U.S. FOREST SERVICE REGION ONE

BLACK-BACKED WOODPECKER ASSESSMENT

Mike Hillis, National Fire Plan Cohesive Strategy Team

Amy Jacobs, Flathead National Forest

Vita Wright, Aldo Leopold Institute

Reviewed by:

Jack Losensky, Ecological Services

Lisa Bate, Wildlife Biologist

Dr. Sallie Hejl, Texas A&M University

Jennifer Taylor, Idaho Panhandle National Forests

Dave Lockman, Bitterroot National Forest

 

INTRODUCTION

Black-backed woodpeckers occupy forested habitats that contain high densities of recently dead or dying trees that have been colonized by bark beetles and woodborer beetles (Buprestidae, Cerambycidae, and Scolytidae).  These beetles and their larvae are most abundant within burned forests.  In unburned forests, bark beetle and woodborer infested trees are found primarily in areas that have undergone natural disturbances, such as wind-throw, and within structurally diverse old-growth forests (Steeger and Dulisse in press, Bull et al. 1986, Goggans et al. 1987, Villard 1994, Hoffman 1997, Weinhagen 1998).

In Montana, black-backed woodpeckers are most abundant in recent stand-replacing burns.  In northern Idaho, where burns have been largely absent for the last 60 years, black-backed woodpeckers are found amid bark beetle outbreaks, although not at the densities found in post-burn conditions in Montana.  Threats to this species, like most woodpecker species in USFS Region One, include snag removal and a lack of snag recruitment.  The greatest concerns for this species, however, are decades of successful fire suppression and salvage logging targeted at recent bark beetle outbreaks.   

Research conducted in western Montana (Hutto 1995, Caton 1996, Hitchcox 1996, Hejl and McFadzen 2000, Powell 2000) indicates that black-backed woodpeckers are strongly associated with recent burns.  Goggans and others (1987), Murphy and Lehnhausen (1998), and Powell (2000) concluded that high densities of bark beetles and woodborers, which occur at high densities in burned forests, provide a unique short-term foraging niche for black-backed woodpeckers.  Caton (1996) concluded that black-backed woodpeckers occupy burns for one to six years following the burn, with peak densities occurring at three and four years following the burn.  This corresponds to the timeframe that bark beetles and woodborers are present in the highest densities (DeNitto et al. 2000).  Powell (2000) concluded that while black-backed woodpeckers forage in areas infested with both bark beetles and woodborer beetles, they strongly select woodborers over bark beetles while foraging.  In burned forests, black-backed woodpeckers prefer large, thick-barked conifers including ponderosa pine, Douglas-fir, and western larch, over small, thin-barked species such as subalpine fir, Englemann spruce, and lodgepole pine (O’Connor and Hillis 1999, Hejl and McFadzen 2000).  Powell (pers comm.) concluded that large, thick-barked trees are more likely to retain green cambium after a severe fire, which provides attractive habitat for bark beetles and woodborers.  The cambium of thin-barked trees like lodgepole pine or subalpine fir, conversely, is more likely to be charred by severe fires, making those thin-barked conifers generally less suitable for bark beetles, woodborers, and black-backed woodpeckers.  Monitoring done after the 2000 Lolo and Bitterroot Fires (Monson and Boniecki 2002), however, suggests that thin-barked conifers should not be totally ignored as potential contributors to black-backed woodpecker habitat.  Monson and Boniecki found low levels of black-backed woodpecker foraging in lodgepole pine that had been killed by low severity fires or secondary beetle attacks.   

Black-backed woodpeckers also occur in unburned landscapes Bull et al.1986, Goggans et al.1987, Bate 1995, Hoffman 1997, Weinhagen 1998, Steeger and Dulisse in press, Taylor unpublished data).  Taylor’s observations of black-backed woodpeckers in unburned forests in northern Idaho suggest that they may occur at substantially lower densities in unburned forests, but no rigorous comparisons between black-backed woodpecker densities in burned and unburned forests have been done.  Hutto (1995) hypothesized that black-backed woodpeckers reproduce at source reproductive levels in burns, but may drop to sink reproductive levels in the intervening periods between large burns. 

Although undocumented, it is thought that black-backed woodpeckers are a relatively short-lived species (similar to other Picoides woodpeckers) and probably only live six to eight years (Dixon and Saab 2000).  This suggests that not only are substantial acreages of burns important to the species, but that burns are needed at relatively short intervals.  Losensky (2002a) concluded that before 1900, very large burns of 1910 magnitude occurred at the rate of one to two per decade within Region One. 

Black-backed woodpeckers appear capable of migrating large distances to exploit rich foraging resources such as those that occur in recent burns or other large-scale natural disturbances (VanTyne 1926, West and Speirs 1958, Bock and Bock 1974, Yunick 1985). It is unknown, however, how far black-backed woodpeckers will migrate to occupy burned forests. In a study that compared percent occurrence of black-backed woodpeckers in burned, mature, and old-growth coniferous forests, black-backed woodpeckers were not detected 50 km away from large (200 ha) stand-replacement burns (Hoyt 2000) although they were detected in old-growth spruce forests 75 and 100 km away. Hoyt (2000) hypothesized that black-backed woodpeckers were drawn away from the unburned forests within a 50 km radius of the burn.

The black-backed woodpecker’s extraordinary degree of dependence on recent fires may not be consistent across the range of this species.  Research conducted in Montana (Hutto 1995, Caton 1996, Hitchcox 1996, Hejl and McFadzen 2000, Powell 2000), suggests Black-backed woodpeckers may require recent burns for long-term survival.  Conversely, in Oregon, Goggans and others (1987) found high densities of black-backed woodpeckers within extensive mountain pine beetle outbreaks that occurred in the absence of fires. This assessment will focus on the recent Montana research.  Therefore, the findings will be most appropriate for the Montana portion of Region One. 

Fire exclusion has potentially reduced the amount of black-backed woodpecker habitat (Hutto 1995, Powell 2000).  The percentages of forests in “stand-initiation” age classes, or young stands that originated from wildfires, have declined substantially since the onset of successful fire suppression (Losensky 1993, Losensky 1995, Hessburg et al. 1999, Wisdom et al. 1999, Losensky 2002,).  In addition, salvage logging often further reduces the suitability of those burned sites for black-backed woodpeckers (USDA 1995, USDA 1999).  Hitchcox (1996), Saab and Dudley (1998), and Hejl and McFadzen (2000) found very few black-backed woodpeckers after salvage logging, even when many of the fire-killed trees were retained.  This assessment will evaluate the degree to which fire suppression has reduced the availability of fire-killed forests for black-backed woodpeckers.  The effects of salvage logging are not evaluated in detail.

METHODS

For highly specialized species such as black-backed woodpeckers, substantial reductions in habitat are presumed to be accompanied by an increased risk that the species will decline to nonviable population levels.  Consequently, an appropriate measure of status is to compare the existing habitat against that which was available in historic periods.  Landres and others (1999), Wilds and others (1999), and Jensen and others (2000) emphasize the need to understand historic habitat availability and natural disturbance regimes when making such comparisons.  For black-backed woodpeckers, historically available habitat is assumed to be represented by that period before fire suppression became effective, or prior to 1900 for this analysis.

Fire suppression clearly was effective by 1940, based on summaries of annual acres burned in Region One (Jones and Barrett in press).  For this analysis, existing habitat conditions are based on the acres of fires burned from 1940 to the present. 

Existing Habitat.  The acreages burned by wildfires since 1940 are summarized annually (Jones and Barrett in press).  A single wildfire can provide black-backed woodpecker habitat for up to six years (Caton 1996), and black-backed woodpeckers migrate substantial distances to exploit fires (Hoyt 2000).  Therefore, it is assumed that high densities of black-backed woodpeckers can be sustained with fairly large fires at random locations across Region One, as long the total acres burned approximates the level that burned historically, and if the current frequency of fires approximates the historical frequency.  Losensky (pers comm.) concludes that the acreage of naturally occurring fires that occurred before fire suppression varied substantially from one year to the next.  In order to recognize this natural variation in historic fire occurrence and the black-backed woodpecker’s ability to survive several low-fire years in a row, existing habitat was determined by totaling the acres burned by 6-year intervals from 1940 to 2000.  For instance, acres were totaled for the period 1940 to 1945, 1946 to 1951, etc.  (The acreages by 6-year period, which comprise existing habitat, are summarized in RESULTS, Table 4.)

The acreages displayed below in Table 1 include burned acres of all severities.  Consequently, some recorded fires include burned acres that presumably did not burn at high enough severities to produce habitat for black-backed woodpeckers.  Additionally, some “islands” within large fires may not have burned at all, but are still included in the acres-burned data.  As a result, existing habitat measured by this method is higher than actual.  Assuming that there is less existing habitat than in historic periods, departures from historic conditions will be underestimated.  Post-burn monitoring conducted on the Bitterroot National Forest (Kamps and Gunn 2002) suggests the overestimation of existing habitat may not be as high as expected.  Kamps and Gunn’s observations of bark beetle activity after the 2000 Bitterroot Fires suggest many surviving trees within low severity burns and trees within unburned islands often succumb to secondary bark beetle mortality anyway, which somewhat reduces the potential discrepancy between fire perimeter data and the actual acres of stands that succumbed to moderate and high severity fires.

Historic Range of Variability (HRV).  Fire Regimes in Region One are defined by severity class and fire return interval (Jones and Barrett, in prep.) and mapped across the Region.  Table 1 illustrates the historic fire regimes and fire return intervals within potential black-backed woodpecker habitat.

Table 1.  Fire regimes and fire return intervals in potential black-backed woodpecker habitat in USFS Region One (excluding east part of the Custer National Forest).

* The 10^th percentile fire return interval is the point at which 90% of all recorded fire return intervals, plotted on a frequency distribution graph (often portrayed as a bell-shaped curve), are longer than that fire return interval.  This represents the minimum fire return interval that might be expected after long-term drought or hotter-than-normal climatic perturbations.  The 90^th percentile fire return interval is the point at which 90% of all recorded fire return intervals, plotted on a frequency distribution graph, are shorter than that fire return interval.  This represents the maximum fire return interval that might be expected from long-term wetter or colder-than-normal climatic perturbations (i.e the little ice age).  The combination of both the 10^th and 90^th percentiles represent the normal range or extremes of variation (or broad peak of the bell-shaped curve) that should be considered when comparing existing to historic conditions.

Potential Vegetation Types (PVTs) define the potential vegetation and are mapped across the Region.  Not all PVTs are black-backed woodpecker habitat.  Wildfires within the climax ponderosa pine PVT are almost always nonlethal (Jones and Barrett in prep) under natural conditions.  Since the climax ponderosa pine PVT is also very open, nonlethal fires will not result in adequate tree mortality and beetle densities to support black-backed woodpeckers.  For the same reason, the whitebark pine/limber pine/alpine larch PVTs also have little potential to support black-backed woodpeckers under natural conditions.  All other forested PVTs have the potential to support high densities of black-backed woodpeckers, if burned at high enough severity (Caton 1996, Hitchcox 1996, Hejl and McFadzen 2000).

The mix of fire regimes in Region One within potential black-backed woodpecker habitat   was determined by the following three steps:

1)   A GIS layer of Forested PVTs, excluding climax ponderosa pine and whitebark pine/limber pine, was overlaid with;

2)  A GIS layer of fire regimes across Region One; and

3)  Where both layers overlapped, the acres for each fire regime were totaled in a summary table.  This distribution of acres by fire regime within potential black-backed woodpecker habitat is summarized in Table 2.

Table 2.  Fire Regime acres in potential black-backed woodpecker habitat in USFS Region One.

Fires occurring within Nonlethal (NL) and Mixed Severity 1 (MS1) fire regimes seldom have mortality levels exceeding 50% (Barrett and Jones in prep).  Since these fire regimes typically also occur in ponderosa pine PVTs (not potential black-backed woodpecker habitat), they can be ignored for this analysis (Kotliar et al. in press).

Calculations for determining HRV.  The following simple formula can derive the mean burned acres up to 6 years following a wildfire that would have been available in Region One in pre-fire-suppression periods:

Mean acres = Acres within the fire regime  x  6

Fire return interval

Considering this formula hypothetically, if we have a million acres of potential black-backed woodpecker habitat, in which stand-replacing fires always occur at 100-year intervals, the following calculation would determine how much of the landscape would have burned within the last 6 years:

1,000,000 acres (6 years)  =  60,000 acres

100 years

Consequently, in our 1,000,000-acre hypothetical landscape, where fires always occurred at 100 year intervals, the 1- to 6-year timeframe (represented in the formula by the 6-year interval) would be present on about 60,000 acres, or 6% of the landscape.

On a real landscape, however, one must consider several other factors.  First, a million acres is too small a landscape to characterize fires in this manner.  For instance, one large fire of 1910 magnitude could burn the entire 1,000,000 acres resulting in 100% of the landscape in black-backed woodpecker foraging habitat--not the 6% predicted.  Secondly, unlike our example in which fires always return every 100 years, real fire return intervals are highly variable and never occur at exact intervals.  Consequently, calculations based on mean fire return intervals only make sense when viewed over long periods of time.  Lastly, stand-replacing fire regimes also include non-lethal or mixed-severity intermediate burns that add more variability to the outcomes than is expressed in the mean fire return interval data (Losensky pers comm.).  Consequently, for the HRV calculations to be meaningful, the following considerations should be applied to any determination of HRV:

1)  Landscapes should be huge for making HRV characterizations;

2)  Comparisons of current habitat to the HRV should be made only in very relative terms;

3)  When the level of existing habitat is substantially lower than the HRV, such comparisons should only be considered meaningful when the departure is several times lower than the HRV; and

4)  Since fire return intervals occur as a range of conditions, departures should also be compared against the 10^th percentile, 90^th percentile, and mean data to reflect the normal range of fire return intervals.  

A complicating variable is that black-backed woodpeckers only exploit burns that occur in stands where the average tree size is 9” dbh or greater (Caton 1996, Powell 2000).  Burns within stands with smaller-sized trees do not support bark beetles and woodborers at densities preferred by black-backed woodpeckers.  The age at which dominant trees in a stand reach 9” dbh is highly variable, but Losensky (pers comm.) and Applegate (pers comm.) suggest 70 years is a reasonable estimate considering the range of Potential Vegetation Types (PVTs) that comprise potential black-backed woodpecker habitat.  Therefore, the following factor is added to the previous equation to adjust the total acres in the 1- to 6-year post-burn timeframe that would have occurred in stands > 9” dbh:

Mean fire return interval  -  70

Mean fire return interval

In this case, if the mean fire return interval was 100 years, only 30% of all wildfires ((100 -  70 ) / 100) would have occurred after the stand had reached 9” dbh.  The factors that will be used in this assessment, based on actual mean fire return intervals by fire regime, are displayed for each fire regime in Table 3.

Table 3.  Factors used in the formula to reflect the probability that naturally occurring wildfires occurred in stands > 9” dbh.

Another complicating factor is that black-backed woodpeckers only exploit fires that burned at moderate or high severities.  Low-severity burns seldom contain the high density of bark beetles and borers needed by black-backed woodpeckers (Powell 2000).  In stand-replacing fire regimes, no adjustment is needed since fires in these fire regimes usually burn at moderate to high severity (Jones and Barrett in prep).

In mixed fire regimes, however, fires can include the full range of low to high severity.  Consequently, for the mixed fire regime MS2 (see Table 1), a factor is needed to reflect the probability that fires occur at moderate or high severity, and to exclude fires that burn at low severity.  Jones and Barrett (in prep) indicate that tree mortality within the MS2 fire regime ranges from 50% to 80%.  Fires that only kill 50% of the stand are substantially less preferred by black-backed woodpeckers based on findings by Hejl and McFadzen (2000) and observations by O’Connor and Hillis (2001).  Tree mortalities of 80%, however, usually provide habitat to support high densities of black-backed woodpeckers.  For this analysis, within fire regime MS2, a factor of 0.5 will be added to the formula to reflect the low range of historical or naturally occurring fire severities.  A factor of 0.8 will be added to the formula to reflect the high range of naturally occurring fire severities.  A factor of 0.65 will be used to reflect mean severity. 

Combining all these variables, we used the following formula to predict the mean HRV:

Acres MS2 (6) (.04) (.65)  +  Acres SR1 (6) (.47)  +  Acres SR2 (6) (.71)   =   Mean HRV

      73                                    133                                244

To reflect the minimum value for the HRV, the mean fire return intervals were replaced with the 90^th percentile fire return intervals (since the longest fire return intervals result in the lowest predicted burned acres), and the mean tree mortality factor for MS2 of 0.65 was substituted with the low tree mortality factor of 0.5.

Substituting these factors, we used the following formula to predict the low range HRV:

Acres MS2 (6) (.04) (.5) +  Acres SR1 (6) (.47)  +  Acres SR2 (6) (.71)  =  Low range

               117                                 180                                 325                        HRV

To reflect the maximum value for the HRV, the mean fire return intervals were replaced with the 10^th percentile fire return intervals (since the shortest fire return intervals result in the highest predicted burned acres), and the mean tree mortality factor for MS2 of 0.65 was substituted with the high tree mortality factor of 0.8.

Substituting these factors, we used the following formula to predict the high range HRV:

Acres MS2 (6) (.04) (.8) +  Acres SR1 (6) (.47)  +  Acres SR2 (6) (.71)  =  High range

               43                                    96                               200                          HRV

The calculations of HRV using these formulas are included in Appendix A.  

RESULTS

Existing Habitat.  The actual acreages of burns totaled by 6-year period, available as black-backed woodpecker habitat in Region One since 1940, are summarized in Table 4 (Jones and Barrett in press).  Figure 1 displays the same results in a bar chart to display relative comparisons.

Table 4.  Summary of available black-backed woodpecker habitat created by wildfires in USFS Region One since 1940.

Figure 1.  Summary of available black-backed woodpecker habitat created by wildfires in USFS Region One since 1940.

Note that even when burned acres are totaled by 6-year period, there is enormous variability from one 6-year period to the next.  For instance, note that in Table 4 the acres burned in the 1988 to 1993 period was over 100 times greater than the acres burned in the 1946 to 1951 period.  Also, note in Figure 1 that the total acreages from 1940 to 1987 are modest.  Note especially in Figure 1, the huge increase in acres burned in the periods from 1988 to 2000.

Predicted HRV.  Based on the formulas described previously and the calculations illustrated in Appendix A, the HRV for pre-fire-suppression levels of black-backed woodpecker are illustrated in Table 5, represented as the mean, low range, and high range HRVs.

Table 5.  Estimated acres of black-backed woodpecker habitat that would have been available within a given 6-year period in the pre-fire-suppression era, considering the normal range of probable variability in USFS Region One.

Note that the mean HRV and low range HRV are substantially higher than the acres of existing habitat (Table 4) for all periods except 1988 to 2000.  Note that only in the period of 1988 to 2000 do existing acres of habitat approximate or exceed the mean HRV.  Table 6 compares the existing habitat by time period as a percent of the mean HRV to provide a relative indication of how much black-backed woodpecker habitat was available by time period.

Table 6.  Comparison of existing habitat by 6-year time period to the mean HRV. 

If we take the average acres of existing habitat for all 6-year time periods in the 1940 to 2000 time frame and compare those against the HRVs calculated using the low range, mean, and high range HRVs (Table 5), we get a relative comparison of how much existing habitat we have, compared to historic levels, considering natural variability.  This comparison, calculated as a percentage of what habitat remains compared to the low, mean, and high range HRV, is illustrated in Table 7.

Because the acres of existing habitat are substantially higher in the 1988 to 2000 time periods, the same comparison was made for the time period of 1940 to 1987, and for the time period from 1988 to 2000.  These latter two comparisons allow us to address the questions, respectively, of: 1) how limited was black-backed woodpecker habitat prior to the big fires of 1988 and 2000?, and 2) how normal were the fires of 1988 and 2000 in terms of providing black-backed woodpecker habitat, or to what degree as a result of those fires did we catch up from the previous 47 years of very successful fire exclusion?    These comparisons are illustrated in Table 7.

Table 7.  Comparisons of existing to historic habitat for the periods 1940-2000, 1940-1987, and 1988-2000.

Note that the percentage of existing habitat compared to the mean HRV for the 1940 to 1987 time period was relatively low (18.8%), i.e. little habitat was available for black-backed woodpeckers compared to normal (mean HRV) levels of burned acres.  Even when the habitat available in 1940 to 1987 is compared to the low range HRV that represents the minimum possible departure, only 25.9% of habitat was available in the time period.  Note too, that the percentage of existing habitat compared to the mean HRV for the 1988 to 2000 time period was substantially higher than normal (284.4%).  This explains why the percentage of existing habitat compared to the mean HRV for the entire 1940 to 2000 time period, while lower than normal (75.4%), constitutes no substantial departure in naturally occurring habitat for the total 60-year time period.  These results suggest the fires of 1988 and 2000 were indeed larger than average compared to historic periods, although probably not outside the normal range of acres burned if we had centuries of burn data with which to compare.

HRV validation using 1900-era data.  Based on fire return interval calculations of the approximately 18 million acres of potential black-backed habitat in Region One (Table 2), only a quarter million (Table 5) of those acres would have provided black-backed woodpecker habitat, defined as burns 1 to 6 years old, within stands > 9” dbh.  This suggests that less than 2% of the landscape historically supported black-backed woodpeckers at any point in time.  To test this prediction, this percentage was compared against 1900-era data to ensure that it was reasonably comparable.  Losensky (1995) took 1930s era timber inventory data and backdated it to 1900 to determine the level of stand initiation (post fire) age classes in 1900.  Hessburg and others (1999) took age classes derived from 1930s aerial photography and backdated them to 1900.  Both analyses provided a 1900-era “snapshot” of the relative amount of recently burned forests on the landscape.

Differences between HRV estimates using both “snap-shot” data and calculations based on fire return intervals.  While the results are not directly comparable, both Losensky and Hessburg and others showed substantially greater amounts of recently burned forests in 1900 than the 2% (234,831 acres out of approximately 18 million acres) calculated in this analysis.  Hessburg and others concluded that 32.9% of the landscape in the interior Columbia basin was within a “stand initiation” size class.  The definition of “stand initiation” included the seed-sapling size class and probably represented age classes of at least 0 to 40 years.  Thus, the 1- to 6-year age class predicted in this analysis might only represent 1/6 of that 32.9%, or about 6%.  Hessburg and others’ estimates also do not suggest what percentage of those stand-initiating fires would have occurred within stands < 9” that would not have provided black-backed woodpecker habitat.  Nonetheless, the 2% HRV calculated using fire return intervals appears very conservative compared to Hessburg and others’ findings.

Role of double-burns.  Losensky (2002a) concluded that 29% of the Montana and Idaho landscape was within stands 0 to 15 years of age in 1900.  Even adjusting that percentage by 1/3 to consider only those stands 1 to 6 years old, or about 9%, the 2% HRV calculated by fire return intervals seems low.  Losensky (pers comm.) suggests that a substantial percentage of his 29% estimate is the result of double-burns, where lethal fires returned 10 to 25 years after a stand-replacing event, when large amounts of snags and coarse, woody debris were present to fuel those fires.  Such double-burns provided no habitat for black-backed woodpeckers.  Snags dead for 10-25 years have no value to black-backed woodpeckers, and their consumption during a double-burn would have no impact.  Recognizing these inconsistencies, Losensky (2002a), in an independent review of this HRV calculation method, concluded that “the analysis portrays a reasonable assessment of the condition of black-backed woodpecker habitat.”  Considering, however, the differences between the calculated HRV and the 1900-era data, the calculated HRV should be considered a “very conservative estimate.”  For that reason, the amount of existing habitat in the 1940-2000 time period is very likely a smaller percentage of the HRV than what our calculations show (Table 7).

CONCLUSIONS

Depending on which timeframe we consider, the results are radically different.  For instance, if we had conducted this analysis back in 1987, and only had 1940 to 1987 fire occurrence data, we might logically have assumed that black-backed woodpeckers were at extreme risk, since only 18.8% of the historically available habitat was available during that previous 47-year period, based on the mean HRV (Table 7).  Conversely, had we conducted this analysis using only 1988 to 2000 data, we would have concluded that fire suppression has had little effect on black-backed woodpeckers, since the level of existing habitat available during that period was 284.4% (Table 7), substantially exceeding the mean HRV. 

The relatively minor decline in existing habitat compared to the mean HRV for the entire 1940 to 2000 time period (75.4% of the HRV, Table 7), simplistically interpreted, might suggest that black-backed woodpeckers are at no risk.  That conclusion is likely grossly understated for five reasons:

1) The levels of existing habitat are inflated due to the presence of unburned islands and pockets of low-severity burns within fire perimeter data (see METHODS, page 3).  Limited sampling of the 2000 fires using aerial photos suggested that the acres of black-backed woodpecker habitat derived from fire perimeter maps was roughly twice that identified using aerial photos and on-the-ground surveys.  The compensatory effect of secondary beetle mortality may be limited to extremely large fires, and may not apply to the more common small and mid-sized fires.  

2) The HRV estimates based on fire-return intervals are substantially lower than actual 1900-era conditions based on Losensky (1995) and Hessburg and others (1999).  While double-burns may account for some of this difference (see RESULTS), the HRVs, calculated using fire return intervals, are probably underestimated, resulting in an underestimate of the departure from HRV.

3) Burned habitats lost to timber salvage have not been considered.  While proposed salvage activities represented less than 15% of total acres burned in 2000 and 2001 (USDA in prep a, USDA in prep b, USDA in prep c), the percentages of habitat removed were clearly higher in smaller burns during low-fire years (USDA 1995, USDA 1999).  Again, existing habitat is probably overestimated.  In addition, the acres that were salvaged likely contained the largest (>9” dbh) trees which are the exact ones needed for black-backed woodpeckers. 

4) The 1 to 6-year post-burn period used for modeling existing black-backed woodpecker habitat reflects the total time black-backed woodpeckers are likely to be present.  Caton (1996) and Powell (2000) concluded that the period in which black-backed woodpeckers are abundant is limited to years three and four following the burn.  If annual acres burned since 1940 were grouped in 2-year or 3-year time periods, instead of the 6-year periods used in the analysis, we’d see a greater percentage of “low fire” time periods.  This again suggests the levels of existing habitat are overestimated (or at least in Losensky’s words “very optimistic”), and the departure from HRV is greater than projected.

5) Lastly, and perhaps most significantly, the interval between large fires has increased substantially.  In the interval between 1940 and 1987 (Table 6), the highest relative percentage of existing habitat to the mean HRV for any period was only 35%.  Consequently, 47 years elapsed during which there was substantially less black-backed woodpecker habitat than normal.  Losensky (pers comm.) suggests that there is no evidence that comparably long intervals existed in Region One prior to fire suppression.   For instance, in the period spanning the turn of the last century, extremely large fires occurred in 1889, 1910, 1917, and 1919, suggesting that big fires should be expected somewhere within Region One at the rate of 1 or 2 per decade.  Losensky (2002b) also compared those intervals with fire return intervals in the Trail Creek area that extended back to the 1700s and still found no evidence that fires were absent for anything approaching 47 years.  Recognizing that Hutto (1995) suggested black-backed woodpeckers are in a declining reproductive rated during low fire years, and black-backed woodpeckers are a short-lived species (Dixon and Saab 2000), such extended fire return intervals could clearly place black-backed woodpeckers at risk.

REFERENCES

Applegate, Vick.  Personal Communication.

Bate, Lisa.  Personal Communication.

Bate, Lisa J. 1995.  Monitoring woodpecker abundance and habitat in the central Oregon Cascades, M.S. Thesis, Univ. of Idaho. Moscow, Idaho.

Bock, C.E., and J. H. Bock. 1974. On the geographical ecology and evolution of the three-toed woodpeckers, Picoides tridactylus and P. arcticus. Am. Widl, Nat. 92:397-405.

Bull, E. L, S. R. Peterson, and J. W. Thomas. 1986. Resource partitioning among woodpeckers in Northeastern Oregon. U.S. For. Serv. Res. Note PNW-444, Portland, OR.

Caton, Elaine L. 1996.  Effects of fire and salvage logging on the cavity-nesting bird community in northwestern Montana.  PhD thesis.  University of Montana. Missoula, MT. 115p.  

Denitto, Gregg, Bill Cramer, Ken Gibson, Blakey Lockman, Tim McConnell, Larry Stipe, Nancy Sturdevant, and Jane Taylor.  2000.  Survivability and deterioration of fire-injured trees in the Northern Rocky Mountains.  USDA Forest Service.  Northern Region.  State and Private Forestry.  Report 2000-13.  27pp.

Dixon, Rita D., and Victoria A. Saab. 2000. Black-backed woodpecker: The birds of North America.  A. Poole and F. Gill editors. Cornell Laboratory of Ornithology and the Academy of Natural Sciences. Cornell, New York.  20p.

Goggans, R., R. D. Dixon, and L. D. S. Seiminara.  1987.  Habitat use by three-toed and black-backed woodpeckers.  Oreg. Dep. Nongame Rep. 87-3-02.  Oreg. Dep. Fish and Wildl., Bend.  49 pp.

Hejl, Sallie J.  Personal Communication.

Hejl, Sallie J., and Mary McFadzen. 2000.  Maintaining fire-associated bird species across forest landscapes in the Northern Rockies.  USDA Forest Service.  Rocky Mountain Research Station--Forest Sciences Laboratory.  INT-99543-RJVA.  21pp.

Hessburg, Paul F., Bradley G. Smith, Scott D. Kreiter, Craig A. Miller, R. Brion Salter, Cecilia H. McNicoll, and Wendall J. Hann.  1995.  Historical and current forest and range landscapes in the interior Columbia River basin and portions of the Klamath and Great Basins. Part 1: Linking vegetation patterns and landscape vulnerability to potential insect and pathogen disturbances.  PNW-GTR-458.  467pp.

Hitchcox, Susan M. 1996.  Abundance and nesting success of cavity-nesting birds in unlogged and salvaged-logged burned forests in northwestern Montana.  M.S. thesis. Univ. of Montana.  Missoula, MT.  89 pp.

Hoffman, N. 1997. Distribution of Picoides woodpeckers in relation to habitat disturbance within the Yellowstone area. Master’s thesis. Montana State University. Bozeman.

Hoyt, J. S. 2000. Habitat associations of Black-backed Picoides arcticus and Three-toed P. tridactylus woopeckers in the Northeastern Boreal Forest of Alberta. Edmonton, Alberta: University of Alberta.  96p.  PhD Thesis.

Hutto, Richard L. 1995.  Composition of bird communities following stand-replacement fires in northern Rocky Mountain (U.S.A.) conifer forests.  Conservation Biology 9(5): 1041-1058.

Jensen, Mark E., Norman L. Christensen, Jr., and Patrick S. Bourgeron.  2001.  An overview of ecological assessment principles and applications.  In: A guidebook for integrated ecological assessments.  Springer.  New York.  p. 13-28.

Jones, Jeff, and Steve Barrett.  In preparation.   Historic fire regimes for the Northern Rockies.  From: Historic fire regimes and departures of fire regimes in Northern Idaho and Western Montana: A first approximation. 

Kamps, A. and S. Gunn .  2002. 2001 Douglas-fir bark beetle survey design and monitoring results.  Unpublished document. Available at Bitterroot National Forest. Hamilton, MT. 10p.

Kotliar, Natasha B., Sallie Hejl, Richard L. Hutto, Victoria Saab, Cynthia P. Melcher and Mary E. McFadzen. In press.  Effects of fire and post-fire salvage logging on avian communities in conifer-dominated forests of the western United States.  Submitted to Avian Biology.  69p.

Landres, Peter B., Penelope Morgan, and Frederick J. Swanson.  1999.  Overview of the use of natural variability concepts in managing ecological systems.  In: Ecological Applications.  9(4):1179-1188.

Lockman, David. Personal Communication

Losensky, B. John.  Personal Communication.

Losensky, B. John.  2002a.  An evaluation of methods to determine the historic range of variability for selected species in the northern Rockies.  Unpublished report.  12pp.

Losensky, B. John.  2002b.  An assessment of vegetation and fire history for the Trail Creek Corridor and Lemhi Pass.  Unpublished report.  71p.

Losensky, B. John.  1995. Historical vegetation types of the interior Columbia River basin.  USDA Forest Service Contract INT-94951-RJVA.  90p.

Monson, C. and E. Boniecki. 2002. 2002 Post-Burn Woodpecker Monitoring. Unpublished Report. Available at Lolo National Forest. Missoula, MT. 14p.

Murphy, E. C. and W. A. Lehnhausen. 1998. Density and foraging ecology of woodpeckers following a stand-replacing fire.  Journal of Wildlife Management. 62(4):1359-1372. 

O’Connor, Patricia M., and J. M. Hillis.  2001. Conservation of post-fire habitat, black-backed woodpeckers, and other woodpecker species on the Lolo National Forest.  Unpublished report.  17pp.

Powell, Hugh D. W.  2000.  The influence of prey density on post-fire habitat use of the black-backed woodpecker.  M.S. thesis. Univ. of Montana.  Missoula, MT.  97pp.

Saab, V. A., and J. G. Dudley. 1998. Responses of cavity-nesting birds to stand-replacement fire and salvage logging in ponderosa pine/Douglas-fir forests of southwestern Idaho.  USDA Forest Service Res. Pap. RMRS-11.  USDA Forest Service, Rocky Mountain Research Station.  Ogden, UT. 

Steeger, C. and J. Dulisse.  In press. Characteristics and dynamics of cavity nest trees in southern British Columbia. Proceedings of the Dead Wood in Western Forests Symposium. November 2-4, 1999. Reno, Nevada.

Taylor, Jenny.  Unpublished nest records in northern Idaho.

USDA Forest Service.  1995.  Little Wolf Fire Salvage and Rehabilitation Environmental Assessment.  Tally lake Ranger District.  Flathead National Forest.

USDA Forest Service.  1999.  Gilbert Fire Salvage Environmental Assessment.  On file at Lolo NF, Bldg 24b Ft Missoula, Missoula, MT. 

USDA Forest Service. 2002. 

USDA Forest Service.  In Preparation a.  Lolo Fire Salvage and Restoration EIS. 

USDA Forest Service.  In Preparation b. Moose Post-Fire Project Draft Environmental Impact Statement.  Glacier View Ranger District.  Flathead National Forest. 

USDA Forest Service. In Preparation. c.  Bitterroot Fire Restoration FEIS.  Bitterroot National Forest. 

VanTyne, J. 1926. An unusual flight of Arctic Three-toed Woodpeckers. Auk 43: 469-474.

Villiard, P. 1994. Foraging behavior of Black-backed and Three-toed woodpeckers during spring and summer in a Canadian boreal forest. Can. J. Zool. 72: 1957-1959.

Weinhagen, A. C. 1998. Nest site selection by the Black-backed Woodpecker in Northeastern Vermont. Masters thesis. University of Vermont. Burlington.

West, J. D., and J. M. Speirs. 1958. The 1956-57 invasion of three-toed woodpeckers. Wilson Bull. 71: 348-363.

Wilds, Stephanie P., and Peter S. White.  2001.  Dynamic terrestrial ecosystem patterns and processes.  In: A guidebook for integrated ecological assessments.  Springer.  New York.  Pages 338-351.

Wisdom, Michael J., Richard S. Holthausen, Barbara C. Wales, Christina D. Hargis, Victoria A. Saab, Danny C. Lee, Wendel J. Hann, Terrell D. Rich, Mary M. Rowland, Wally J. Murphy, and Michelle R. Eames.  1999.   Source habitat for terrestrial vertebrates of focus in the interior Columbia basin: broad-scale trends and management implications.  PNW-GTR-485.

Yunick, R. P. 1985. A review of recent irruptions of the Black-backed Woodpecker and Three-toed Woodpecker in eastern North America. J. Field Ornithol. 56:138-152.

Click here for MS Word version of Assessment.