A Global Community‐sourced Assessment of the State of Conservation Technology

Metadata

• Author: conbio.onlinelibrary.wiley.com
• Full Title: A Global Community‐sourced Assessment of the State of Conservation Technology
• Category: articles
• Document Tags: #planet
• URL: https://conbio.onlinelibrary.wiley.com/doi/epdf/10.1111/cobi.13871

Highlights

• The technologies with the highest perceived potential were machine learning and computer vision, eDNA and genomics, and networked sensors.
• The most pressing challenges affecting the field as a whole were competition for limitedfunding, duplication of efforts, and inadequate capacity building.
• The key opportuni-ties for growth identified in focus groups were increasing collaboration and informationsharing, improving the interoperability of tools, and enhancing capacity for data analy-ses at scale. Some constraints appeared to disproportionately affect marginalized groups.
• These advancements havebrought attention to the emerging field of conservation tech-nology, previously defined as the “devices, software platforms,computing resources, algorithms, and biotechnology methodsthat can cater [to] the needs of the conservation community”(Lahoz-Monfort et al.,2019). We suggest the discipline is fur-ther defined by the developers and users of these tools, aswell as the market intermediaries that support their engage-ment. Although no singular technology can solve the currentglobal ecological crisis, devices such as camera traps, acousticsensors, drones, biologgers, and satellites, as well as increas-ingly powerful genomic and artificial intelligence applications,hold the potential to empower conservationists to better under-stand and manage the socioecological systems in which theywork.
• Most respondents reported working at conservation non-governmental organizations (NGOs) (37%), universities orresearch institutions (31%), or technology companies (17%).The remainder worked at private nontechnology institutions(9%), government agencies (4%), or other (1%; e.g., open-source community or retired) (Table1). Respondents’ primaryroles at their places of employment were mainly conserva-tion practitioners (32%), technologists (30%), and professors orresearchers (22%), followed by students or early career profes-sionals (12%) and other (3%; e.g., investors or policy makers )
• Survey respondents rated GIS and remote sensing, drones, andmobile apps highest in overall performance; 77% rated GIS and remote sensing as good or very good, and 74% rated bothdrones and mobile apps as good or very good (Figure2a). Thetechnology groups with the lowest overall performance ratingswere networked sensors, data management tools, and eDNAand genomics: 42%, 46%, and 35% of respondents, respectively,rated them as acceptable and 12%, 6%, and 10%, respectively,rated them as poor or very poor.
• In addition to overall performance, survey respondents were asked to rate the technologies’ potential capacity to advanceconservation if current problems were addressed. Respon-dents rated machine learning and computer vision, eDNA andgenomics, and networked sensors highest in this category: 95%,94%, and 92% of respondents, respectively, rated them as veryhelpful or game changers (Figure2b). The technology groupswith the lowest capacity ratings were mobile apps, data manage-ment tools, and camera traps: 20%, 16%, and 13% of respon-dents, respectively, rated them as nice to have, somewhat help-ful, or helpful.
• Many respon-dents also mentioned problems with data analytics (particularlythe integration and use of machine learning tools;n=52), thecost of technologies (n=48), and power and battery life con-straints limiting functionality (n=43)
• Insufficient technical skills ranked as the next most important user constraint(44%), followed by the time required to engage (e.g., learn newtechnologies; 41%). Respondents identifying as conservationtechnology developers and testers reported that financing wasalso a significant barrier for them. They rated securing fund-ing throughout the development cycle (67%) and securing seedfunding for projects (62%) as their top 2 constraints (Figure3b).Understanding the conservation tool landscape (who is doingwhat and where the gaps exist) was rated the next most pressingdeveloper constraint (31%)
• Participants across all 7 focus groups reinforced many ofthese themes when asked to identify opportunities for growth,most frequently highlighting increasing collaboration and infor-mation sharing (n=43), improving the interoperability oftools and data streams (n=32), and enhancing capacity formeaningful data analyses at scale (n=30) (Table2). Oneof the most explicit calls for action across 6 of the 7 focusgroups was improving data sharing (n=25) with particu-lar emphasis from participants working with biologgers andacoustics on establishing open data repositories to facilitatethe storage, curation, and analysis of global data sets. Underthe umbrella code of collaboration and information sharing,respondents saw an opportunity for a convening body, estab-lished following a national lab model, that with sufficient fund-ing could facilitate the level of global collaboration and coordi-nation needed to capitalize on the suggestions above.
• However, recurring problems were also reported to behampering the utility of conservation technologies in practice,including their reliability and performance in challenging con-ditions, limited power and data storage capacities, reliance onlandscape connectivity for data transmission, and accessibilityto conservation end users
• Although overlooking ethical concerns was ranked lowest bysurvey respondents in overall challenges facing the field, insuf-ficient evaluation of the impact was a top constraint across all7 focus group discussions, suggesting that this may be a higherlevel concern considered more frequently by experts than theaverage user or developer. Previous literature also indicates thisis an area of increasing importance to the discipline, raising con-cerns that, if implemented inappropriately, conservation tech-nologies may reinforce historical injustices and further sepa-rate conservation data and decision-making from those mostaffected by them (Adams,2019;Bryant,2002).
• For example, trepidations aboutthe social risks of tools such as drones and camera traps beingdeployed without appropriate legislative and ethical frameworks(Humle et al.,2014; Sandbrook et al.,2018; Wallace et al.,2018)recently led to the development of guidelines for the sociallyresponsible use of surveillance technologies (Sandbrook et al.,2021) and an ethical code of conduct for camera traps in wildliferesearch (Sharma et al.,2020).
• The first formal guidance onaddressing data privacy concerns when using social media datain conservation science was also recently published (Di Mininet al.,2021).
• Feedback from the global conservation technology commu-nity of practice describes an ideal vision of this emerging field10–20 years from now in which collaboration trumps com-petition; solutions are open, accessible, and interoperable; anduser-friendly data processing and management tools empowerthe rapid translation of data insights to conservation action.
• lthoughwillingness to collaborate and share information appears to begrowing, the infrastructure to support broader engagement inthese activities is still mostly lacking. For this reason, establish-ing open, community-curated data repositories was one of themost frequently identified opportunities in focus groups, partic-ularly by acoustics and biologging experts who do not currentlyreap the benefits of tools such as Wildlife Insights. Previouscalls have been made for similar infrastructure to accommodateglobal, multiyear acoustic data sets (Gibb et al.,2019). Notably,community science platforms (e.g., Zooniverse) have massivelyadvanced public engagement with conservation-data process-ing but have been similarly dominated by camera trap imagery.