Reprinted from Wildlife Society BulletinVolume 26 Number 1 Spring 1998

Silvicultural systems and red-cockaded

woodpecker management:

another perspective

Larry D. Hedrick, Robert G. Hooper, Dennis L. Krasac,

and Joseph M. Dabney

Rudolph and Conner (1996) maintained that a mod-ified even-aged silvicultural system using irregularshelterwood as the method of regenerating newstands provides greater benefits for red-cockadedwoodpeckers (Picoides borealis) than uneven-agedsystems. Their argument was confined to loblolly (Pi-nus tuedu) and shortleaf (R echinatu) pine foresttypes and emphasized public lands. In a reply toRudolph and Conner (1996) Engstrom et al. (1996)stressed the virtues of uneven-aged management, butrather than adhering to the context established by theformer authors, framed their arguments largelyaround longleaf pine (P. pahstris) on private lands.This resulted in some disparate comparisons whichobscured some issues and overlooked others.

In an attempt to bring sharper focus to the issues,we offer the following thoughts. The context for ourargument is the management of all southern pine for-est types on national forest lands for red-cockadedwoodpecker recovery.

Relative importance of southernpines to the recovery of the red-

cockaded woodpecker

Implicit in the argument of Rudolph and Conner(1996) is the belief that shortleaf and loblolly pine for-est types will be. important in the recovery of the red-cockaded woodpecker. Engstrom et al. (1996) reiter-ated the importance of longleaf pine for red-cockadedwoodpeckers and correctly pointed out that the For-

est Service has given high priority to restoring long-leaf on land previously converted to other foresttypes. However, Engstrom et al. (1996:334) erro-neously stated, “Most National Forests that have largewoodpecker populations are dominated by longleafpine although loblolly pine is dominant in some ar-eas.” The large populations to which Engstrom et al.(1996) referred are those reported by James (1995).Of those populations, only the Vernon and Talladegapopulations occur in habitats dominated by longleafpine (Table 1; U.S. Dep. Agric. For. Serv., unpubl.data). Eighty-three percent of national forest popula-tions reported by James (1995) are dominated by pinespecies other than longleaf.

Recovery in some locales will be dependent onproper management of existing loblolly or shortleafpine types that are not the result of past conversionfrom longleaf, and which will not be converted tolongleaf in the future. These locales include theOconee National Forest in Georgia, Daniel Boone Na-tional Forest in Kentucky, Bat&head National Forestin Alabama, Bienville National Forest in Mississippi,the Davy Crockett and Sam Houston National Forestsin Texas, and the Ouachita National Forest inArkansas (U.S. Dep. Agric. For. Serv., unpubl. data).Collectively, these areasare expected to support ap-proximately 1,900 red-cockaded woodpecker groupsin the future (LJS Dep. Agric. For. Serv. 1995). Onthe Francis Marion National Forest in South Carolina,the Talladega National Forest in Alabama, theKisatchie Ranger District in Louisiana, and the An-gelina and Sabine National Forests in Texas, substan-

Address for Larry D. Hedrick: USDA Forest Service, Ouachita National Forest, P.O. Box 1270, Hot Springs, AR 71902, USA. Addressfor Robert G. Hooper: USDA Forest Service, Southern Research Station, 2730 Savannah Highway, Charleston, SC 29414, USA.Address for Dennis L. Krusac: USDA Forest Service, Southern Region, 1720 Peachtree Road NW, Atlanta, GA 30367, USA. Addressfor Joseph M. Dabney: USDA Forest Service, Ozark-St. Francis National Forests, P.O. Box 1279, Mountain View, AR 72560, USA.

Key words: even-aged management, national forest lands, red-cockaded woodpecker habitat, silviculture, uneven-aged management

Understory conditions in a shortleaf pine stand under even-aged management forred-cockaded woodpecker habitat in the Poteau Ranger District, Ouachita NationalForest. Photo court&y of W. Montague.

tial areas of existing forest types are scheduled foreventual longleaf restoration (U.S. Dep. Agric. For.Serv., unpubl. data). However, restoration will re-quire more than a century to effect. In the meantime,the existing pine types other than longleaf will becritical components of red-cockaded woodpeckerhabitat.

Silvicultural systems and red-cockaded woodpecker habitat

Engstrom et al. (1996:336) posed the critical ques-tion, “How can a forest be structured to allow regen-eration and provide woodpecker cavity trees simulta-neously?” They then argued that uneven-aged man-agement systems in general, and the Stoddard-Neeluneven-aged method as practiced in the RedHills area of Georgia and Florida in particular, aremore suitable than modified even-aged methods formaintaining red-cockaded woodpecker habitat. Weagree with Lennartz (1988), Rudolph and Conner(1996) the Red-cockaded Woodpecker RecoveryPlan (U.S. Fish and Wildl. Serv. 19S5), and the South-ern Region’s Longterm Strategy for Red-cockadedWoodpecker Recovery (U.S. Dep. Agric. For. Serv.1995) that both even-aged and uneven-aged strate-gies can successfully provide suitable habitat. How-ever, we think that even-aged strategiesprovide more potential cavity trees, offer better re-generation potential, result in better stand structure(i.e., fewer midstory problems), and provide bet-

ter opportunities for the use of fre-quent prescribed fire than uneven-aged strategies.

The 2 most critical components ofred-cockaded woodpecker habitatconsist of potential cavity trees andsuitable stand structure. Potential cav-ity trees are living pines with heart-wood of sufficient diameter to containexcavated cavities in conjunctionwith a relatively narrow band of sap-wood (Hooper 1988, Conner et al.1994). In addition, the pool of poten-tial cavity trees must have a high fre-quency of heart rot (Conner andLocke 1982, Hooper 1988, Hooperet al. 1991, Rudolph et al. 1995). Gen-erally, SO- to lOO-year-old loblollyand shortleaf pines, and lOO- to 120-year-old longleaf pines have thesecharacteristics (Clark 1992, 1993).

Stand structure is as important aspotential cavity trees. Especially in the vicinity ofcavity trees, stands need to be park-like and essen-tially devoid of hardwood and pine midstories. En-croaching midstories lead to woodpecker abandon-ment of cavity trees and ultimately to declines in thepopulation of woodpeckers (Locke et al. 1983, Con-ner and Rudolph 1989, Costa and Escano 1989, andLoeb et al. 1992).

Walker (1995) modeled various strategies for pro-viding red-cockaded woodpecker habitat. He foundthat even-aged strategies provided morepotential cavity trees than uneven-aged strategies.Several even-aged strategies (clearcutting, shelter-wood, irregular shelterwood, shelterwood withrelic patches) in loblolly pine provided 50-63% morepotential cavity trees than either of 2 uneven-agedmethods (group selection and single-tree selection;Walker 1995; Table 4, lOO-yr rotation). For longleafpine, the even-aged strategies (clearcutting, shelter-wood, shelterwood with relic patches) produced70-129% more potential cavity trees than the un-even-aged approaches (group and single-tree selec-tion; Walker 1995; Table 6, 120;yr rotation; Fig. 5).

The reason even-aged strategies can provide morepotential cavity trees than uneven-aged approaches isthat the 3 southern pines most important to the red-cockaded woodpecker are intolerant of competitionfor moisture, nutrients, and light. They survive,grow, and develop best in full sunlight and whenthey are relatively free of competition (Baker andBalmer 1983, Burns 1983, Baker 1987, Boyer andWhite 1990).

Table 1. Area of pine forest types on national forests (NF) with major red-cockaded woodpecker populations”

LongleafLongleaf SlashSlash LoblollyLoblolly ShortleafShortleaf Pond Total

Population ha (%) ha (%) ha (%) ha (%) ha (X) ha

Apalachicola NFFrancis Marion NF

Vernon, Kisatchie Ranger District?

Vernon

KisatchieSam Houston NF

Talladega NFCBienville NF

Croatan NF

Evangeline District Kisatchie NF

Osceola NFAngelina, Sabine NFs

Piedmont NWR, Oconee NFd

Total

54,67254,672 (46.2%)(46.2%) 61,094 (51.6%)(51.6%) 2,0532,05312,16612,166 (24.8%)(24.8%) 145 (<l(<l .O%) 34,52134,521

16,89511,68011,680

858523,90823,9081,5341,5345,1285,1287,4117,4117,8507,850

10,31010,31000

151,639

(68.5%)(68.5%)(39.0%)(39.0%)(<l(<l .O%)(59.2%)(59.2%)(2.9%)(2.9%)(9.6%)(9.6%)

(26.8%)(26.8%)(19.6%)(19.6%)(10.5%)(10.5%)

(24.4%)(24.4%)

4,0383,098

960

25

6,72632,046

2,373

9

109,651109,651

(16.4%)(10.3%)(10.3%)(<l.O%)(<l.O%)

(<l .O%)

(<l(<l .O%)(24.3%)(80.2%)

(2.4%)

(<l .O%)(17.6%)

3,7203,720 (15.1%)(15.1%) 0

14,58314,583 (48.7%)(48.7%) 58758749,16549,165 (88.5%)(88.5%) 6,2386,23816,22216,222 (40.2%)(40.2%) 25325348,92148,921 (92.2%)(92.2%) 2,5652,56511,67511,675 (22.0%)(22.0%) 8813,17713,177 (47.7%)(47.7%) 327327

7777 (<l.O%)(<l.O%) 0055,49655,496 (56.5%)(56.5%) 30,09130,09130,60530,605 (97.3%)(97.3%) 854854

280,215280,215 (45.1%)(45.1%) 40,92340,923

(1 (1 .7%).7%)(70.5%)(70.5%)

0

0

(2.0%)

(11.2%)

(<l .O%)(4.8%)(4.8%)

(<l.O%)(<l.O%)(1 (1 .2%).2%)

(30.6%)(30.6%)(2.7%)(2.7%)(6.6%)(6.6%)

539 (< 1 ,000) 118,358118,3582,1632,163 (4.4%)(4.4%) 48,99548,995

0 24,65324,6530 29,94829,9480 55,58455,5840 40,38340,3830 53,04553,045

36,37336,373 (68.4%)(68.4%) 53,18553,18500 27,64127,64166 (<l.O%)(<l.O%) 39,97939,97900 98,27098,27000 31,46831,468

39,08139,081 (6.3%)(6.3%) 621,509621,509

a Population names are from James (1995). Forest type is the existing tree species and is not necessarily the species that historically occurred on a given site.

’ In 1991 it was determined that the Vernon and Kisatchie Ranger Districts were 2 separate populations (U.S. Dep. Agric. For. Serv. 1991).‘The Talladega National Forest consists of 2 red-cockaded woodpecker populations. The population referenced by James (1995) is the Oakmulgee population of the Talladega National

Forest.d Figures are only for Oconee National Forest habitats.

.-. _ _ II

Midstory and understory conditions in a shortleaf pine standunder uneven-aged management in the Womble Ranger District,Ouachita National Forest. Photo courtesy of R. Raines.

Of the 3 species, longleaf is the most intolerant.Boyer (1993) found that as little as 2 m2/ha of overstorybasal area reduced growth of the younger trees by 55%.He also found that >6.2 m2/ha of overstory basal areaprevented retention of adequate regeneration in long-leaf. The effect of competition from mature trees is ap-parent up to 30 m into openings (Farrar 1993, Platt andRathbun 1993). Young longleaf pines in openingsof 0.1 ha are suppressed by adjacent large trees, whilein a 2.0-ha opening they are free from such competitiononly in the center 1.25 ha of the opening (Fat-tar 1993).Platt et al. (1988) reported that openings from thedeath of single trees in the Red Hills colonized slowly.Platt and Rathbun (1993) found the minimum-sizeopening for recruitment in the Red Hills was about 30m in diameter. Except in sparsely stocked stands, re-moval of a single tree would not usually create an open-ing that large. On the other hand, group selection meth-ods, in which large patches are created, can easily meetthe minimum patch-size criterion.

Suppressed, slow-growing longleaf seedlings maybe killed by prescribed fires before reaching a resis-

tant size (Boyer 1974, Boyer and White 1990,Boyer 1993). Seedlings in large openings grow to afire-resistant size faster than those in smaller open-ings and thus have a greater chance of surviving fire.In addition, fuel loads from needle cast are heaviestnear mature trees, which increases the mortality rateof seedlings from fire (Boyer 1974, Platt and Rathbun1993).

Because of its intolerance to competition and theinfluence of fire on seedling survival, longleaf pinestypically grow in even-aged patches. Patch size of acohort in old-growth forests has been found to varyfrom ~0.4 ha to several hundred ha (Schwarz 1907,Chapman 1923, Wahlenberg 1946, Boyer and White1990, Platt and Rathbun 1993). It may be possible toperpetuate longleaf pine using small patch sizes if re-generation is already present prior to cutting, or if ex-treme care is used with fire, and one is willing to waitlong enough for regeneration to become establishedin a given patch. However, the diameter distribu-tions reported by Lindeman (1994) for the Red Hills,where small patch sizes have been used for >50years, suggest that reproduction is marginal at best.This condition may reflect the greater difficulty inburning so as not to destroy suppressed regenerationwhile at the same time attempting to control hard-woods. Also, in some cases reproduction has beenintentionally destroyed in the Red Hills area to main-tain the open stand structure preferred by landown-ers for quail hunting (U.S. Dep. Agric. For. Serv.1995). In any case, we are lacking examples of un-even-aged longleaf forests with adequate numbers ofsmall-diameter trees. Such forests would have muchless of the open ‘structure typically associated withpreferred red-cockaded woodpecker habitat. Towhat extent these denser forests would affect thewoodpecker is unknown.

Loblolly and shortleaf pines are somewhat moretolerant of competition than longleaf and thus aremore adaptable to uneven-aged management (Bakeret al. 1996). However, at least for large-scale applica-tion, we agree with Rudolph and Conner (1996) thatthe problem of midstory control makes uneven-agedmanagement undesirable for providing habitat forred-cockaded woodpeckers in loblolly and shortleafpine forests. Prescribed burning is severely con-strained because fire-susceptible regeneration is pres-ent across the landscape. Baker et al. (1996) statedthat herbicides were necessary every lo-20 years tocontrol hardwoods for pine management. Similarherbicide use would be required for red-cockadedwoodpecker management (U.S. Dep. Agric. For.Serv. 1995). Some publics object to use of herbicidesand, as Rudolph and Conner (1996) pointed out,

burning has ecosystem benefits that herbicide usedoes not.

Ecosystem management andsilvicwltural systems

Engstrom et al. (1996) articulated several key goalsof ecosystem management that include maintainingnative biota, and perpetuating natural forest structureand function when resources are used by humans.We agree with those goals. They then described thekinds of natural disturbances to which upland south-ern pines have been subjected throughout their evo-lutionary history and the natural even-aged and un-even-aged forest structure that resulted. Their de-scription of natural structure in pine forests of theSouth in the pre-exploitation era is consistent withother accounts (Schwarz 1907, Chapman 1909, Mat-toon 1915, Chapman 1923, Turner 1935, Wahlen-berg 1946, Boyer and White 1990, Platt and Rathbun1993).

However, we are left with the impression that Eng-Strom et al. (1996) find no place in ecosystem man-agement for silvicultural systems that mimic less fre-quent, larger-scale disturbance events which result ineven-aged forest structure. They argued that it makesecological sense for management “to mimic the leastsevere form of disturbance because the most severeforms of disturbance will occur regardless of humanactivities” (Engstrom et al. 1996:335), thus implyingthat it makes no ecological sense for management tomimic any larger-scale disturbances. The logic in thisescapes us, for it is equally true that certain small-scaledisturbance events, such as lightning strikes andwindthrow of single or small groups of trees, will oc-cur regardless of the silvicultural system in practice.

It is certainly true that large-scale events like hurri-canes and tornadoes will occur regardless of humanactivities (Hooper and McAdie 1995). However, atleast 2 natural disturbance agents with the potentialto be large in scale (i.e., wildfires and southern pinebeetle [Dendroctonusfrontalis]) outbreaks, are rou-tinely suppressed by human activities while still smallin scale. In no way do these particular agents operateon modern landscapes in the same manner thatthey likely operated over evolutionary time scales.

Silvicultural systems that mimic frequent, small-scale disturbance events and those that mimic lessfrequent, large-scale disturbance events have an eco-logical basis and make ecological sense. In terms ofproducing sufficient numbers of ‘pines of sufficientage and size to meet foraging and nesting needs, eachsystem can be made compatible with the habitat re’

quirements of the red-cockaded woodpecker(Walker 1995).

We suggest that the issue is to decide how best towork within the evolutionary biology of southernpine ecosystems to maintain ecological integrity, in-cluding recovery of the red-cockaded woodpecker,and allow for human resource use. Criteria thatshould discriminate among silvicultural systems forapplication in any specific case include suitability ofthe resultant stand structure for red-cockaded wood-peckers, operational feasibility, the ease or difficultywith which prescribed fire can be integrated, thewellbeing of other biota, and compatibility withother land-management objectives. Several of thesecriteria are likely to vary among ownerships.

Technical considerations versusoperational feasibility

Engstrom et al. (1996) responded to issues aboutthe technical difficulty of selection managementraised by Rudolph and Conner (1996) in 2 ways.First they stated that implementation of selectionmanagement requires a higher level of ecologicalknowledge than that required for even-aged systems,and second, that Forest Service employees often lacksuch knowledge. We disagree on both counts.Any silvicultural system requires application of eco-logical knowledge, but if there is a higher level of un-derstanding required, it is in planning even-aged re-generation treatments, where landscape-level aspectsof juxtaposition, isolation, and connectivity comeinto consideration. These are not normally importantconsiderations in planning uneven-aged treatments.

Engstrom et al. (1996) were particularly disparag-ing in regard to Forest Service timber markers. Thisis unjustified. Forest Service timber markers imple-ment prescriptions prepared by professionalforesters, biologists, and ecologists. They are welltrained and undergo a rigorous certification processbefore being allowed to mark trees. And, they rou-tinely and successfully mark trees under both un-even-aged and even-aged management prescriptions.

Rudolph and Conner (1996) discussed opera-tional differences between even-aged and uneven-aged management. We concur with their assess-ment. However, we would mention 2 other major operational differences not covered in their paper:marking efficiency and inventory procedures. From 3to 5 times more total land area must be traversed andmarked annually under uneven-aged systems in orderto produce a timber volume equal to that under even-aged systems. Also, inventory procedures under un-

even-aged systems are more time consuming to con-duct, and thus, more costly. These additional require-ments mean that more personnel and funding are re-quired to practice uneven-aged management. In-creases in either Forest Service personnel or fundingare unlikely. Implementation problems with selec-tion management are not rooted in technical difficultyor a lack of ecological knowledge among Forest Ser-vice employees as claimed by Engstrom et al. (1996)but rather in operational practicality and efficiency.

Use of prescribed fireRudolph and Conner (1996) correctly pointed out

the difficulty of using prescribed fire with sufficient fre-quency and intensity in uneven-aged management pro-grams because vulnerable reproduction is present at alltimes. Engstrom et al. (1996) replied that if done care-fully, it is possible to burn young loblolly pine standswithout destroying them, as has been demonstrated inresearch plots in the Tallahassee, Florida area.

However, it does not necessarily follow that care-fully applied treatments on research plots are also fea-sible at operational scales. Weather conditions withsuitable combinations of temperature, humidity, andwindspeed may occur frequently enough to allow“careful” burning of research plots, but be altogethertoo rare to allow for operational burning of the hun-dreds of thousands of hectares of red-cockadedwoodpecker habitat that need treatment each year.We agree with both Rudolph and Conner (1996) andEngstrom et al. (1996) that effective prescribed burn-ing on national forest lands for red-cockaded wood-pecker habitat management is essential. However,we doubt that this is logistically compatible withlarge-scale uneven-aged management in pine foresttypes, especially in loblolly, shortleaf, and slash pine.

The Forest Service strategy for redcockaded wood-pecker recovery calls for 365,000 ha of prescribedburning annually in habitat management areas alone(U.S. Dep. Agric. For. Serv. 1995). In recent years theamount of prescribed burning for all purposes in thesouthern region of the Forest Service has averagedabout 250,000 ha and has never exceeded 285,000 ha(U.S. Dep. Agric. For. Serv., unpubl. data). Any addi-tional restrictions on burning imposed by uneven-agedmanagement will widen the gap between what needsto be burned and what can be burned each year.

The Red Hills populationThe Red Hills, an area of approximately 12 1,500 ha

in southwestern Georgia and northwesternFlorida with a red-cockaded woodpecker population

of about 179 groups (Engstrom and Baker 1995) wasoffered by Engstrom et al. (I 996) as a model for man-aging public lands. However, the observed grossdensity of active red-cockaded woodpecker clustersin the Red Hills area is about I/680 ha. This densityis <25% of the overall targeted density for nationalforest lands (U.S. Dep. Agric. For. Serv. 1995).

To put this into perspective, the Forest Service ex-pects a recovered red-cockaded woodpecker popula-tion, approximately 500 groups, of which 250 are ac-tively breeding (U.S. Dep. Agric. For. Serv. 1995) torequire about 80,000 ha. At the observed Red Hillsdensity, a recovered population would require ap-proximately 340,000 ha, an area1 extent that not eventhe largest National Forest in the South can provide.The current red-cockaded woodpecker density inthe Red Hills may not be at the maximum level; onthe other hand, the capability of this area to supporta greater density under existing management cannotbe assumed.

A word about regulationThe goal of nearly all forest management programs

is the sustained yield of products and benefits (Smith1986). The way that sustained yields of products andbenefits, including red-cockaded woodpecker habitator ecosystem conditions in general, is assured isthrough regulation of the forest (Walker 1995). Reg-ulation is achieved when the area occupied by eachage class or size class of trees is represented in equalproportion and consistently grown to provide a con-tinual and approximately equal amount of outputs in-defmitely (Davis 1966). This holds for either even-aged or uneven-aged silvicultural strategies.

Engstrom et al. (1996) briefly discussed regulationfor selection-management systems, but made no men-tion of whether or not regulation is a goal for man-agers and landowners in the Red Hills, or whether ornot the presently favorable habitat situation for red-cockaded woodpeckers there derives from regulatedforest conditions. There are indications (IJS. Dep.Agric. For. Serv. 1995; Lindeman 1994) that few, ifany, of the uneven-aged forests in the Red Hills arenear a regulated condition. Further, a model devel-oped by Lindeman (1994) for the Red Hills area,which simulated trees per unit area over 1,000 years,suggests that wide fluctuations can be expected.These fluctuations would likely result in “boom andbust” flows for both timber yields and red-cockadedwoodpecker habitat over time (Seagle et al. 1987).

For national forest lands, assuring sustained yieldsis a requirement of law (Multiple Use-Sustained YieldAct, 74 Stat. 215; I6 U.S.C. 528-531). Therefore,

managers of national forests must work toward a goalof forest regulation. A regulated uneven-aged struc-ture will be very different from the present prevailingconditions in the Red Hills: there would be manymore smaller and middle-sized trees, and conditionsfor the red-cockaded woodpecker would likely bemuch less favorable than they now are.

Management examples fromnational forests

Engstrom et al. (1996) implied that even-aged man-agement of the National Forests has failed to providerecovery of the red-cockaded woodpecker, and thatfuture recovery using even-aged management is notfeasible. We feel that the following examples are suf-ficient to counter such notions.

The Francis Marion National ForestAfter having its old-growth pine forests

removed between 1885 and 1930, the land for theFrancis Marion National Forest was purchased by thefederal government and designated a National Forestin 1936 (U.S. Dep. Agric. For. Serv. 1977). Based onsurviving timber volumes (A. A. Grumbine, Manage-ment Plan, Francis Marion National Forest, unpubl.rep., U.S. Dep. Agric. For. Serv., 1936), a fairamount of woodpecker habitat was present in 1936(Watson et al. 1995) and could have supported >lOOgroups of woodpeckers (R. G. Hooper, unpubl.data). Prescribed burning, mostly dormant-seasonbums but with some growing-season fires, was be-gun on a regular basis in 1944. Even-aged manage-ment was begun in 1950 (U.S. Dep. Agric. For. Serv.1977). The first documented increase in any popula-tion of red-cockaded woodpeckers occurred in theFrancis Marion National Forest between 1981and 1988 (Hooper et al. 1991). Prior to HurricaneHugo in 1989, the Francis Marion supported 477 red-cockaded woodpecker groups, a population density3 times higher than that ofthe Red Hills (1 group/2 12ha vs. 1 group/680 ha, respectively).

The Vernon DistrictWhen it was established in 1935, the Vernon

Ranger District of the Kisatchie National Forest inLouisiana was a 34,460-ha “wasteland” in the middle ofa 2 million-ha “wasteland” (Cruikshank 1939). E. J.S&latter (Boundary revision of Vernon Unit KisatchieNational Forest, unpubl. rep., U.S. Dep. Agric. For.Serv., 1933) stated that on the area that would becomethe future Vernon Ranger District: (1) all the virginlongleaf pine had been cut, (2) all the merchantable

second growth had been cut, and (3) 90% of the polesand pilings had been cut, with only the smallest sizesremaining. Analysis of aerial photos taken in 1935 re-vealed only 2.7 sawtimber-size pines and 8.4 pulp-wood-size pines per hectare (R. G. Hooper, unpubl.data). In 1936 I1 2 red-cockaded woodpecker groupscould have survived, and it is possible that none did.Under even-aged management and an aggressive pre-scribed burning program, the population expanded to175 groups by 1992. Today, it is the largest populationwest of the Mississippi River (James 1995) and proba-bly has the highest density of any extant large popula-tion. The population density of red-cockaded wood-peckers on the Vernon District is >3 times higher thanthe Red Hills population (1 group/l97 ha vs. 1group/680 ha, respectively).

The Appalucbicola DistrictThe Appalachicola Ranger District in Florida had

an early history similar to the Francis Marion NationalForest (E. R. DeSilvia, Management report for the Ap-palachicola Purchase Unit, unpubl. rep., U.S. Dep.Agric. For. Serv., 1935) but apparently had less sur-viving habitat than the Francis Marion. It hasbeen managed by even-aged strategies and has had avery aggressive prescribed-bum program. Today, theAppalachicola District supports the largest popula-tion of red-cockaded woodpeckers in existence (504groups).

Other national forestsRed-cockaded woodpecker populations on

other national forests have not done as well as thoseon the 3 national forests discussed above. In mostcases the reasons that populations have either de-clined or never increased since the exploitive era oflogging (1880- 1930; Williams 1989) are relatively cer-tain. These include a lack of sufficiently old trees foruseable cavities, a lack of midstory control, and isola-tion of small populations (Hooper 1988, Conner andRudolph 1989, Costa and Escano 1989, Krusac et al.1995). Some of these same problems exist in the RedHills, where 1 subpopulation was extirpated (Baker1983), possibly as a result of isolation.

SummaryThe advantages and disadvantages of even-aged

and uneven-aged systems for red-cockaded wood-pecker management can be summarized as follows.

Even-aged systemsUnder even-aged management, forest stands are

comprised of trees of approximately the same

age. Regulation is achieved among stands at the land-scape level and is accomplished by equalizing thearea occupied by the various age classes of trees. Forexample, at regulation, a landscape being managedunder a 120-year rotation and a lo-year entry cyclewould have about 8.5% of the total area in each of 12IO-year age classes, and regeneration cutting eachyear would be applied in stand-sized units on ap-proximately 0.85% of the total area. In order to per-petuate the system and assure sustainability, repro-duction need be secured only once during the rota-tion for any particular stand.

For red-cockaded woodpecker management, theadvantages of even-aged systems are most apparentin the ease with which prescribed fire can be inte-grated and open park-like conditions can be main-tained over most of the landscape. At any given time,most of the landscape can be burned without regardto its effects on tree reproduction. At any given time,the relatively small portion of the landscape in standscomprised of vulnerable small trees can be easily ex-cluded from burns.

The supposed disadvantages of these systems arisefrom regeneration cutting. The use of clearcutting orstandard seedtree and shelterwood cuts eliminatesthe foraging potential of a stand for as long as 30years and its potential as a cluster site for as long asSO- 100 years. While the use of irregular shelterwoodor seedtree cutting, where seed trees are retained in-definitely, does not immediately eliminate foraging orcluster-site potential; unsuitable conditions do occur15-20 years later when a dense sapling midstory de-velops (Gresham 1996). However, under rotations of120 years for longleaf and shortleaf pines, and80-100 years for other pine species (U.S. Dep. Agric.For. Serv. 1995) an abundance of high-quality habitatwith open stand conditions exists (Walker 1995).Several populations have increased under even-agedmanagement, 2 of these to a recovered level.

Uneven-aged systemsUnder uneven-aged management trees of various

ages and sizes are grown within each stand. Regula-tion is achieved by assuring that the diameter distrib-ution of trees within each stand is structured so thatthere are adequate numbers of trees in alldiameter classes sufficient to sustain yields throughperiodic cutting entries, usually 5-10 years apart. Aregulated uneven-aged stand will have a large num-ber of seedlings, many saplings, fewer medium-sizedtrees, and a small number of large trees per unit area.In order to perpetuate the system and assure sustain-ability, reproduction needs to be secured within eachstand with each cutting cycle.

For red-cockaded woodpeckers, the chief advantageof uneven-aged systems is that foraging and potentialnesting habitat, at least in terms of numbers and sizesof trees, could theoretically be present in all stands allthe time. Therefore, no portion of the landscapewould be lacking basic habitat resources at any time.

A possible disadvantage of uneven-aged systems forred-cockaded woodpecker management is related todifficulty in maintaining open park-like stand condi-tions. One reason for this is that prescribed fire gen-erally cannot be used with the same frequency or in-tensity as in even-aged stands because vulnerable re-generation is present in all stands all the time. But byfar, the more important reason is that the targetedstand structure in regulated uneven-aged stands in-cludes a large number of small- and medium-sizedtrees in subordinate canopy positions, a conditionthat is inherently dense and closed. This condition isantithetical to the open park-like conditions preferredby the red-cockaded woodpecker. The Red Hills areawhich supports a sizeable and stable red-cockadedwoodpecker population has been under uneven-agedmanagement for >50 years. However, regulation hasnot been achieved there. We have no examples of sig-nificant red-cockaded woodpecker populations thriv-ing under regulated uneven-aged conditions.

ConclusionRecovery of the red-cockaded woodpecker is the

mutual goal of all parties to the current debate overwhat silvicultural system should be used on nationalforest lands to effect that end. We intend no criticismof the Stoddard-Neel method as it has been applied inthe Red Hills area. For >50 years it has met the pri-mary objectives of the landowners, while providingsome level of timber products, supporting habitat fora significant population of red-cockaded woodpeck-ers, and providing general protection for native biotain those ecosystems. We agree with Engstrom et al.(1996) that research on its potential applicability andthat of other uneven-aged systems on public lands iswarranted. However, until the results of such re-search are available, we feel that advocating wide-spread use of selection systems for red-cockadedwoodpecker habitat management on national forestlands is premature. In the meantime, we haveenough research information and case history resultsto know that properly designed even-aged systemswill meet all of the habitat requirements of the red-cockaded woodpecker.

Acknowledgments. We are indebted to W. Pell, W.Boyer, C. Rudolph, R.Conner, J. Guldin, and M. Roun-

saville for reviewing and making helpful commentson a previous draft of this manuscript.

Literature citedBaker, J. B. 1987. Production and financial comparisons of un-

even-aged and even-aged management of loblolly pine. Pages267-273 in D. R. Philips, editor. Proceedings of the fourth bi-ennial southern silvicultural conference. U.S. Department ofAgriculture Forest Service, General Technical Report SE-42.

Baker, J. B., and W. E. Balmer. 1983. Loblolly pine. Pages 148-152in Silvicultural systems for major forest types of the UnitedStates. U.S. Department of Agriculture Handbook 445.

Baker, J. B., M. D. Cain, J. M. Guldin, P. A. Murphy, and M. G. Shel-ton. 1996. Uneven-aged silviculture for the loblolly and short-leaf pine forest cover types., U.S. Department of AgricultureForest Service, General Technical Report SO-1 18.

Baker, W. W. 1983. Decline and extirpation of a population ofred-cockaded woodpeckers in northwest Florida. Pages 44-45in D. A. Wood, editor. Proceedings of the Red-cockadedWoodpecker Symposium II. 27-29 January 1983, Panama City,Florida. Florida Game and Fresh Water Fish Commission.

Boyer, W. D. 1974. Impact of prescribed fire on mortality of re-leased and unreleased longleaf pine seedlings. U.S. Depart-ment of Agriculture Forest Service, Research Note SO-182.

Boyer, W. D. 1993. Regenerating longleaf pine with natural seed-ing. Pages 299-309 in S. M. Hermann, editor. The longleaf pineecosystem: ecology, restoration, and management. Proceedingsof the Tall Timbers fire ecology conference, Number 18.

Boyer, W. D., and J. B. White. 1990. Natural regeneration of long-leaf pine. Pages 94- 113 in R. M. Farrar, editor. Proceedings ofthe symposium on management of longleafpine. U.S. Depart-ment of Agriculture Forest Service, General Technical Reportso-75.

Bums, R. M. 1983. Introduction. Pages I-2 in R. M. Bums, editor.Silvicultural systems for major forest types of the United States.U.S. Department of Agriculture Handbook 445.

Chapman, H. H. 1909. A method of studying growth and yield oflongleaf pine applied in Tyler County, Texas. Proceedings ofthe Society of American Foresters 4:207-220.

Chapman, H. H. 1923. The causes and rates of decadence instands of virgin longleaf pine. The Lumber Trade Journal84:(6)11,16,17.

Conner, R. N., and B. A. Locke. 1982. Fungi and red-cockadedwoodpecker cavity trees. Wilson Bulletin 94:64-70.

Conner, R. N., and D. C. Rudolph. 1989. Red-cockaded wood-pecker colony status and trends on the Angeliia, Davy Crock-ett, and Sabine National Forests. U.S. Department of Agricul-ture Forest Service Research.Paper SO-250.

Conner, R. N., D. C. Rudolph, D. Saenz, and R. R. Schaefer. 1994.Heartwood, sapwood, and fungal decay associated with red-cockaded woodpecker cavity trees. Journal of Wildlife Man-agement 58:728-734.

Costa, R., and R. E. F. Escano. 1989. Redcockaded woodpecker:status and management in the southern region in 1986. U.S.Department of Agriculture Forest Service, Technical Publica-tion RS-TP12.

Cruikshank, J. W. 1939. Forest resources of Southwest Louisiana.U.S. Department of Agriculture Forest Service, Southern ForestExperiment Station, Forest Survey Release 43.

Davis, K. P. 1966. Forest management: regulation and valuation.McGraw-Hill Book Co., New York, New York.

Engstrom, R. T., and W. W. Baker. 1995. Redcockaded wood-peckers on Red Hills hunting plantations: inventory, manage-

ment, and conservation. Pages 489-493 in D. L. Kulhavy, R. G.Hooper, and R. Costa, editors. Red-cockaded woodpecker: re-covery, ecology and management. Center for Applied Studies,College of Forestry, Stephen F. Austin State University, Nacog-doches, Texas.

Engstrom, R. T., L. A. Brennan, W. L. Neel, R. M. Farrar, S. T. Lin-deman, W. K. Moser, and S. M. Hermann. 1996. Silviculturalpractices and red-cockaded woodpecker management: a replyto Rudolph and Conner. Wildlife Society Bulletin 24:334-338.

Farrar, R. M. 1993. Growth and yield in naturally regeneratedlongleaf pine stands. Pages 31 l-335 in S. M. Hermann, editor.The longleaf pine ecosystem: ecology, restoration, and man-agement. Proceedings of the Tall Timbers fire ecology confer-ence, Number 18.

Gresham, C. A. 1996. Loblolly pine (Pinus tuedu) irregular shel-terwood development and red-cockaded woodpecker manage-ment-a case study. Southern Journal of Applied Forestry20:90-93.

Hooper, R. G. 1988. Longleaf pines used for cavities by red-cock-aded woodpeckers. Journal of Wildlife Management52:392-398.

Hooper, R. G., D. L. Krusac, and D. L. Carlson. 1991. An increasein a population of red-cockaded woodpeckers. Wildlife SocietyBulletin 19:277-286.

Hooper, R. G., and C. J. McAdie. 1995. Hurricanes and the long-term management of the red-cockaded woodpecker. Pages148-166 in D. L. Kulhavy, R. G. Hooper, and R. Costa, editors.Red-cockaded woodpecker: recovery, ecology, and manage-ment. Center for Applied Studies, Department of Forestry,Stephen F. Austin State University, Nacogdoches, Texas.

James, F. C. 1995. The status of the red-cockaded woodpecker in1990 and the prospect of recovery. Pages 439-451 in D. L.Kulhavy, R. G. Hooper, and R. Costa, editors. Redcockadedwoodpecker: recovery, ecology and management. Center forApplied Studies, College of Forestry, Steven F. Austin State Uni-versity, Nacogdoches, Texas.

Krusac, D. L., J. M. Dabney, and J. J. Petrick. 1995. An ecologicalapproach to recovering the red-cockaded woodpecker onsouthern national forests. Pages 61-66 in D. L. Kulhavy, R. G.Hooper, and R. Costa, editors. Red-cockaded woodpecker: re-covery, ecology and management. Center for Applied Studies,College of Forestry, Stephen F. Austin State University, Nacog-doches, Texas.

Lennartz, R. M. 1988. The red-cockaded woodpecker: old-growthspecies in a second-growth landscape. Natural Areas Journal8:160-165.

Lindeman, S. T. 1994. An evaluation of ecological forest manage-ment for the Red Hills region of Georgia and Florida. Master ofForestry Project, School of the Environment of Duke Univer-sity, Durham, North Carolina.

Locke, B. A., R. N. Conner, and J. C. Kroll. 1983. Factors intluenc-ing colony site selection by redcockaded woodpeckers. Pages46-50 in D. A. Wood, editor. Proceedings of the RedcockadedWoodpecker Symposium II. 27-29 January 1983, Panama City,Florida. Florida Game and Fresh Water Fish Commission.

Loeb, S. C., W. D. Pepper, and A. T. Doyle. 1992. Habitat charac-teristics of active and abandoned red-cockaded woodpeckercolonies. Southern Journal of Applied Forestry I6:120-125.

Mattoon, W. R. 1915. Life history of shortleaf pine. U.S. Depart-ment of Agriculture Forest Service, Bulletin Number 244.

Platt, W. J., G. W. Evans, and S. L. Rathbun. 1988. The populationdynamics of a long-lived conifer (Pinus palustris). AmericanNaturalist 131:491-525.

Platt, W. J., and S. L. Rathbun. 1993. Dynamics of an old-growthlongleaf pine population. Pages 275-297 in S. M. Hermann, ed-

itor. The longleaf pine ecosystem: ecology, restoration andmanagement. Proceedings of the Tall Timbers tire ecologyconference, Number 18.

Rudolph, D. C., and R. N. Conner. 1996. Red-cockaded wood-peckers and silvicultural practices: is uneven-aged silviculturepreferable to even-aged? Wildlife Society Bulletin 24:330-333.

Rudolph, D. C., R. N. Conner, and R. R. Schaefer. 1995. Red-cock-aded woodpecker detection of red heart infection. Pages338-342 in D. L. Kulhavy, R. G. Hooper, and R. Costa, editors.Red-cockaded woodpecker: recovery, ecology and manage-ment. Center for Applied Studies, College of Forestry, StephenF. Austin State University, Nacogdoches, Texas.

Seagle, S. W., R. A. Lancia, D. A. Adams, M. R. Lennartz, and H. A.Devine. 1987. Integrating timber and red-cockaded wood-pecker habitat management. Transactions of the North Ameri-can Wildlife Natural Resource Conference 52:41-52.

Schwarz, G. F. 1907. The longleaf pine in virgin forests: silvicul-tural study. John Wiley and Sons, New York, New York.

Smith, D. M. 1986. The practice of silviculture. John Wiley andSons, New York, New York.

Turner, L. M. 1935. Catastrophes and pure stands of southernshortleafpine. Ecology 2:213-215.

U.S. Department of Agriculture Forest Service. 1977. Francis Mar-ion National Forest Land Management Plan-Final Environ-mental Impact Statement. US. Department of Agriculture For-est Service, Coiumbia, South Carolina.

U.S. Department of Agriculture Forest Service. 1991. Decisionnotice, finding of no significant impact and supplement to theenvironmental assessment interim standards and guidelines forprotection and management of RCW habitat within 3/4 mile ofcolony sites as it pertains to the Apalachicola and Kisatchie Na-tional Forests. U.S. Department of Agriculture Forest Service,Atlanta, Georgia.

U.S. Department of Agriculture Forest Service. 1995. Final envi-ronmental impact statement for the management of the red-cockaded woodpecker and its habitat on national forests in thesouthern region, Management Bulletin R&MB 73. U.S. Depart-ment of Agriculture Forest Service, Atlanta, Georgia.

U.S. Fish and Wildlife Service. 1985. Red-cockaded woodpeckerrecovery plan. U.S. Fish and Wildlife Service, Atlanta, Georgia.

Wahlenberg, W. G. 1946. Longleaf pine: its use, ecology, regen-eration, protection, growth and management. Charles LathropPack Forest Foundation, Washington, D.C.

Walker, J. S. 1995. Potential redcockaded woodpecker habitat pro-duced on a sustained yield basis under different silvicultural sys-tems. Pages 112-130 in D. L. Kulhavy, R. G. Hooper, and R.Costa, editors. Redcockaded woodpecker: recovery, ecologyand management. Center for Applied Studies, College ofForestry, Stephen F. Austin State University, Nacogdoches, Texas.

Williams, M. 1989. Americans and their forests. Cambridge Uni-versity Press, New York, New York.

Larry D. /-/edrick, wildlife biologist and forester, is Team Leaderfor Integrated Resource Management on the Ouachita NationalForest in Arkansas and Oklahoma. His interests include land-scape ecology and ecosystem restoration, in particular, restorationof the shortleaf pine-bluestem ecosystem. Since 1974 he hasmanaged habitat for the red-cockaded woodpecker on nationalforest lands in Texas, Alabama, and Arkansas. Robert G. l-looper,a research wildlife biologist with the Southern Research Station,began research on the red-cockaded woodpecker in 1976. Hehas focused on foraging habitat requirements, prey-habitat rela-tionships, cavity tree selection, cluster initiation, population mon-itoring techniques, hurricane risk assessment, spatial behavior,land-use history, and artificial cavity technology. He is a memberof the Red-cockaded Woodpecker Recovery Team. Dennis L.Krusac, wildlife biologist, is an endangered species specialist withthe U.S. Department of Agriculture Forest Service, Southern re-gion. He began working with red-cockaded woodpeckers in1985. His expertise is in translocation strategies and the role ofsilviculture and prescribed burning as tools to sustain red-cock-aded woodpecker habitat. Dennis was a member of the team thatdeveloped the Forest Service red-cockaded woodpecker recoverystrategy, is the Forest Service red-cockaded woodpecker recoverycoordinator, and a member of the Red-cockaded Woodpecker Re-covery Team. joseph M. Dabney, wildlife biologist, is DistrictRanger of the Sylamore Ranger District of the Ozark-St. FrancisNational Forests in Arkansas. Joe led the interdisciplinary teamthat developed the Forest Service red-cockaded woodpecker re-covery strategy during 1989-l 994.