RANKING SPECIES OCCURRENCES - A GENERIC APPROACH
Problems with Previous Rank Specifications
A Simplified Approach
Advantages of Generic Occurrence Rank Guidelines
GENERIC GUIDELINES FOR THE APPLICATION OF OCCURRENCE RANKS
A: Excellent Viability
B: Good Viability
C: Fair Viability
D: Poor Viability
E: Verified Extant
F: Failed to Find
NR: Not Ranked
HYPOTHETICAL EXAMPLES OF OCCURRENCES WITH SUGGESTED RANKS
Geoffrey A. Hammerson, Dale Schweitzer, Larry Master, and Jay Cordeiro
January 11, 2008
Element occurrence (EO, hereafter simply "occurrence") ranks provide a succinct assessment of the estimated viability (probability of persistence) of occurrences of a given species. They provide an estimation of the likelihood that, if current conditions prevail, a species occurrence will persist for a period of time. Because occurrence ranks are used to represent the relative overall “quality” of an occurrence as it currently exists, they are based solely on criteria that reflect the present status of that occurrence. These criteria can be broadly specified as “rank factors,” namely size (including population size and/or occupied area), abiotic and biotic conditions, and landscape context. Future threats should not be used to “downgrade” an occurrence rank, but ongoing events (e.g., successional changes, periodic unfavorable management) that result in inexorable degradation of occurrence quality should be considered.
Problems with Previous Occurrence Rank Specifications
NatureServe zoology staff recently reviewed all existing occurrence rank specifications for vertebrates and invertebrates. Many of the rank specifications superficially looked good; they often included quantitative criteria that seemed to appropriately "scientific." However, on closer inspection, these occurrence rank specifications generally turned out to be arbitrary, highly subject to change, of uncertain utility for distinguishing occurrence viability, and impractical or impossible to apply. When viewed as a whole, the occurrence rank specifications lacked a clear and consistent conceptual framework.
One of the major problems involved the use of specific measures of abundance, such as catch or observation rates, to distinguish occurrence viability. Catch rates are known to vary greatly with surveyor, methods, season, or other factors, and so the resulting data may not accurately reflect abundance. Consequently, it is generally unreliable to use such data for determining occurrence ranks. Strict adherence to specified sampling protocols might alleviate this problem, but in the real world such consistency rarely occurs.
Another problem with using quantitative occurrence rank criteria derives from the demographic characteristics of various plant and animal species. In many species, substantial variations in population size occur over periods of multiple years. If we establish precisely defined population criteria (e.g., >1,000 = B, < 1,000 = C) for ranking occurrences, we cannot simply use current population size to rank occurrences because the rank of some occurrences would change through the course of normal annual population fluctuations. Additionally, occurrences may be incorrectly ranked because the year or years sampled represent extreme conditions. One could circumvent this problem by using averages or modal or worst-year conditions, but such information is rarely if ever available. In fact, most of the occurrence rank specifications that were written in the past often appear to be unusable for most occurrences because the required information does not exist, rarely will be obtained, and sometimes cannot be obtained by known methodologies.
Even if abundance could be determined in a meaningful, repeatable way, we would still lack a secure scientific basis for specifying precisely defined, objective occurrence viability criteria. Conservation biologists have had enormous difficulty in determining or agreeing on quantitative population viability criteria for various taxonomic groups. And long-term population trends from the real world frequently are at odds with theoretical considerations.
Because of these factors, past efforts to establish useful, reliable, and stable occurrence rank criteria were largely unsuccessful.
A Simplified Approach
NatureServe scientists have concluded that elaborate or highly specific quantitative criteria are not required in order to rank species occurrences usefully for conservation purposes. Instead, categorical, qualitatively defined rank guidelines should be sufficient for most occurrence ranking. For a small minority of well-studied species or groups of species it may be possible to develop and employ meaningful, quantitative occurrence rank criteria, and the generic occurrence rank guidelines described in the following section do not preclude the use of more quantitative alternatives. In fact, for some particular species or species groups, the previously existing occurrence rank specifications were modified and retained. Many of these specifications basically offer suggestions as to how to apply the generic concepts to the species or group. When species- or group-specific occurrence rank specifications are available they should be consulted, and the ranker should decide whether these or the generic guidelines (or a combination) work best for the information at hand.
Advantages of Generic Occurrence Rank Guidelines
The generic occurrence rank guidelines for species address the problems mentioned above, and they circumvent additional problems. For example, for some species, the viability of populations of equal size may not be the same in two different regions or even in different habitats of the same region. The qualitative generic criteria deal effectively with species that exhibit substantial ecogeographical variations in demographic characteristics and eliminate the need to write multiple occurrence rank guidelines for single species. By focusing on probability of persistence, the criteria should work equally well for occurrences that attempt to represent populations as well as those that are arbitrary conservation units (e.g., occurrences of many migratory birds). Also, the generic criteria allow one to consider all of the variables that affect occurrence viability without having to anticipate them or incorporate them into the occurrence rank criteria. The generic occurrence rank guidelines are much less susceptible to change than are specific quantitative (but arbitrary) criteria. Additionally, the generic occurrence ranking guidelines make it likely that occurrences can be assigned to a rank other than "E," particularly if combination ranks (e.g., AB, AC) are employed.
Most importantly, we believe that the occurrence ranks derived from the generic criteria will be sufficient for conservation prioritization – for identifying a set of target occurrences for conservation action. They should allow users of the ranks to distinguish among occurrences with excellent viability, other robust (good viability) occurrences, occurrences with fair viability, and poor occurrences that have a high risk of extirpation.
Acknowledgements. We thank Marilyn Anions, Bill Bosworth, Sara Cairns, Nicole Capuano, Anne Chazal, Karen Cieminski, Leo Collins, Pat Comer, Todd Crabtree, Melissa Cullina, Jeremy Deeds, Phillip deMaynadier, Erik Endrulat, Don Faber-Langendoen, Mark Ferguson, Gretchen Fowles, Chris Frye, John Gamon, Kelly Gravuer, Steve Grund, Steve Hall, Ron Hellmich, Julie Holling, Dale Jackson, Amy Jenkins, Colin Jones, Jimmy Kagan, Doug Keinath, Harry LeGrand, Suzanne Mason, Larry Master, Kat Maybury, Roger McCoy, Dawn McKay, Sarah McRae, James Morefield, Bill Nichols, Mike Oldham, Leah Oliver, Tom Patrick, Eric Peterson, Bob Popp, Ken Popper, Rich Ring, Dan Salzer, Mike Schaefle, Matthew Schlesinger, Sue Schuetze, Tim Simmons, Tim Smith, Beth Swartz, Jeffrey Tash, Deborah White, Erin White, Steve Young, and others behind the scenes for their thoughtful comments on previous versions of the generic rank guidelines.
The generic approach to ranking species occurrences focuses on occurrences that are potentially rankable as A, B, C, and D. Many occurrences, such as those based solely on old museum records or on a recent observation with scant data, are not rankable as A, B, C, or D, but they could be ranked as E, H, X, U, F, or NR.
In the absence of known disturbance and with the habitat still extant, H is generally recommended for occurrences that have not been reconfirmed for 20 or more years, but for many short-lived insects a shorter interval may be appropriate, and for unusually stable habitats (like undisturbed caves), or for certain plants whose seeds may persist and remain viable in the soil for decades, a longer interval, up to 40 years, may be used. With very few exceptions, occurrences are to be regarded as H after 40 years without confirmation, even with no effort to locate the species. The time frame for H occurrences is necessarily arbitrary, and the values specified here should be regarded as generally appropriate but somewhat flexible rules. The professional judgment of the assessor should determine when resurveys with negative results have been sufficient in quantity and quality to warrant updating an occurrence rank from F to H or from H to X. Deviations from the suggested time frame should be explained in the EO RANK Comment field.
In some cases, H may indicate occurrences with imprecise locational information such that it may be difficult or impossible to determine whether subsequent observations are of the same occurrence; many of these occurrences may remain H indefinitely. Nevertheless, occurrences with imprecise locational information sometimes may be mapped using an appropriate and reasonable indication of the degree of locational uncertainty.
Note that certain combination ranks (i.e., AB, AC, BC, and CD) are encouraged and should be used to indicate the range of uncertainty regarding the appropriate rank for an occurrence. In fact, due to pervasive limited information about most occurrences, the appropriate rank for most occurrences will be a combination rank. It may be relatively easy to determine an appropriate rank by eliminating clearly inappropriate ranks (e.g., an occurrence is clearly not an A nor a D, so it's BC; or an occurrence appears to be viable and is clearly better than a D, but little else is known, so it's AC). The ranks AD and BD are uninformative regarding conservation value so their use is strongly discouraged; generally E should be used instead.
Occurrence ranking benefits from multiple opinions and may be accomplished most effectively in an "expert's workshop" setting. Occurrence ranks are best determined by persons who have a good understanding of the population characteristics of the species or who at least have good basic knowledge of the biology and ecology of the group of organisms to which the species belongs. Such knowledge allows the ranker to make a good forecast about the viability of a particular occurrence. The rationale for each rank should be recorded in the EO Rank Comments field in Biotics.
For purposes such as monitoring the response of an occurrence to management actions, habitat mitigation, and ecological restoration, occurrence ranks, whether qualitatively or quantitatively defined, will be insufficient. Instead, efforts will need to be made to identify the key ecological attributes (subcomponents of the rank factors) that play an important (driving) function in the viability of the species. For each key ecological attribute, specific indicators or metrics can then be selected that will reflect changes in the viability of a species. For example, “reproductive success” is an indicator of the “reproduction and health” key ecological attribute (a subcomponent of condition), and the specific metric is “number of fledged young.”
HYPOTHETICAL EXAMPLES OF OCCURRENCES WITH SUGGESTED RANKS
Colonial seabird: occurrence consists of 5 adjacent islands, all of which are protected; nesting population recently was estimated at 400,000 pairs, with no evidence of decline compared to estimates made 25 years ago. Suggested rank: A. An occurrence with these highly favorable characteristics is very likely to persist for the foreseeable future (i.e., at least 20-30 years) in its current condition or better.
Pool-breeding ambystomatid salamander of forested landscapes: occurrence is represented by several breeding sites; pools contain dozens to hundreds of egg masses; breeding sites are surrounded by an extensive tract of mature forest. Suggested rank: AB. This occurrence, which likely represents a metapopulation, can be expected to exhibit at least good viability. Further information on occurrence size, condition, and landscape context might allow one to distinguish between A or B.
Pond-breeding odonate: occurrence encompasses a complex of several neighboring wetland patches of various sizes of up to a few hectares; species was first discovered in this location 75 years ago; recent quick surveys yielded multiple adults and exuviae in nearly every patch; wetland and adjacent upland habitat are protected and not subject to pesticide applications. Suggested rank: AB. This occurrence, which likely represents a metapopulation and for which long-term persistence has been documented, can be expected to exhibit at least good viability. Further information on population size, habitat condition, and landscape context might allow one to distinguish between A or B.
Small mammal of arid basins: recent random surveys indicate that the occurrence extends over at least 100 square kilometers; habitat appears to be relatively unaltered, stable, and compatibly managed; population size and density are unknown, but the species is readily detected. Suggested rank: AB. This evidence suggest an at least good probability of continued occurrence persistence in the present condition or better. Nothing here appears to compromise the viability of the occurrence.
Snake of rocky desert mountains: recent 1-day survey found 9 individuals scattered over 2 patches of talus habitat in a single remote mountain range (not readily accessible by road); suitable habitat encompasses several additional adjacent patches totaling a few square kilometers of basically undisturbed talus. Suggested rank: AB. Snakes in such habitats generally are difficult to observe in quantity, so seeing this many in one day suggests an ample population. Evidence indicates that the habitat is not subject to much disturbance and can be expected to persist in at least good condition for the foreseeable future.
Tadpole shrimp of ephemeral pools: known occurrence consists of a single 1-hectare basin in a landscape altered only by seasonal grazing of livestock, which has occurred over several decades; 25 specimens were collected from the site 20 years ago; last year a single specimen was collected (abundance unknown); the basin was dry this year; unknown whether or not there are any nearby occupied pools. Suggested rank: AC. Tadpole shrimp eggs can persist for years in a dry basin, and these shrimp can be dispersed among basins by birds and other mobile animals. In a particular location, tadpole shrimp may be present for one or more years, then absent for a variable number of years and, when present, they may range from scarce to abundant. Presence of tadpole shrimp in a basin in multiple years over a wide time frame suggests that the basin will continue to support a population (although perhaps irregularly) for the foreseeable future, if current conditions prevail, so the rank--even for this pool alone--is better than D. Further information on population size and landscape context (e.g., presence of nearby occupied pools) might allow one to define the rank more precisely. In particular this pool might or might not prove to be merely a portion of a larger A or B quality metapopulation.
Seed-banking annual plant of river sand bars: isolated relict species; population of many 10,000s of mature individuals occupies much of a 90-km stretch of river corridor; abundance fluctuates greatly from year to year; population appears to be negatively affected by hydrological and sediment transport alterations associated with present dam operations. Suggested rank: BC. The occurrence exhibits good size characteristics but current circumstances indicate that a small or large decline might be occurring. Though the occurrence may persist for the foreseeable future in good or excellent condition, it is also possible that current conditions are resulting in a major decline that could significantly jeopardize occurrence viability. The combination rank reflects this substantial uncertainty.
Pool-breeding amphibian of forested landscapes: occurrence consists of a single, isolated breeding pool and surrounding, mostly intact uplands; 20 percent of the upland area contains rural residential development; single recent survey found several hundred egg masses in the breeding pool. Suggested rank: B. Evidence indicates a substantial population that is, however, confined to a single pool, and the upland habitat has been somewhat degraded, so the occurrence does not meet the criteria for an A rank. On the other hand, an occurrence such as this is likely to persist if current conditions prevail, hence it is not a D occurrence. Given the large number of egg masses, the occurrence has good prospects for persistence and appears to be better than a marginal or C occurrence.
Turtle of freshwater wetlands: population of approximately 80-100 mature individuals, plus various age classes of immatures, occupies most of a 15-hectare wetland; no indication of a major increase or decline but few data are available for trend estimation; wetland is in a wildlife sanctuary and protected from outright destruction. Suggested rank: B. Ideally, A-ranked occurrences should be represented by thousands of adults, but in the case of long-lived species such as turtles, a population of hundreds of adults likely would exhibit excellent viability. This occurrence does not exhibit optimal population size for excellent viability, yet the prospects for long-term persistence without additional protection or management appear to be very good, so the rank should be less than A and higher than C.
Tiger beetle of coastal sand dunes: cursory surveys during the appropriate season consistently yield observations suggesting that a population of at least a few hundred adults (and undetermined numbers of larvae) occupies a suitable habitat patch of 25 hectares in a national park; museum collection records for the area extend back at least 50 years; habitat has natural processes largely intact but is partially encroached on by a small asphalt parking lot that has been present for at least 30 years, and a small portion of the habitat is subject to light foot traffic by humans. Suggested rank: BC. Nonoptimal habitat conditions, and evidence of long-term persistence of a currently significant population, eliminate A and D ranks, respectively. Further information on population size and landscape context (e.g., presence of nearby occupied patches) might allow one to define the rank more precisely.
Small mammal of coastal dunes: occurrence extends along 3 kilometers of protected coastal dunes; surrounding areas are heavily developed (hotels, parking lots, etc.); species was first documented at the site 30 years ago; each of several live-trapping efforts in recent years yielded at least a few individuals per 100 trap-nights. Suggested rank: BC. Limited size and less than highly favorable landscape context eliminate an A rank, and documented persistence over at least 30 years indicate that the occurrence rank is better than D. Further information on population size is needed to determine whether the rank should be B or C.
Perennial herb of calcareous habitat: occurrence restricted to limestone barrens on isolated peninsula of 363 hectares (864 acres) on marine coast; outlier population from main range (~1,600 kilometers away); past quarrying stopped with protection as provincial natural reserve in 1998; harsh environment resulting in short growing season but possible annual seed set and seeds dispersed by wind; no recent surveys for numbers of plants but observed sporadic distribution throughout peninsula. Suggested rank: BC. Available information indicates that this occurrence does not exhibit optimal or highly favorable size, condition, and landscape context, and it does not suggest an occurrence with a high probability of extirpation, so the A and D ranks can be ruled out. Further information on population size (100s?, 1,000s?) should allow one to define the rank more precisely.
Butterfly associated with a wetland food plant: populations are naturally fragmented throughout the range; species has been known from this site for more than 100 years, and the size and condition of the site appears to have changed little over the past several decades; population occupies a small patch of wetland (stable bog or fen, about 1 hectare); nearest occupied habitat is 50 km away; 5-30 adults seen during recent 2-hour surveys during peak flight season (note that this is the number observed, not the total number of adults present); wetland is on a nature preserve with good prospects for appropriate habitat management. Suggested rank: C. The small size of this occurrence precludes the A and B ranks, while long-term persistence and stable conditions eliminate the D rank.
Passerine bird associated with old field conditions: recent surveys (3-5 years old) found 25-30 singing males during the nesting season; occupied habitat patch is 100 hectares, surrounded by forest and residential development; reforestation is slowly occurring; land is divided among several private owners. Suggested rank: CD. Certainly the size, condition, and landscape context are not optimal or highly favorable, so it is not an A occurrence. If existing conditions (including ongoing reforestation) prevail, the occurrence is likely to undergo habitat degradation and thus not to maintain its current condition, so a rank of B can be eliminated. Appropriate management is at least plausible and probably could maintain this population for the foreseeable future in its current condition or better, so the C rank cannot be eliminated. Because persistence of this occurrence for the foreseeable future depends on appropriate management that may not occur (in which case the occurrence may be extirpated), a rank of CD is warranted.
Long-lived perennial plant of wetlands: occurs only on the edges of wet sphagnum bogs; recent survey found 19 plants, 60% in leaf only, 40% in fruit; 40% seedlings, 20% immature, 40% mature; occupied area is less than 1 hectare; in 2005 there were three main patches of plants, and at least one subsequently was destroyed by bulldozing to expand a cranberry bog; two remaining patches are extant but need follow-up survey. Suggested rank: C or CD. The small size of this occurrence, and its poor condition and landscape context, indicate that A or B ranks are not appropriate. Under current conditions, the probability of persistence seems low, so the occurrence may warrant a rank of D, but evidence of ongoing reproduction suggests that a C rank might be appropriate, especially if only a small proportion of the remaining plants were destroyed and there is a reasonable chance that the remaining habitat can be protected and appropriately managed.
Freshwater mussel of riverine habitat: extensive searching by experienced mussel biologists yields only a few mature individuals and no younger age classes; despite several surveys, no other occupied habitat patches have been found in this river in recent decades. Suggested rank: CD or D. This occurrence might not be completely devoid of viability; some mussels have a long life span, and it is plausible that conditions could change for the better. For example, riverine fish faunas are notoriously dynamic, and arrival of a new host fish could allow a renewal of successful reproduction.
Perennial plant of wooded landscapes: historical records indicate that the species was fairly common in the several hectares encompassed by the occurrence; recent targeted surveys yielded only a few individuals, all browsed by deer; deer population in the area is large and unlikely to decrease. Suggested rank: CD. Clearly not an occurrence with excellent or good estimated viability (too small, poor condition and landscape context), so not an A or B occurrence. Construction of a deer exclosure might prevent total loss of the remaining population, but otherwise the occurrence has a high probability of extirpation in the foreseeable future, so the appropriate rank is CD.
Passerine bird associated with grassland habitat: in recent years 2-3 singing males have been present in May in a 5-hectare hayfield surrounded by residential development and young forest; a few fledglings were observed in July in one of the years. Suggested rank: D. This information substantiates the existence of a legitimate occurrence, but very small size and poor landscape context indicate a high probability of extirpation in the foreseeable future. Feasible management could not do much to improve the prospects for persistence.
Short-lived perennial desert plant, emergence dependent upon spring rains: occurrences and abundance vary widely from year to year in response to variations in precipitation): 78 plants found in intensive survey in 1989, 15 plants observed in brief 1990 survey, 4 plants in 1991 late-season survey, 5 plants observed in 1998 at peak season with fairly intensive surveys after favorable precipitation conditions; site is seasonally grazed by sheep, which decreases the abundance of the plant, and change in management is unlikely. Suggested rank: D. The small size and poor condition of this occurrence immediately rule out A and B ranks. Available data suggest a significant declining trend and high probability of extirpation if current conditions prevail.
Pond-breeding toad: occurrence encompasses a single pond and surrounding uplands; pond is protected but most habitat around the pond has been converted to residential development over the past decade; recent surveys in appropriate season yielded a total of 3 egg masses in the pond. Suggested rank: D. Extremely small size and poor condition and landscape context make it highly unlikely that this occurrence will persist, let alone ever become a viable occurrence.
Remnant grassland butterfly: small habitat scrap of 4 hectares of unmanaged dry native grassland, surrounded by residential development and with a cleared party spot with scraps of burned wood near the middle; absentee owner; neighbors recently started using part of the area for motorcycle recreation; the area mark-release-recapture studies estimated 30, 40, 35 and 50 adults within the past decade; 20 years ago there were three other comparable occupied habitats within 2 kilometers but now none are known within 50 km; populations of this species commonly decline (often in dry years) to as low as 10-20% of the mean during a 10-20 year period; according to old timers and anecdotal literature from several states colonies often suddenly appeared or disappeared; data indicate that immature stages are usually reduced by 90-100% following fires; a huge majority of suitable habitat patches in the state no longer support the species. Suggested rank: D. With a mean of about 39 adults this apparently stable population is expected to fall to around 4-8 adults within 10-20 years from just routine fluctuations and certainly could not recover from two consecutive bad years, and might not from one. Furthermore, fires are nearly inevitable considering current context and uses and are most likely during already bad (dry) years. This occurrence has a high risk of extirpation during the next 20-30 years, and it appears to be implausible that the occurrence will be restored to at least fair viability through appropriate management.
Wetland turtle: Information from 10 years ago established an occurrence in a difficult-to-access wetland; this occurrence was then regarded as a viable occurrence and was ranked BC. Aerial photos taken last year indicate extensive alteration of both the wetland and upland habitat adjacent to the wetland over the past decade. Last year, a reliable person observed and photo-documented a single adult turtle on the site, but other recent information about the population is not available. Suggested rank: E. This information indicates that the rank might be as high as B (if the turtle population has not been negatively affected by the habitat changes) or as low as D (e.g., if the only suitable nesting habitat has been destroyed). An occurrence rank of BD is uninformative for conservation purposes. Because available information is too limited to determine the estimated viability of this occurrence, the occurrence should be categorized only as extant.
Perennial herbaceous plant of upland forests: A botanist collected the species last year and deposited a specimen in an herbarium, but nothing else is known about the occurrence. Suggested rank: E. If we assume the collected specimen was not the last remaining individual, we can consider the occurrence to be extant. Further information is required in order to categorize the occurrence status more precisely.
Wetland turtle: Collection information from 50 years ago established an occurrence in a difficult-to-access wetland; this occurrence was then regarded as a viable occurrence and was ranked B. No further information about the turtle population is available. Recent aerial photos of the area indicate extensive alteration of both the wetland and upland habitat adjacent to the wetland over the past two decades. Suggested rank: H. This occurrence needs to be surveyed in order to verify that it still supports a turtle population and to determine current occurrence viability.
Herbaceous plant of interior, stabilized sand dunes: A university museum includes a specimen collected in the 1940s from a vague location in a dune area, which still exists and is largely intact. Nothing else is known about the occurrence, and the area has not been well surveyed. Suggested rank: H. This information is too old to support an occurrence rank other than H. The occurrence needs to be surveyed in order to verify that it still supports a population and to ascertain occurrence viability.
Wetland orchid: In the 1960s a small population of a wetland orchid was documented on a site on private land that was formerly accessible through landowner permission. Today the site is no longer accessible due to an uncooperative new landowner. Suggested rank: H. Because this occurrence is based only on old information, the appropriate rank is H. The occurrence rank could be updated in the future if access to the site improves.
Cave invertebrate: Surveys 50 years ago found several individuals in a cave pool. Subsequently, the opening used for access to the cave was sealed and likely will remain so indefinitely. No further information on the occurrence is available. Suggested rank: H. Because this occurrence is based only on old information, the appropriate rank is H. In the future, improved remote sensing/survey techniques might allow the rank to be updated. If prior to sealing of the cave entrance the occurrence had been ranked A, B, C, D, or E based on recent surveys, that rank would still apply until the time frame suggested invoking the H rank, unless there is a significant probability that cave sealing might have affected the viability of the occurrence, in which case the appropriate rank would be H.
Rattlesnake that uses communal hibernacula: Museum specimens and a published account from 15 years ago document the existence of a denning population somewhere on remote Snake Mountain. Last year a well-timed, one-day survey failed to find any rattlesnakes, but much suitable, difficult-to-access habitat was not searched. Suggested rank: F. The recent brief survey does not preclude the possibility that the occurrence is still extant, so the appropriate rank is F.
Toad that uses ephemeral bodies of water for breeding: Surveys in the 1970s found a several breeding toads in and around a shallow pool in a prairie landscape then used as cattle pasture. Today the site has been completely replaced by commercial development, and the former breeding site has been filled and covered with asphalt. Recent searches and examination of recent aerial photographs found no evidence of suitable breeding habitat anywhere within a few kilometers of the previous breeding site, and there are no other records of the occurrence of this species in the area. Suggested rank: X. This evidence indicates that it is appropriate to regard the occurrence as extirpated.
Grassland songbird: Surveys in the late 1970s found a few nesting pairs around the margins of a large pasture. Since then, the species has not been detected in that area or in other locations in the vicinity, despite surveys at the appropriate time by expert birders in each of the past three years. Suggested rank: X. This evidence is sufficient to regard the occurrence as extirpated. If the evidence of extirpation were not so conclusive (i.e., there is a reasonable probability that the species is still there but was not detected), then the appropriate rank would be H.
Spring-dwelling fish: Surveys done 5 years ago documented a robust population in a natural spring-fed pool; on this basis the occurrence was ranked B. Examination of aerial photos taken last year and discussions with local residents indicate that the pool has been significantly modified in size and shape, but new fish surveys have not been conducted. Suggested rank: NR. Available information is insufficient to determine whether the changes have been positive, negative, or neutral; at one extreme, the fish may have been eradicated (perhaps in conjunction with the introduction of a predatory species in the modified habitat), or perhaps the changes have benefited the native fish population. Because of the altered conditions, the basis for the previous rank of B no longer applies, and the status of the occurrence needs to be redetermined; meanwhile, the rank should be updated to NR.