A social media-based framework for quantifying temporal changes to wildlife viewing intensity DOI Creative Commons
Kostas Papafitsoros, Lukáš Adam, Gail Schofield

et al.

Ecological Modelling, Journal Year: 2022, Volume and Issue: 476, P. 110223 - 110223

Published: Nov. 29, 2022

Documenting how human pressure on wildlife changes over time is important to minimise potential adverse effects through implementing appropriate management and policy actions; however, obtaining objective measures of these their impacts often logistically challenging, particularly in the natural environment. Here, we developed a modular stochastic model that infers ratio actual viewing consecutive periods (years) using social media, as this medium widespread easily accessible. Pressure was calculated from number times individual animals appeared media pre-defined windows, accounting for time-dependent variables influence them (e.g. people with access media). Formulas confidence intervals ratios were rigorously validated, corresponding uncertainty quantified. We applied framework calculate loggerhead sea turtles (Caretta caretta) at Zakynthos island (Greece) before during COVID-19 pandemic (2019–2021) based 2646 entries. Our ensured temporal comparability across years data grouped window sizes, by correcting interannual increase use. Optimal sizes windows delineated, reducing while maintaining high time-scale resolution. The optimal around 7-days peak tourist season when more available all three years, >15 days low season. In contrast, raw exhibited clear bias quantifying pressure, unknown uncertainty. here allows widely-available be used objectively pressure. Its modularity allowed quantified combined, or subsets (different groups, situations locations), could any site supporting exposed tourism.

Language: Английский

Mapping indicators of cultural ecosystem services use in urban green spaces based on text classification of geosocial media data DOI
Madalina Gugulica, Dirk Burghardt

Ecosystem Services, Journal Year: 2023, Volume and Issue: 60, P. 101508 - 101508

Published: Jan. 13, 2023

Language: Английский

Citations

37

A framework to integrate innovations in invasion science for proactive management DOI
Charles B. van Rees, Brian K. Hand, Sean C. Carter

et al.

Biological reviews/Biological reviews of the Cambridge Philosophical Society, Journal Year: 2022, Volume and Issue: 97(4), P. 1712 - 1735

Published: April 22, 2022

ABSTRACT Invasive alien species (IAS) are a rising threat to biodiversity, national security, and regional economies, with impacts in the hundreds of billions U.S. dollars annually. Proactive or predictive approaches guided by scientific knowledge essential keeping pace growing invasions under climate change. Although rapid development diverse technologies has produced tools potential greatly accelerate invasion research management, innovation far outpaced implementation coordination. Technological methodological syntheses urgently needed close gap facilitate interdisciplinary collaboration synergy among evolving disciplines. A broad review is necessary demonstrate utility relevance work fields generate actionable science for ongoing crisis. Here, we such advances relevant including remote sensing, epidemiology, big data analytics, environmental DNA (eDNA) sampling, genomics, others, present generalized framework distilling existing emerging into products proactive IAS management. This integrated workflow provides pathway scientists practitioners disciplines contribute applied biology coordinated, synergistic, scalable manner.

Language: Английский

Citations

30

Identifying wildlife observations on twitter DOI
Thomas C. Edwards, Christopher B. Jones, Padraig Corcoran

et al.

Ecological Informatics, Journal Year: 2021, Volume and Issue: 67, P. 101500 - 101500

Published: Nov. 30, 2021

Language: Английский

Citations

28

Updated distribution of Zoropsis spinimana (Dufour, 1820; Araneae: Zoropsidae) in Germany and novel insights into its ecology based on a citizen science survey DOI Creative Commons
Alexander Wirth, Gaby Schulemann‐Maier

Frontiers in Arachnid Science, Journal Year: 2024, Volume and Issue: 3

Published: March 15, 2024

In Germany, Zoropsis spinimana (Dufour, 1820) is an introduced, likely synanthropic spider species. Here, we report the results of a nationwide mapping appeal conducted by citizen science platform NABU-naturgucker.de, used to assemble live distributional data for species in Germany. With help media interest this species, gathered valuable dataset and large image gallery just five weeks, received more than 15,000 records, representing 2.3-fold increase occupied territory compared previous knowledge. By analyzing detail, obtained novel insights into ecology eco-geography Z. including information on prey, coloration, potential predators, altitudinal distribution temporal appearance, along with two cases accidental human translocation.

Language: Английский

Citations

4

Social media and citizen science records are important for the management of rarely sighted whales DOI
Annabelle S. Cranswick, Rochelle Constantine, Hannah Hendriks

et al.

Ocean & Coastal Management, Journal Year: 2022, Volume and Issue: 226, P. 106271 - 106271

Published: July 1, 2022

Language: Английский

Citations

18

A novel approach to Indian bird species identification: employing visual-acoustic fusion techniques for improved classification accuracy DOI Creative Commons
Pralhad Gavali,

J. Saira Banu

Frontiers in Artificial Intelligence, Journal Year: 2025, Volume and Issue: 8

Published: Feb. 21, 2025

Accurate identification of bird species is essential for monitoring biodiversity, analyzing ecological patterns, assessing population health, and guiding conservation efforts. Birds serve as vital indicators environmental change, making critical habitat protection understanding ecosystem dynamics. With over 1,300 species, India's avifauna presents significant challenges due to morphological acoustic similarities among species. For monitoring, recent work often uses sensors collect sounds an automated classification system recognize Traditional machine learning requires manual feature extraction model training build system. Automatically extracting features now possible advances in deep models. This study a novel approach utilizing visual-acoustic fusion techniques enhance accuracy. We employ Deep Convolutional Neural Network (DCNN) extract from images Long Short-Term Memory (LSTM) network analyze calls. By integrating these modalities early the process, our method significantly improves performance compared traditional methods that rely on either data type alone or utilize late strategies. Testing iBC53 (Indian Bird Call) dataset demonstrates impressive accuracy 94%, highlighting effectiveness multi-modal approach.

Language: Английский

Citations

0

The wildlife nextdoor: Socioeconomics and race predict social media carnivore reports DOI Creative Commons
Wilson C. Sherman, Christopher J. Schell, Christine E. Wilkinson

et al.

The Science of The Total Environment, Journal Year: 2025, Volume and Issue: 977, P. 179227 - 179227

Published: April 14, 2025

Social media and other internet-based, community generated datasets are emerging as valuable tools in advancing our understanding of biodiversity distributions across urban environments. However, it is unclear how best to harness these data for managing mitigating human-wildlife conflicts an urbanizing world. In this study, we analyzed 2584 posts comments on carnivore sightings, human-carnivore interactions, attitudes towards carnivores via the neighborhood-based social platform Nextdoor, focusing 52 peri-urban neighborhoods near Angeles National Forest California. We focused two most frequently discussed species: coyote (Canis latrans) American black bear (Ursus americanus). social-ecological covariates potential predictors reports, also compared sightings species collected logging application, iNaturalist. found that whiter, wealthier, less densely populated closer national forest tended report more conflict, while conflict did not show a clear relationship with metrics racial makeup or intensity. wealthier had higher percentages population registered indicating possible bias participation. Comments expressing positive bears were almost five times common than coyotes. Finally, number Nextdoor reports both 11 numerous observations iNaturalist within same window time locations. conclude can be viable predicting interactions. utility coexistence will nullified if researchers managers do fully account socioeconomic biases influencing who participates reporting process. Building inclusive accessible could therefore beneficial equity wildlife engaging diverse public nature.

Language: Английский

Citations

0

Social media and deep learning reveal specific cultural preferences for biodiversity DOI Creative Commons
Ilan Havinga, Diego Marcos, P.W. Bogaart

et al.

People and Nature, Journal Year: 2023, Volume and Issue: 5(3), P. 981 - 998

Published: March 20, 2023

Abstract Social media has created new opportunities to map cultural ecosystem services (CES) related biodiversity at large scales. However, using these novel data understand people's preferences in relation CES remains a challenge. To address this, we trained deep learning model capture interactions with selected flora and fauna on Flickr as service compared this citizen science iNaturalist, photos of individual species considered human–species interactions. After mapping the distribution Great Britain find significant spatial differences two platforms. Using second, pretrained model, were also able identify different for groups such birds social versus science. better preferences, richness abundance group 36 bird species, sometimes finding between ecological measures. Our findings demonstrate that can be used include wider range assessments along‐side data. reflect only limited first‐hand experience biodiversity. Read free Plain Language Summary article Journal blog.

Language: Английский

Citations

9

Surveillance of coastal biodiversity through social network monitoring DOI Creative Commons
Pablo Otero,

Eva Velasco,

J. Valeiras

et al.

Ecological Informatics, Journal Year: 2024, Volume and Issue: 80, P. 102515 - 102515

Published: Feb. 8, 2024

Knowledge of marine biodiversity is vital for developing appropriate conservation policies. In the current Information Age, data shared by citizens in social networks are a cost-effective alternative to complement on-going monitoring programs, as well understand human interactions with natural environment from perspective. This information can be obtained transparent way citizen (passive science approach) after sharing relevant content such as: rare catches recreational fishermen, sightings invasive species, stranding cetaceans, sea turtle entanglements, episodes massive arrival jellyfish or between organisms, among others. study has analyzed posted on networking site X (formerly known Twitter) its launch 2007 2022, focusing those posts that apparently reported observation along Spanish coast. To avoid an initial bias, generic messages asking "who knows" if "anyone what they have found were captured, stating had something interesting. After retrieving ~11 K tweets potential information, 597 finally identified validation. Most observations (21%) corresponded gelatinous animals, fish (11%) and mammals also being frequent. 57% these adequately located over coast, drawing first coastal map Spain based this methodology. The results show technique low-cost tool complementary existing which allows studying occurrence temporal variability non-indigenous sensitive alert case arrivals jellyfish, cetaceans turtles,

Language: Английский

Citations

3

The value of social media wildlife sightings for elusive species monitoring: a population assessment of servals in a South African urban nature reserve DOI Creative Commons
Kyle Smith,

Michael J. Somers

Urban Ecosystems, Journal Year: 2025, Volume and Issue: 28(2)

Published: Feb. 21, 2025

Abstract Data sourced from social media platforms is an underutilised resource for wildlife research, especially in studying enigmatic species. This study evaluates the potential of such data to provide population and behavioural insights into elusive species, serval ( Leptailurus ), Rietvlei Nature Reserve, urban protected area South Africa. We collected 153 visitor sightings servals within reserve different online spanning June 2011 August 2024, which we identified 30 individual servals, including three long-term residents. Analysis these revealed a stable with evidence reproduction migration through permeable border fence. Behavioural information sightings, as prey captured habitat use, align existing knowledge ecology. Even though passive contributions by public generally falls short terms quality detail, this demonstrates that well-supported community can be valuable source basic species specific area. approach allows cost-effective research beneficial both management formulation conservation strategies.

Language: Английский

Citations

0