragmentation per se are neutral, positive, or negative, and the reasons for these differences. Ultimately this will improve efficacy of area-based conservation policies, to the benefit of biodiversity and people.

Effect of IUCN Red List category on public attention to mammals

Cultural data is a powerful tool to analyze public awareness of key societal issues, including the conservation of nature. I used two publicly available repositories of cultural data, Google Trends and Google Ngram, to quantify the effect of the International Union for the Conservation of Nature (IUCN) Red List conservation status on public attention toward 4539 mammal species. With Google Trends, I calculated whether Google searches for their common and scientific names have been increasing or decreasing over time. I also ran an anomaly detection analysis to investigate whether a change in red-list status directly results in an increase in Google searches. Additionally, I quantified the mentions of species' common and scientific names in English texts with Google Ngram. Overall, Google searches for most mammal species remained at similar levels or increased since 2008. The severity of species' IUCN Red List status was a significant predictor of increasing Google searches, although the effect size was relatively small. Red-list status seemed strongly confounded with mammal body size. Species that moved to a higher-risk category spiked significantly in Google searches directly after the new designation. The mention of species' common names in the Google Ngram's English 2019 corpus significantly increased as the red-list category increased. These results provide valuable insight into the importance of the IUCN Red List for increasing public awareness and the usefulness of publicly available cultural data on examining the effectiveness of specific conservation efforts and thus evaluating targets for support and funding.

Who is reporting non-native species and how? A cross-expert assessment of practices and drivers of non-native biodiversity reporting in species regional listing

Each year, hundreds of scientific works with species' geographical data are published. However, these data can be challenging to identify, collect, and integrate into analytical workflows due to differences in reporting structures, storage formats, and the omission or inconsistency of relevant information and terminology. These difficulties tend to be aggravated for non-native species, given varying attitudes toward non-native species reporting and the existence of an additional layer of invasion-related terminology. Thus, our objective is to identify the current practices and drivers of the geographical reporting of non-native species in the scientific literature. We conducted an online survey targeting authors of species regional checklists-a widely published source of biogeographical data-where we asked about reporting habits and perceptions regarding non-native taxa. The responses and the relationships between response variables and predictors were analyzed using descriptive statistics and ordinal logistic regression models. With a response rate of 22.4% (n = 113), we found that nearly half of respondents (45.5%) do not always report non-native taxa, and of those who report, many (44.7%) do not always differentiate them from native taxa. Close to half of respondents (46.4%) also view the terminology of biological invasions as an obstacle to the reporting of non-native taxa. The ways in which checklist information is provided are varied, but mainly correspond to descriptive text and embedded tables with non-native species (when given) mentioned alongside native species. Only 13.4% of respondents mention to always provide the data in automation-friendly formats or its publication in biodiversity data repositories. Data on the distribution of non-native species are essential for monitoring global biodiversity change and preventing biological invasions. Despite its importance our results show an urgent need to improve the frequency, accessibility, and consistency of publication of these data.