What you don't see is as important as what you do, when it comes to mapping Antarctic seabird breeding sites.
According to Australian Antarctic Division seabird expert, Dr Colin Southwell, knowing whether a species is present or absent from a site is critical for population and biodiversity studies, detecting change in species' distributions, and prioritizing sites for protection.
However, while wildlife literature is peppered with observations of breeding seabird presence, reports rarely include absence, or search effort - where a scientist looked and didn't find anything, or didn't look ('ignorance').
"You may not see anything because the animal isn't there, it's hidden, or you didn't search everywhere, and it can be difficult to record this," Dr Southwell said.
The distinction between whether a breeding seabird is absent or just hidden from view is important, as it skews the accuracy of observations towards species like penguins, that are large, easily detectable and widely distributed.
This could affect decisions around environmental and conservation management.
Dr Southwell and a team of Australian and international researchers address this issue in a new study that pulls together 100 years of information from published papers, unpublished reports, data held in the Australian Antarctic Data Centre, handwritten field notes, and oral histories.
The data covered breeding habitat for eight bird species, across 5000 km of East Antarctic coastline, and extending 2000 km inland.
"The distinction between presence, absence and ignorance is a central concept of the study, which is widely acknowledged in ecology but seldom addressed," Dr Southwell said.
"We tried to address this issue by considering where researchers have searched, how thoroughly they are likely to have searched, and how easy each species was to see.
"We then made inferences about the certainty of absence. If there was no evidence of searching at a place, we concluded there was ignorance."
The team used this approach to assess the thoroughness of surveys over the past 100 years, and identify knowledge gaps in the species and places surveyed.
"Unsurprisingly, observations were clustered close to permanently occupied research stations, most of which are located on the coast," Dr Southwell said.
"They were also biased in favour of species like Adélie penguins, which are monitored as a sentinel species for fisheries and climate change impacts.
"There was limited information on smaller species such as the Wilson's storm petrel and snow petrel, likely because they are well camouflaged or hidden in rock crevices and tend to occupy less accessible locations.
"We found that the coverage of observations in recent decades would be insufficient to effectively detect change in breeding locations into the future."
The team then looked at the effect these knowledge gaps had on environmental management and conservation actions aimed at mitigating the impacts of human activities - specifically aviation and fisheries - on seabirds.
As aviation can disturb breeding seabirds, understanding where knowledge is limited or absent is important in refining flight paths.
"Mapping where we have no knowledge is important, as breeding seabirds could be mistakenly assumed to be absent when they're not. This could lead to inadvertent disturbance," Dr Southwell said.
Similarly, management of krill fisheries by the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR) uses information on the distribution and abundance of predators, including seabirds, to distribute krill catches across location and time.
"Existing seabird population counts at known breeding sites are likely to underestimate the total populations in the CCAMLR Management Areas," Dr Southwell said.
"Identifying knowledge gaps in seabird breeding populations could inform new surveys where count data are missing or uncertain, to better represent seabird populations in management decisions."
The team say future monitoring should report absences as well as presence.
"For practical reasons we will never have perfect knowledge or ecological information, but conservation science should strive to improve knowledge and hence management outcomes," Dr Southwell said.
