The foundational concept of habitat lies at the very root of the entire science of ecology, but inaccurate use of the term compromises scientific rigor and communication among scientists and nonscientists. In 1997, Hall, Krausman & Morrison showed that ‘habitat’ was used correctly in only 55% of articles. We ask whether use of the term has been more accurate since their plea for standardization and whether use varies across the broader range of journals and taxa in the contemporary literature (1998–2012). We searched contemporary literature for ‘habitat’ and habitat‐related terms, ranking usage as either correct or incorrect, following a simplified version of Hall et al.'s definitions. We used generalized linear models to compare use of the term in contemporary literature with the papers reviewed by Hall et al. and to test the effects of taxa, journal impact in the contemporary articles and effects due to authors that cited Hall et al. Use of the term ‘habitat’ has not improved; it was still only used correctly about 55% of the time in the contemporary data. Proportionately more correct uses occurred in articles that focused on animals compared to ones that included plants, and papers that cited Hall et al. did use the term correctly more often. However, journal impact had no effect. Some habitat terms are more likely to be misused than others, notably ‘habitat type’, usually used to refer to vegetation type, and ‘suitable habitat’ or ‘unsuitable habitat’, which are either redundant or nonsensical by definition. Inaccurate and inconsistent use of the term can lead to (1) misinterpretation of scientific findings; (2) inefficient use of conservation resources; (3) ineffective identification and prioritization of protected areas; (4) limited comparability among studies; and (5) miscommunication of science‐based findings. Correct usage would improve communication with scientists and nonscientists, thereby benefiting conservation efforts, and ecology as a science.
Keywords: conservation implications, critical habitat, habitat, habitat‐related terms, misuse, operational terminology, standardization
Habitat is “the resources and conditions present in an area that produce occupancy – including survival and reproduction – by a given organism. Habitat is organism‐specific; it relates the presence of a species, population, or individual (animal or plant) to an area's physical and biological characteristics. Habitat implies more than vegetation or vegetation structure; it is the sum of the specific resources that are needed by organisms.”
(Hall, Krausman & Morrison, 1997)
1. INTRODUCTIONSince 1970, it is estimated that on erage vertebrate species’ populations (including mammals, birds, reptiles, amphibians, and fish) he declined by 52% as a result of human activities (Living Planet Index; WWF International, 2014). The primary driver causing these species’ population declines and extinctions is loss of habitat (Brook, Sodhi, & Bradshaw, 2008; Kerr & Deguise, 2004; Millennium Ecosystem Assessment, 2005; Venter et al., 2006). Loss of habitat could interact in complex, unforeseen ways with the looming global threat of climate change (Mantyka‐Pringle et al., 2015). But despite the recognition that habitat plays a key role in stemming global population declines, even the primary ecological literature may not support a consistent definition for the term (Hall, Krausman & Morrison, 1997; Guthery & Strickland, 2015; Mathewson & Morrison, 2015; see Krausman & Morrison, 2016).
‘Habitat’ is used to describe virtually every kind of location occupied by organisms—from the small‐scale microcosm habitat of hot spring pools which are the home of the Banff Spring Snail (Physella johnsoni—Lepitzki & Pacas, 2010) to the vast plains of the Serengeti Mara Ecosystem and other African sannas, where migratory herds of Wildebeest (Connochaetes taurinus) and Elephants (Loxodonta africana) track new green vegetation (Bohrer, Beck, Ngene, Skidmore, & Douglas‐Hamilton, 2014; Boone, Thirgood, & Hopcraft, 2006; Jarman & Sinclair, 1979). In some cases, correct use and identification of habitat could he direct implications for human life. For example, identifying temporal and spatial habitat for Great White Sharks (Carcharodon carcharias) can help mitigate threats to humans in inshore waters (Kock et al., 2013) and similarly delineating Tiger (Panthera tigris) habitat in the Sundarban in India and Bangladesh can oid conflicts with people (Naha et al., 2016).
Used today almost ubiquitously in the ecological and conservation literature, the word ‘habitat’ is a Panchreston problem, similar to that identified for another commonly used ecological term, ‘fragmentation’ (Bunnell, 1999; Fischer & Lindenmayer, 2007). The Panchreston problem refers to “an explanation or theory used in such a variety of ways as to become meaningless.” According to a search on Google Scholar, the term ‘habitat’ has been deployed in at least 2.4 million publications. Moreover, habitat‐related terminology pervades the scientific literature (Krausman, 1999) and includes a whole gamut of terms—among them; habitat type, habitat use, suitable habitat, habitat requirements, habitat fragmentation, and habitat heterogeneity. Ironically, if used correctly, most of these habitat‐related terms are self‐contradictory or oxymorons, for example, ‘habitat heterogeneity,’ ‘unsuitable habitat,’ and ‘habitat type’ (Mathewson & Morrison, 2015; Krausman & Morrison, 2016; see Box 1).
Box 1. Definitions for habitat‐related terms (from Krausman & Morrison (2016)—used with permission). 1. Habitat—The resources and conditions present in an area that produce occupancy, which may include survival and reproduction by a given organism. Habitat is organism‐specific and is more than vegetation or vegetation structure. Thus, suitable habitat is redundant and unsuitable habitat is a misnomer; if it was unsuitable, it would not be habitat! Neither term should be used. Instead reference unsuitable areas or unsuitable vegetation types. Habitat abundance—The amount of habitat ailable regardless of its ailability to animals. Because researchers can measure abundance, using this term is more accurate than using habitat ailability. Habitat ailability—The accessibility and procurability of physical and biological components of a habitat by animals. Habitat quality—The ability of the environment to provide conditions appropriate for individual and population persistence. Quality should be based on the demographics of the population and not necessarily numbers alone. Quality is an outcome (e.g., survival and productivity) and is not a user‐defined inherent property of a location. For example, Hall et al. (1997) suggested low habitat quality represents the resources ailable for survival, medium habitat quality represents resources ailable for survival and reproduction, and high‐quality habitat includes resources ailable for population persistence. These are critical distinctions because a geographic location (e.g., study area) could fluctuate from year‐to‐year in some critical resource (e.g., berry or insect production) yet retain the same basic vegetation composition and structure. Thus, habitat quality could vary from year to year. Habitat type—This term refers only to the type of vegetation association in an area or to the potential of vegetation to reach a specified climax stage (Daubenmire, 1968:72–73). Habitat type describes vegetation, not an animal species’ habitat. To describe vegetation in an animal's habitat, use vegetation types and associations to oid confusion with habitat type. Habitat use—The way an animal uses or consumes a collection of physical and biological resources. Habitat selection—Habitat selection is a hierarchical process that an animal uses to choose habitat components (Hutto, 1985; Johnson, 1980) Open in a new tab‘Habitat’ has been similarly defined by several authors: Hall et al. (1997) described it as “The resources and conditions present in an area that produce occupancy – including survival and reproduction – by a given organism” (Hall et al., 1997, 175). Garshelis (2000, 112) defined it as “…. the type of place where an animal normally lives or, more specifically, the collection of resources and conditions necessary for its occupancy.” More recently, Morrison (2009, p. 61) described habitat as “an area with a combination of resources (e.g., food, cover, water) and environmental conditions (e.g., temperature, precipitation, presence or absence of predators and competitors) that promotes occupancy by individuals of a given species (or population) and allows those individuals to survive and reproduce.” Hall et al. (1997) argued that “Habitat is organism‐specific; it relates the presence of a species, population, or individual (animal or plant) to an area's physical and biological characteristics. Habitat implies more than vegetation or vegetation structure; it is the sum of the specific resources that are needed by organisms.” Moreover, habitat is invariably measured by ecologists from a human perspective, which may be quite different to what organisms perceive. Here, we provide support and endorsement for the definition of ‘habitat’ used in the original paper by Hall et al. (1997). Twenty years ago, Hall et al. (1997) found that only nine of 50 (18%) of the articles they reviewed defined and used the term ‘habitat’ and habitat‐related terms consistently according to category 1 (strictest definition) in their standard terminology (see Methods). They found that use of the term ‘habitat’ varied for two main reasons: (1) Habitat relations varied according to spatial extent, with few authors documenting the scale of their investigations or limiting their discussions to the scale of their work (the implication being that results could not be extrapolated to other situations and contexts); (2) many authors used the term ‘habitat’ to refer to vegetation association or vegetation type, but ‘habitat’ is not synonymous with these terms.
So why is it important that the term ‘habitat’ be used correctly today, and why should scientists care? First, inaccurate and inconsistent use of the term can lead to misinterpretation of scientific findings (Herrando‐Pérez, Brook, & Bradshaw, 2014). Given that habitat loss is the primary driver of species’ population declines and extinctions, identifying and correctly defining habitat for organisms is necessary for its protection, management, and restoration. Not only do we need to evaluate and quantify the effects of human activities on species’ habitat, but we also must weigh the relative roles of species’ habitat loss compared to other threats, such as land cover fragmentation, invasive organisms, climate change, contaminants, or overexploitation (Wilcove, Rothstein, Dubow, Phillips, & Losos, 1998; Evans et al., 2011; Lesbarrères et al., 2014). Furthermore, we should measure the cumulative effects of those threats (e.g., Halpern & Fujita, 2013).
Second, and related to the above, incorrectly defining ‘habitat’ could lead to inefficient use of conservation resources. In a world where resources for conservation planning and biodiversity conservation are always limited, cost is a critical consideration (Carwardine et al., 2008; Daily & Ellison, 2002; Keith, Vardon, Stein, Stein, & Lindenmayer, 2017; Naidoo et al., 2006). Thus, we need to efficiently define and protect habitat in the face of limited conservation resources and increasing development pressures.
Third, misusing the term ‘habitat’ could lead to ineffective identification and prioritization of protected areas. Understanding the features and parameters that comprise habitat for individual species and how environmental changes impact those factors is part of this identification and prioritization. We need to be able to define and map habitat for as many species as possible, particularly for species at risk (Reed et al., 2006).
Armed with this information, we can then make predictions and, in turn, take management and conservation action. Advances in technology he meant that our ability to map habitat for species is greatly enhanced. Perhaps the potential to misuse the term is even greater now that such mapping has become so widely adopted; the temptation is to call GIS‐derived features, such as forest and wetland, ‘habitat’ (e.g., Evans, Costa, Tomas, & Camilo, 2014), whereas in fact ‘land cover’ would be a more accurate term. Another issue is that habitat by definition is species‐specific, but there has been a trend in ecology toward viewing ecosystems as a whole, rather than individual species, and perhaps this is why use of the term has become even more muddled. Examples are the use of the habitat‐related terms ‘habitat type’ and ‘habitat heterogeneity’ where ‘habitat’ is used in a generic sense.
Fourth, misusing the term ‘habitat’ means that there is potential for comparability among studies to become limited and, by extension, conducting meta‐analyses and complementarity to identify generic insights into species–habitat relationships.
Fifth and finally, we need to be able to communicate the concept of habitat effectively among scientists and between scientists and nonscientists. This means we need to define the term ‘habitat’ appropriately and adhere to this definition consistently (Guthery & Strickland, 2015). Hall et al. (1997) concluded that the inappropriate use of the word ‘habitat’ hinders communication with other scientists, managers, and the public. This ambiguity surrounding the use of the term can he widespread repercussions. For example, in Canada, there is a legal requirement to identify critical habitat for species at risk—defined as the habitat necessary for the survival or recovery of a listed endangered, threatened, or extirpated species in Schedule 1 of the Species at Risk Act (SARA, 2002).
The need for operationalizing scientific terms has never been more important, and yet according to Mathewson and Morrison (2015), little has changed in this regard over the last 50 years. While others he recently researched use of the term (e.g., Guthery & Strickland, 2015) and reiterated pleas for standard terminology (Krausman & Morrison, 2016), ours is the first systematic and quantitative study of how use of the term has changed over time. In this study, we searched the contemporary literature to compare our findings with Hall et al.'s results.
We conducted similar searches to those conducted by Hall et al., with three important modifications: (1) We asked whether analyzing articles in a wider range of journals than those used by Hall et al. would yield different conclusions. (2) We investigated a broader range of taxa to see whether there were taxon‐related differences in use of the term ‘habitat.’ (3) We asked whether it is the use of the term ‘habitat’ itself (the umbrella term) that is incorrect, or whether the problem lies in misuse of the numerous habitat‐related terms. We tested the effect of period (articles analyzed by Hall et al. vs. contemporary articles published since their publication), on use of the term. Within the contemporary literature (post‐1997), we tested whether the correct use of the term ‘habitat’ differed between articles that dealt with animals (mammal, bird, reptile, amphibian, and fish) versus those that included discussions of habitat use by plants. Finally, we assessed whether articles that specifically cited Hall et al. (1997) were more likely to use the term correctly than those that did not on the assumption that the authors of these articles would be more careful or consistent in their use of the term than those authors not citing Hall et al. (1997).
Like Hall et al. (1997), we believe that the need to operationalize1 the term is important, while also recognizing the limitations of the concept (Mathewson & Morrison, 2015). We agree with Peters (1991) that without a clear, operational definition, users of terms are able to develop their own interpretations of what they mean. And there is no justification for the sloppy use of terms in ecology (Morrison & Hall, 2002). Articles in peer‐reviewed journals play a pivotal role in adoption of science‐based evidence for practical conservation management (Ewen, Adams, & Renwick, 2013), which makes it critical to operationalize our terms, including use of the word ‘habitat.’
2. MATERIALS AND METHODS 2.1. Sampling articles that use the term ‘habitat’Hall et al. (1997) examined the use of the term ‘habitat’ in 50 articles from 11 journals and several scientific books and technical manuals between 1980 and 1994. They chose these journals and books because they represented (1) current (as of 1994) important wildlife publications (e.g., Wildlife Techniques Manuals, 4th and 5th editions—Schemnitz, 1980; Bookhout, 1994) and (2) bird–habitat and mammalian–habitat relationships (Table 1 in their paper, p. 175).
2.1.1. Post‐1997 survey of contemporary articlesTo objectively review the scientific literature, we used a search engine (Web of Science) for the time period post‐Hall et al.'s publication (1998 to 2012, hereafter ‘post‐1997’ or ‘contemporary’). We deliberately omitted the 4 years post‐1994 (the endpoint year examined by Hall et al.) because we also wanted to see whether the publication of Hall et al.'s, paper influenced usage of the term. This enabled us to compare their data (Table 2 in Hall et al., 1997, p. 176) and our own data, and to determine whether the use of the term ‘habitat’ had improved since Hall et al.'s plea for standardization.
We carried out exploratory analyses to determine how many articles we could find within journals containing the word ‘habitat.’ In the first phase of investigations, we replicated Hall et al.'s methods using the same nine journals (Table S1) and 22 habitat‐related search terms (Table S2) for the post‐1997 period. This initial search yielded 2,385 articles containing the word ‘habitat.’ We randomly selected 50 of these articles (the same number analyzed in Hall et al., 1997) in equal proportion from each journal for analysis. This represented 2% of the total search results.
For the second phase of investigations, we expanded the journal selection to include those that were not sampled in phase 1, but had a high impact factor (in the Web of Science category of ‘Ecology’), and that included studies on nonmammal vertebrate taxa. We included journals with a high impact factor because these may be more influential in how the term was used and theoretically should set a higher standard for correct usage. Many new ecology and conservation journals he been launched since Hall et al.'s publication, and we wanted to make sure we sampled a broad range of these journals; for example, 17 of the 50 journals we chose were first published in or after 1994 (Tables S1 and S2). We initially chose the top 30 journals (out of 129) that had the highest impact factor. Of those, two journals were not ailable in the Web of Science database, an additional two did not contain articles with ‘habitat’ in the title, and five were included in phase 1. We therefore chose the nine journals with the next highest impact factor to retain a total of 30 journals. A search of these 30 journals employing the same search terms used in phase 1 and Hall et al. yielded a total of 8,858 articles containing the word ‘habitat.’ To sample the same proportion of articles as phase 1, our goal was to analyze at least 177 articles in equal proportion from each of the journals chosen. In total, we analyzed 185 articles for phase 2.
The above exploratory analyses provided us with a starting point for which journals to include in the analysis. We selected a total of 50 journals that we felt were representative and where we expected the term ‘habitat’ to be used. Of these, 32 were selected (Table S1) and were all listed among the top 76 ecological and conservation journals based on their 2014 ISI ranking and Google ranks (Bradshaw, 2014). We included an additional 18 journals to cover some specific taxonomic groups not included in the former list (e.g., that specialized on fish) and also to include more specialized or regional journals not listed by Bradshaw (2014) but nonetheless contained articles describing the habitat associations of different taxa (Table S2). We reviewed 235 articles in total for phases 1 and 2, in addition to 80 articles that cited Hall et al. directly (see below) making a grand total of 315 articles.
2.1.2. Analysis of habitat‐related termsWe examined each article to see how the term ‘habitat’ was used and then identified all habitat‐related terms using the search tool in Adobe Acrobat. While each term (‘habitat’ or habitat‐related term) was assessed based on its use throughout the article, we focused particular attention on the initial few uses of the word within each article, assuming that the first few use(s) of the word would reflect how the term is applied for the remainder of the article and should therefore be consistent thereafter. To start rating how the term is used, we began our search in the Introduction since abstract word limits prevent authors from providing detailed definitions. In the Introduction, we expected authors to elaborate on and explain their meaning of the term so readers can understand the context for the remainder of the article. We scored the usage of the term ‘habitat’ according to the scoring system below.
On a scale of 1–4, we indicated whether the usage of the term ‘habitat’ was (1) consistently and correctly used throughout the article with an explicit definition similar or identical to the one given by Hall et al., 1997; (2) used acceptably and consistently but with no definition; (3) used inconsistently (weak usage) with no definition; or (4) used incorrectly—no definition supplied and ‘habitat’ was confused with other terms (e.g., ‘vegetation type’ and ‘land cover’). We then read through the article for other habitat‐related terms and scored them in the same way. Multiple uses of different terms were common in the articles we reviewed. For example, ‘habitat’ may be used as well as ‘habitat use’ or ‘suitable habitat.’
Some habitat‐related terms we found were not listed by Hall et al. For example, the term ‘habitat fragmentation’ has been used extensively post‐1997. We deemed it important to include such terms in our statistical analyses. However, we simplified many terms in the models as they essentially had the same meaning (see column 2 in Table S3). Of the terms in Table S3, 12 were used more than 10 times in the contemporary data. We combined all other habitat‐related terms (many of which had