Do not publish - New research shows that publishing papers on rare and endangered species could drive them to extinction due to poachers who are trawling scientific papers

The Chinese cave gecko, Goniurosaurus luii, was extirpated from its type locality shortly after it was first described scientifically. PHOTO: J. SARTORE/GETTY IMAGES

Biologists have long valued publishing detailed information on rare and endangered species. Until relatively recently, much of this information was accessible only through accessing specialized scientific journals in university libraries. However, much of these data have been transferred online with the advent of digital platforms and a rapid push to open-access publication. Information is increasingly also available online in public reports and wildlife atlases, and research published behind paywalls can often be found in the public domain. Increased data and information accessibility has many benefits, such as helping to improve repeatability in scientific studies and enhancing collaboration (1, 2). However, such readily accessible information also creates major problems in the context of conserving endangered species.

Increasingly, the dual-use research dilemma (3), in which research can have both substantial positive but also negative impacts, is pervading research on rare and endangered species, with information intended to aid conservation fueling illegal actions that harm biodiversity. Biologists must urgently unlearn parts of their centuries-old publishing culture and rethink the benefits of publishing location data and habitat descriptions for rare and endangered species to avoid unwittingly contributing to further species declines. Restricting information entails some costs, but these must be weighed against the increasing harm of unrestricted information accessibility.

At least three key issues associated with unrestricted access to information on rare and endangered species warrant careful attention. These risks are not new but are greatly exacerbated in an era of digital proliferation and open access. First, unrestricted access to species location information is facilitating a surge in wildlife poaching (4, 5), with many species at risk (6). Poaching has been documented in species within months of their taxonomic description in journals (4). For example, more than 20 newly described reptile species have been targeted in this way, potentially leading to extinction in the wild. Indeed, when the names of some of these species—such as the Chinese cave gecko, Goniurosaurus luii (see the photo)—are typed into a search engine, the text autopopulates to suggest a search to purchase these animals.

Second, unrestricted access to location data and habitat descriptions can disrupt the often delicate relationships between scientists and landowners. We have personal experience of this. Our research in Australia on restoring farmland biodiversity requires repeated access to farms and depends on high levels of trust among landholders. We have detected populations of endangered species such as the pink-tailed worm-lizard (Aprasia parapulchella). Our research permits demand that location records be uploaded to open-access government wildlife atlases. Soon after uploading records, people seeking the rare worm-lizard were caught trespassing, upsetting farmers, damaging important rocky outcrop habitats, and jeopardizing scientist-farmer relationships that have taken years to establish.

Third, unrestricted access to species information has the potential to accelerate habitat destruction and create other negative disturbances. The digital age has brought a desire among many nature enthusiasts to observe, photograph, and sometimes remove animals and plants (7). Animal behavior and habitats are often heavily disturbed in the process (8).

Decisions to publish sensitive information on endangered and newly described rare species must be based around a careful assessment of whether its publication will benefit or harm the target species (see the figure). Key trade-offs must be weighed. For example, easily accessible data can help amass the evidence to challenge development proposals that may affect endangered species. Increased data accessibility can also foster improved scientific repeatability and greater collaboration. Although withholding information may have some negative consequences, this action is increasingly needed (9), given that calls for better regulation and law enforcement to protect atrisk species have met limited success (4, 5).

Where species have high economic value (such as in the case of the Chinese cave gecko), withholding information may be the only option. Relevant government or regulatory agencies should be notified of scientific discoveries, and pathways for access from legitimate persons remain open. In moderate risk situations, spatial data might be buffered and only very broad location data provided. Where there is low risk of perverse outcomes, unrestricted publication of habitat descriptions and location information remains appropriate (see the figure).

Much information on endangered and newly described species can still be published without location data being provided and without undermining the integrity or repeatability of the scientific work [akin to the notion that the rediscovery of Lazarus species can be validated without the collection of voucher specimens (10)]. As such, negative trade-offs arising from the dual-use research dilemma are not as pronounced as in other fields. For example, restrictions on publishing methodological advances in the study of pathogen virulence can inhibit scientific research that can have considerable human health benefits but is sometimes deemed necessary because of the potential for this information to facilitate perverse outcomes (such as bioterrorism) (11, 12).

Assessing risks from publication Endangered or newly discovered species can be at risk from poachers if their location data is published. This scheme helps to assess whether publication should be restricted in particular cases. GRAPHIC: A KITTERMAN/SCIENCE

Some fields such as paleontology and archaeology have long maintained restrictions on the publication of site locations and promoted government policies and regulations to limit collection and trade in fossils, artefacts, and culturally sensitive and/or scientifically important material (13). Organizations such as the U.S. Forest Service do not disclose geospatial data in order to protect research sites (14). Other solutions include modification of research permits so that endangered species locations are not automatically uploaded into wildlife databases and masking such records on private land, as presently occurs in some states in the United States. Some of these approaches are already in place in conservation; for example, the open-access journal PLOS ONE has data exemptions for endangered species. However, current policies are specific to individual journals, data repositories, or organizations and lack consistent enforcement. A major benefit to the author-led self-censorship that we advocate is that restrictions of the dissemination of sensitive information can be implemented widely and immediately.

There are signs that this problem is beginning to be addressed. Journals such as Zootaxa that carry taxonomic descriptions of new species now publish new descriptions without location information (15). More researchers, journal editors, and data custodians need to follow their lead. Otherwise, the potential benefits of open-access scientific information and data for biodiversity conservation will be outweighed by the perverse effects of exposing wild populations to substantial added conservation threats. Although much information on endangered and rare species is already available online, it remains crucial to change our actions now to avoid unwittingly contributing to further species declines.

Notes Editors' Note Science authors concerned about disclosing location information of species in Science submissions should discuss this with their editor.