Wilson Bull., 111(1), 1999, pp. 56-60 RESPONSE OF BROWN-HEADED NUTHATCHES TO THINNING OF PINE PLANTATIONS MICHAEL D. WILSON^ - AND BRYAN D. WATTS' — ABSTRACT. Brown-headed Nuthatches (Sitta pusilla) reached their highest abundance within loblolly pine (Finns taecki) plantations in the hrst year after thinning and declined in subsequent years. Commercial thinning of plantations resulted in a reduction of canopy cover, hardwood basal area, and understory density. Overall, the detection rates of nuthatches were low (19% of points surveyed) and no nuthatches were detected in stands before thinning. Nuthatches were more than three times as likely to be detected within survey points containing snags compared to those that did not. However, snag density did not vary signihcantly between stand ages. These patterns suggest that nuthatch distribution within stands may be influenced by snag distribution but that distribution among stands may be determined by the density and height of understory vegetation. Received 13 March 1998, accepted 15 Sept. 1998. Prior to European settlement of North management practices that produce a habitat America, the Southeastern Coastal Plain was structure similar to the historic southeastern characterize<J by old-growth pine forests that pine ecosystem (Gobris 1992, Plentovich et al. covered more than 24 million ha (Croker 1998). 1979). This ecosystem was maintained by The Brown-headed Nuthatch has also ex- low-intensity ground fires caused by lightning perienced a contraction of its former range strikes (Komarek 1964, 1974) and indigenous (Jackson 1988), and according to data from people (Bartram 1791, Ware et al. 1993). Fires the U.S. Fish and Wildlife Service’s Breeding occurred over vast areas at approximately 3- Bird Survey has been declining at a rate of 5 year intervals (Chapman 1932, Krusac et al. more than 1.5% per year throughout much of 1995) and maintained forests with an open the Southeast (Sauer et al. 1997). Very little midstory and dense ground cover of forbs and is known about the ecology and habitat re- grasses (Platt et al. 1991). quirements of the Brown-headed Nuthatch Land clearing for agriculture, harvesting of and even less is known about how current for- longleaf pine (Pinus palustris) for the naval est management practices may affect its dis- stores industry, and the suppression of wild- tribution. The purpose of this paper is to pre- fires severly reduced the extent of the south- sent some information on the use of pine plan- eastern pine ecosystem by the early 1800s tations by Brown-headed Nuthatches relative (Ashe 1894, 1915; Pinchot and Ashe 1897). to stand age and commercial thinning. Currently, natural stands of longleaf pine are METHODS restricted to only about 1% of their former range (Ware et al. 1993). This study was conducted in managed loblolly pine Brown-headed Nuthatches {Sitta pusilla) (Finns taeda) plantations in eastern North Carolina are among a small group of species including (approximately 35° 50' N, 77° 00' W). These planta- the Red-cockaded Woodpecker (Picoides ho- tions are managed for pulpwood and sawtimber pro- realis) and the Bachman’s Sparrow (Aimophi- duction on a 30—35 year rotation. After canopy clo- sure, the plantations are thinned twice before final har- la aestivalis) that are endemic to the south- vest. Thinnings reduce the number of trees, open the eastern pine ecosystem (Jackson 1988). The forest canopy, and allow for growth of understory veg- Red-cockaded Woodpecker and the Bach- etation. We man’s Sparrow have experienced significant selected stands that repre.sented seven different population declines within the southeast re- ages and relation to thinning: (1)9-11 yearold stands gion (Lennartz and Henry 1985, Dunning with closed canopies, (2) 13—16 year old stands within 1993); however, both have benefited from one year after the first commercial thinning, (3) l6-f8 year old stands that were three years after first thin- ning, (4) 19-21 year old stands that were 5 years after first thinning, (5) 22-26 year old stands that were with- lia'mCeanntderMaFr*oyr.CoWni.lsleiravmastbiuorng,BiVolAog2y,31C8o7l-l8e7g9e5.of Wil- sintanIdsyetahratawfeterre.3seyceoanrds(ahfitnenrisneg,con(6d)t2hi8n-n2i9ng,yeaanrdo(l7d) ^Corresponding author; E-mail: [email protected] 30—35 year old stands that were 5 years after second 56 VV/7.sv;/( and Walls • BROWN-HEADHD NUTHATCH HABn'AT USE 57 thinning. Six replicate stands (each > 24 ha) were se- RESULTS lected for each stand type. Within each stand age, Stand age had a significant influence (Krus- stands were chosen to minimize variation in planted kal-Wallis test: df = 5, P 0.01) on all hab- stocking level and basal area of pine. Stands within each type were separated by at least 500 m. itat variables measured except the density of Seven minute, rtxed-radiiis (50 m) point counts were snags (Table 1). Canopy height and ground used to measure the density and frequency of occur- cover height were positively related to stand rence of Brown-headed Nuthatches within study plan- age whereas pine density was negatively re- tations. Four point counts were established within each lated to stand age. All other significant vari- stand and distributed evenly between edge and interior ables increased with stand age but were also m locations. Edge points were positioned 50 from the influenced by commercial thinning. stand edge such that the plot perimeter was tangential Detection rates for Brown-headed Nut- to the stand edge. For all stands, edge points were hatches within pine plantations were relatively positioned on stand edges that were adjacent to log- ging roads. Interior points were positioned 150 m from low. Nuthatches were detected in 15 of 42 the stand edge. Stands were surveyed three times be- (35.7%) pine stands included in the study and tween 1 June and 4 July 1997. Surveys were initiated 32 of 168 (19%) individual point counts sur- 0.5 hr after sunrise and concluded within four hours. veyed. Stand age had a significant influence The vegetation was sampled within all point count on the detection of Brown-headed Nuthatches pglrootwsintog dpeetrieordm,in(e2)( 1ve)gveetgaettiaotniornescphoansnegsestoactrhoisnsnintgh,e (1X).NatNesoCorrnecutitohna=tch1e2s-3,wedrfe=d5e,tePcte<d 0i.n05;foFriegs.t and (3) relationships between nuthatch distribution and vegetation. Linear transects were used for vegetation patches prior to first thinning. The number of points where nuthatches were detected was sampling parallel to the long, regularly distributed can- opy openings created by row thinning. The length of greatest in the year immediately following vegetation transects was standardized to 25 m and the thinning and declined with time after thinning. width varied between 4 and 7 m to accommodate var- Using survey points as statistical units, nut- iation in thinned and non-thinned longitudinal rows hatches were significantly associated with within stands. Four vegetation transects were estab- habitat variables that were directly influenced lished within each point count and equally distributed by thinning events. For example, nuthatch between thinned and non-thinned rows. abundance was negatively correlated with asencdHtsad.beiCatoadutnsdttaastnadoiwfnegarlelstlceaomrlsglee(cswtneoadogasdt)ytbwpyloatnlytepsveel((s>hwai8rtdhcwimonotddrbahvn)s- Pcan<op0y.0c3)o,vehra(rKdewnodoadlldtens=ity—0.(t12=, 7—70=.141,44n, pine) and stem diameter class (8-23, 24—38, > 38 cm = 144, P < 0.02), and basal area of hard- dbh) were made over the entire 25 m transect. Pine woods (t = —0.13, n = 144, P < 0.02). In and hardwood basal areas were estimated using the addition, nuthatch abundance was positively midpoint dbh for the two smaller diameter classes and correlated with groundcover density (t = 38 cm for the larger class (few trees were larger than 0.19, n = 144, P < 0.001). Nuthatch density 38 cm dbh). Additional information was collected was not significantly correlated with canopy within 2 X 2 m quadrats established at opposite ends height (t = —0.04, n = 144, P > 0.05), pine ocofpoymepacacoshvsetrrbay(nsmececoatns.vuerIxenfdodreimnnastfiioooumnretcceaorrl)dlienacantledddciiarnneccoltpuiydoendshecoiafgnh-ta bdaesnaslitayr(eta =(t0=.04—,0n.02—, 1n44=, P14>4,0P.05>) o0r.0p5)i.ne (measured using a clinometer), groundcover height and Although stand type did not have a signif- counts of all stems, shrubs, and saplings (> 0.5 m in icant influence on the number of survey plots height and < 8 cm dbh). Counts were summed to rep- containing standing snags, and snags did not resent total groundcover density (stem.s/m^). appear to result from thinning, nuthatches A Kruskal-Wallis test was used to test for the influ- were positively correlated with standing snags ceonucnetsofofstsannadgsa.geBeocnausalel ohfabtihteatmvaanryiabzleerso evxacleupets ffoorr (T = 0.15, 77 = 144, P < 0.009). In fact, nut- hatches were over three times more likely to counts of both Brown-headed Nuthatches and snags be detected within survey plots containing (i.e., data were distributed as a negative binomial), fre- quency of occurrence values were used to assess pat- standing snags (12 of 32 plots, 37.5%) com- terns among stand types. The relationship between nut- pared to plots that did not (13 of 122 plots, hatches and habitat variables was assessed at the level 11.6%; = 7.35, df = 1, P < 0.007). of the point count using Kendall’s rank correlation. DISCUSSION Nuthatches were not detected in 9-10 year old stands, so this stand age was eliminated from all analy.ses and It is generally thought that partially rotted used only for descriptive purposes. wood is a prerequisite for cavity excavation ^ 58 THE WILSON BULLETIN • Vol. 111. No. 1. March 1999 where by Brown-headed Nuthatches, and the major- —cq— q— qoooo—oo—o— dcdddddlOdd ity of cavities reported have been located in VVVVVVAVV snags (McNair 1984). In general the popula- snags tion density of cavity-nesting birds is posi- for tively related to snag density (Cunningham et al. 1980, O’Meara 1984, Raphael and White except 1984). In Florida, a large percentage of the — VO r- rj (N variation in the density of cavity-nesting birds (including Brown-headed Nuthatches) was ex- variables plained by snag density and dispersion (Land et al. 1989). Snag density has been shown to ^ O O O O habitat (N On r- (N ID — On (N be lower in pine plantations than in natural stands (McComb et al. 1986), and was low in all the plantations we surveyed. Brown-headed for m O' O On O On O' cD Nuthatches were significantly more likely to (IND ON OPn fON CON »D NO CN be detected within survey points that con- presented tained snags. The possibility that snag density may serve to limit overall nuthatch density are Dj O (N O NO ^ o within loblolly pine plantations requires fur- CD (N O <—N O 00 d (N ther investigation. Because nuthatch density (N 00 so values was influenced by thinning and snag density, snag density alone does not explain nuthatch distribution among pine stands. Median n—ON‘ 0nT0of —dO ——00 OdoN rnOD CD ID Brown-headed Nuthatches exhibited a rapid (N response to thinning. Nuthatches were not de- each). tected within pine plantations prior to the first 6 thinning but reached their highest densities («— O—^n 0O0N CfdoDD p—- O—do o nO — IDi swiptohnisne tshueggfeisrstts ytehaart tahftienrnitnhginancitnigv.itiTehsismarye,- in some way, enhance habitat structure for study nuthatches. Thinning activities were shown to for q ^ od Tf «—D reduce canopy cover, reduce the density and 0—0 00 r- 00 00 basal area of hardwoods, and increase ground- selected cover density. Although the importance of canopy cover types O'I D>Js I—D NOpOn OOd CC^DD dp dO »D 00 O' htoattchheesusheaosfnpoitnebleaenndsexbpyloBrreodw,nE-nhgesatdreodmNauntd- coworkers (1984) reported that nuthatch abun- stand 7 dance declined as the density of midstory of hardwoods increased. This result is consistent •o with our observations that nuthatches were B common oo less in years after thinning, as the presented. characteristics nflj rME density and basal area of hardwoods in- eg eg creased. is eg 'oC-S/) (=Nrc5* u.OoD-S. ^^BcE/) • useThbey eBfrfoecwtns-hoefabduerdniNnugthhaatrcdhweosodisssoimnilsatrantdo Habitat occurrence 6 =tt c .•2? that of thinning. Nuthatches used (45 nut- wVi eg «C rt g' T3 J= hatches/km^) mature longleaf pine stands that T1AB.LE foreqfuency U-‘Joc>aOCSZfN)UCo>C31L T^Sc>3/;> e(u^U aVocOO<aU H•-OOC2o3 ^^c0<e(3cCgUU0r"o0o^OOcOuo>-a03o^oCcuO>JJ NdawuenetvdrheealLtoarcpebhgeiudsldkaeyrunlnsydi1et9rby8s5ut)dro,enrceibrdeuestaisnn(eoHdFitlrwotsrihtitadhnaedtts(imRawlee.iptfeho1nl9nlw9ioe1n)wlg.-l cu ix j: c/5 ing burning (Engstrom et al. 1984, Wilson et Wilson and Watts • BROWN-HEADED NUTHATCH HABITAT USE 59 FIG. 1. Detection frequency (# of point counts) of Brown-headed Nuthatches in commercially thinned pine plantations in eastern North Carolina (n = 24 point counts for each stand type). Frequency distribution between stand ages was significantly different from an even distribution (x^vaiescorrection = 12.3, df = 5, P < 0.05). LITERATURE CITED al. 1995), sirnilar to the decline in nuthatches we observed following thinning. Ashe, W. W. 1894. The forests, forest lands, and forest One possible explanation for the inverse re- products of eastern North Carolina. N.C. Geol. lationship between the density of understory Surv. 5:1-128. vegetation and numbers of Brown-headed Ashe, W. W. 1915. Loblolly or North Carolina pine. Nuthatches is that vegetation may obscure po- N.C. Geol. Surv. and Econ. Surv. 24:1—176. tential cavity locations. Brown-headed Nut- Bartram, W. 1791. Travels through North and South hatch cavities are frequently excavated in rel- Carolina, Georgia, east and west Florida. Dover atively low positions; usually below 3.66 m Publications, New York. (n = 309; McNair 1984). Most (68%) cavities Chapman, H. H. 1932. Is the longleaf type a climax? Ecology 13:328-334. were located in tree stumps (McNair 1984) Conner, R. N. and D. C. Rudolph. 1991. Effects of suggesting that the potential for increasing midstory reduction and thinning in Red-cockaded cavity height may be limited. Regeneration of Woodpecker cavity clusters. Wildl. Soc. Bull. 19: understory and groundcover vegetation was 63-66. rapid in the current study such that low cavity Croker, T. C., Jr. 1979. Longleaf pine. J. For. Hist. 23:32-43. positions could be obscured quickly, but the impact on patch use by Brown-headed Nut- Cunningham, J. B., R. P. Balda, and W. S. Gaud. 1980. Selection and use of snags by secondary hatches is unknown. cavity-nesting birds of the ponderosa pine forest. ACKNOWLEDGMENTS U.S. For. Serv. Res. Pap. RM-222:1-15. Dunning, J. G. Jr. 1993. Bachman's Sparrow in The We thank T. Melchiors for administrative support birds of North America, no. 38 (A. Poole. P. Set- and assistance through all aspects ofthe field work and tenheim, and F. Gill, Eds.). The Academy of Nat- for providing editorial comments on the manuscript. ural Sciences, Philadelphia, Pennsylvania; The We also thank B. Barber and J. Hughes for educating American Ornithologists' Union, Washington, us about forest tracts and silvicultural techniques and D.C. for assistance in selecting study sites. D. Bradshaw, G. Engstrom, R. T, R. L. Crawford, and W. W. Baker. Levandoski, and B. Piccolo provided valuable field as- 1984. Breeding bird populations in relation to sistance, and T. Rafiq helped with data management. changing forest structure following fire exclusion: This study was funded through a cooperative agree- a 15-year study. Wilson Bull. 96:437-450. ment between the Weyerhaeuser Company and the Gobris, N. M. 1992. Habitat occupancy during the College of William and Mary. breeding season by Bachman's Sparrow at Pied- 60 THE WILSON BULLETIN • Vol. Ill, No. 1, March 1999 mont National Wildlife Refuge in central Georgia. O’Meara, T. E. 1984. Habitat-island effects on the avi- M.Sc. Thesis, Univ. of Georgia, Athens. an community in cypress ponds. Proc. Annu. Hirth, D. H., L. D. Harris, and R. E Noss. 1991. Conf. Southeast. Assoc. Fish W'ildl. Agencies 38: Avian community dynamics in a peninsular Flor- 97-110. ida longleaf pine forest. Fla. Field. Nat. 19:33-48. Platt, W. J., J. S. Glitzenstein, and D. R. Streng. Jackson, J. A. 1988. The southeastern pine forest eco- 1991. Evaluating pyrogenicity and its effects on system and its birds: past, present, and future. Pp. vegetation in longleaf pine savannahs. Proc. Tall 119-159 in Bird conservation 3 (J. A. Jackson, Timbers Fire Ecol. Conf. 17:143-161. Ed.). Univ. of Wisconsin Press, Madison. PiNCHOT, G. AND W. W. AsHE. 1897. Timber trees and Komarek, E. V., Sr. 1964. The natural history oflight- forests of North Carolina. N.C. Geol. Surv. 6:1- ning. Proc. Tall Timbers Fire Ecol. Conf. 3:139- 227. 183. Plentovich, S., j. W. Tucker, Jr., N. R. Holler, and Komarek, E. V., Sr. 1974. Effects of fire on temperate G. E. Hill. 1998. Enhancing Bachman’s Sparrow forests and related ecosystems: southeastern Unit- habitat via management of Red-cockaded Wood- ed States. Pp. 251-278 in Fire and ecosystems (T. peckers. J. Wild. Manage. 62:347-354. T. Kowalski and C. E. Ahlgren, Eds.). Academic Raphael, M. G. and M. White. 1984. Use of snags Press, New York. by cavity-nesting birds in the Sierra Nevada. Krusac, D. L., j. M. Dabney, and J. J. Petrick. 1995. Wildl. Monogr. 86:1-66. An ecological approach to recovering the Red- Repenning, R. W. and R. F. Labisky. 1985. Effects of cockaded Woodpecker on southern National For- even-age timber management on bird communi- ests. Pp. 61-66 in Red-cockaded Woodpecker: re- ties of the longleafpine forest in northern Florida. cover, ecology and management (D. L. Kulhavy, J. Wildl. Manage. 49:1088-1098. R. G. Hooper, and R. Costa, Eds). Center for Ap- Sauer, J. R., J. E. Hines, G. Gough, I. Thomas, and plied Studies in Forestry, College ofForestry, Ste- B. G. Peterjohn. 1997. The North American phen E Austin State Univ., Nacogdoches, Texas. breeding bird survey results and analysis, version Land, D., W. R. Marion, and T. E. O’Meara. 1989. 96.4. Patuxent Wildlife Research Center, Laurel, Snag availability and cavity nesting birds in slash Maryland. URL = http://www.mbr.nbs.gov/bbs/ pine plantations. J. Wildl. Manage. 53: 1 165-1171. bbs.html Lennartz, M. R. and V. G. Henry. 1985. Recovery Ware, S., C. Frost, and P D. Doerr. 1993. Southern plan for the Red-cockaded Woodpecker. U.S. Fish mixed hardwood forest: the former longleaf pine and Wildlife Service, Atlanta, Georgia. forest. Pp. 447-494 in Biodiversity of the south- McComb. W. C., S. a. Bonney, R. M. Sheffield,—and eastern United States (W. H. Martin, S. G. Boyce, N. D. Cost. 1986. Snag re.sources in Florida are and A. C. Echternacht, Eds.). John Wiley and they sufficient for average populations of primary Sons, Inc., New York. cavity nesters? Wildl. Soc. Bull. 14:40-48. Wilson, C. W., R. E. Masters, and G. A. Bukenho- McNair, D. B. 1984. Clutch-size and nest placement FER. 1995. Breeding bird response to pine-grass- in the Brown-headed Nuthatch. Wilson Bull. 96: land community restoration for Red-cockaded 296-301. Woodpeckers. J. Wildl. Manage. 59:56-67.