Journal of Fish and Wildife Management Vol 1 #2

Volume 1, Number 2, November 2010

ISSN 1944-687X

JOURNAL OF FISH AND WILDLIFE MANAGEMENT

A Biannual Public Domain Publication Developed by the U.S. Fish and Wildlife Service

EditorialIn This IssueWelcome to the second issue of the Journal of Fish and Wildlife Management! In this issue we have more than doubled the number of papersonce again they are impressive in their scope and breadth. Study locations for Articles, Notes, and Surveys range from the Sandhills of Nebraska to the Prairie Potholes of Minnesota, from the Great Lakes of the Northeast to the Big Woods of Arkansas, and from the coasts of Texas to the alkaline waters of Walker Lake, Nevada. The research topics and focal species are similarly diverse, including assessments of wetland changes, habitat usage, and recreational carrying capacity, as well as investigations of predator prey relationships, population dynamics, and density estimates, to name a few; species include fish, mammals, birds, invertebrates, and reptiles, such as bluegill Lepomis macrochirus, yellow perch Perca flavescens, brook trout Salvelinus fontinalis, Lahontan cutthroat trout Oncorhynchus clarki henshawi, elk Cervus elaphus, bobcats Lynx rufus, mottled ducks Anas fulvigula, Lake Erie watersnakes Nerodia sipedon insularum, western pearlshell mussels Margaritifera falcata, and several species of bats, woodpeckers, and turtles. This issue also contains two interesting Issues and Perspectives essays. One makes a compelling argument for changes to the implementation of endangered species laws and the other comprehensively details the history of avian captive breeding and provides insights for current and future conservation efforts. release once the services are active. Second, we will soon be checking each accepted manuscript against the CrossCheck antiplagiarism database provided by CrossRef (for more information see http://www.crossref.org/ crosscheck.html) in an effort to actively engage in the prevention of plagiarism. Finally, we are developing a comprehensive data archiving policy for all papers published in the Journal of Fish and Wildlife Management and North American Fauna in order to promote the preservation and maximum use of data, similar to the efforts of many other top journals and societies (e.g., The American Naturalist, Evolution, Molecular Ecology, Ecological Society of America). In fact, for most of the studies in this issue, the data required to replicate the results are included directly in the paper or in the accompanying Supplemental Materials. The new policy will be detailed in the next Journal of Fish and Wildlife Management editorial (i.e., June 2011; see the editorials by Pullin and Salafsky [2010] and Whitlock et al. [2010] for compelling arguments on the critical need to develop ubiquitous data archiving policies).

Reviewers and Editors WantedWe will continue to need more experts in a variety of conservation subjects to augment our current group of talented editors and reviewers. For those of you who are interested and have not signed up to be a reviewer, I encourage you to do so by registering at our manuscript submission website (https://jfwm.allentrack.net). Contact me directly if you are interested in serving as a Subject Editor. Please also explore our home page (http:// fwspubs.org) to view all Journal of Fish and Wildlife Management and North American Fauna papers (all freely available in the public domain), sign up for table of content alerts, track and export citations, share articles, and more. Enjoy! Dr. John K. Wenburg Editor-in-Chief Journal of Fish and Wildlife Management [email protected]

Journal UpdatesThere are several journal updates of note. First, we are currently working with leading abstracting and indexing services to register for uploading of all content published in the Journal of Fish and Wildlife Management (and our sister monograph journal, North American Fauna), including Thomson Reuters Web of Knowledge (Web of Science, BIOSIS, Zoological record); Elseviers Scopus; ProQuests Environment Complete; PubMed; EBSCOs Fish, Fisheries & Aquatic Biodiversity Worldwide; and Wildlife and Ecology Studies Worldwide. This will greatly increase the profile and citation of scientific works published here, and more importantly, it will expand their utility for the conservation community at large. Some of these services should be active before this editorial is published, but the registration process is extremely time consuming and rigorous, requiring up to three published issues and many months to complete for some service providers. Rest assured that all of the past and future content from both journals will be incorporated into these services as soon as possible and new content will be uploaded almost immediately upon

ReferencesPullin AS, Salafsky N. 2010. Save the whales? Save the rainforest? Save the data! Conservation Biology 24: 915917. Whitlock MC, McPeek MA, Rausher MD, Rieseberg L, Moore AJ. 2010. Data archiving. The American Naturalist 175:145146.

Journal of Fish and Wildlife Management | www.fwspubs.org

November 2010 | Volume 1 | Issue 2 | 62

J.K. Wenburg

Special Thanks to Reviewers and EditorsStarting an applied conservation journal from scratch has been a rewarding, exciting, and challenging process, one that would not have succeeded without the substantial efforts of our volunteer editors and reviewers. Rigorous peer review for the wide variety of papers published in the first two issues required a large and diverse cadre of experts from many fields across many agencies, organizations, and academic institutions. The following individuals gave of their valuable time and expertise to serve as reviewers, editors, or both in order to help make our first year a success. Thank you! Layne Adams Roger Applegate Meredith Bartron Rick Bennett Mark Bertram John Bigelow Sean Blomquist Wes Bouska Pat Brandes Mike Brasher Jeffrey Bromaghin Tom Cade Don Campton Steven Castleberry Steven Chadwick Steven Chambers McCrea Cobb John Coll Michael Colvin Warren Conway R. Corace, III Scott Costa Richard Crossett Paul Cryan Thomas Dahl John Daigle Robert Daniels Brian Davis Phil Delphey Jeff Denman Duane Diefenbach Andrew Dolan Chris Dwyer Cheryl Dykstra Diane Elam Janet Ertel Joy Evered Tim Fannin David Flaspohler John Fletcher Scott Foott Mark Ford Robert Ford Steve Fritts Katie Frohardt William Gale John Galvez Ken Gates Orin Gelderloos Mark Gernes Erin Gillam Charles Goebel Michael Gregg James Haas Andy Hafs Suzanne Hagell Eric Hallerman David Haukos Mike Hayes Patricia Heglund Rob Holbrook Marty Holtgren Brian Huberty Joshua Hull Lawrence Igl Eric Johnson Philip Johnson Rex Johnson Stephanie Jones Matina Kalcounis-Ruppell Mike Kaller Daniel Kashian Melinda Knutson John Kocik James Layzer Kent Livezey Jim Lyons Michael Mace Bruce Manny Bruce Marcot Doug Markle Steven Martin Alec Maule Jason May John McDonald Lyman McDonald Steven Morey Mike Millard Brian Millsap Stephen Miller Leopoldo Miranda Steven Morey Robert Murphy Chris Nicolai Clay Nielson Doug Norris Dana Ohman Jeffrey Olsen Bradley Onken John Organ Frank Panek Mary Parkin Abigail Pattishall David Payer Grey Pendleton David Perkins Roger Perry Ray Petering Kristine Pilgrim Carol Pollio John Posey Madison Powell Charles Rabeni Barnett Rattner Steve Reagan Stewart Reid Sarah Rinkevich Nathan Roberts Karyn Rode Becky Rosamond Joe Rydell Mike Shingleton Jeffrey Simmons Rod Simmons Thomas Simon David Smith Bill Starkel Mark Steingraeber Todd Sutherland John Sweka Kirsty Swinnerton Jennifer Szymanski Bill Thompson Bruce Thompson Jonathan Thompson Stan Trauth Nick Trippel Tyler Wagner Brian Wakeling Patrick Walsh William Walter Jeffrey Warren Keith Weaver Lisa Williams John Willson Teresa Woods Michelle Workman Christian Zimmerman



Articles

Initial Effects of Prescribed Burning and Understory Fertilization on Browse Production in Closed-Canopy Hardwood StandsChristopher E. Shaw,* Craig A. Harper, Michael W. Black, Allan E. Houston C.E. Shaw, C.A. Harper Department of Forestry, Wildlife and Fisheries, University of Tennessee, Knoxville, Tennessee 37996 M.W. Black Aerospace Testing Alliance Conservation, Arnold Air Force Base, Tennessee 37389 Present address: Northern Bobwhite Conservation Initiative, University of Tennessee Institute of Agriculture, Knoxville, Tennessee 37996 A.E. Houston Ames Plantation, Grand Junction, Tennessee 38039

AbstractForage production for white-tailed deer Odocoileus virginianus is often limited in closed-canopy forests. We measured browse production and nutritional carrying capacity after prescribed burning and understory fertilization in closedcanopy hardwood stands one growing season after treatment in two physiographic regions of Tennessee. Nutritional carrying capacity estimates for prescribed burning, fertilization, and prescribed burning with fertilization were greater than in controls on the Cumberland Plateau. However, the cost per pound of forage produced after fertilization exceeded US$26. In the Coastal Plain, fertilization did not affect nutritional carrying capacity, and prescribed burning and prescribed burning with fertilization lowered nutritional carrying capacity from controls. At both sites, prescribed fire had minimal effect on soil pH or soil phosphate and potash levels. Our results suggest prescribed fire and fertilization are of limited utility for increasing browse production in closed-canopy hardwood forests. Keywords: fertilizer; fire; nutrient; Odocoileus virginianus; white-tailed deer Received: October 18, 2009; Accepted: July 12, 2010; Published Online Early: August 2010; Published: November 2010 Citation: Shaw CE, Harper CA, Black MW, Houston AE. 2010. Initial effects of prescribed burning and understory fertilization on browse production in closed-canopy hardwood stands. Journal of Fish and Wildlife Management 1(2): 6472; e1944-687X. doi: 10.3996/102009-JFWM-016 Copyright: All material appearing in the Journal of Fish and Wildlife Management is in the public domain and may be reproduced or copied without permission. Citation of the source, as given above, is requested. The findings and conclusions in this article are those of the author(s) and do not necessarily represent the views of the U.S. Fish and Wildlife Service. * Corresponding author: [email protected]

IntroductionAn increasing number of nonindustrial private landowners in the eastern United States actively manage their property for wildlife (Measells et al. 2005, 2006). The majority of these landowners manage for white-tailed deer Odocoileus virginianus (hereafter deer), and the most popular land management practice is planting food plots (Schweiss and Dwyer 2008). Acreage dedicated to food plots, however, is a small fraction of the property, and practices to improve forested areas could increase nutritional carrying capacity (NCC).Journal of Fish and Wildlife Management | www.fwspubs.org

Regeneration methods, such as clearcutting, and timber stand improvement practices can improve forage availability for deer (Blymyer and Mosby 1977; Miller et al. 2009). However, many landowners are not interested in harvesting their timber or removing any trees. Prescribed fire also has been used to enhance forage availability for deer in forested areas (Dills 1970); however, most work concerning use of prescribed fire for increased deer browse has followed some level of canopy removal to increase available sunlight (Masters et al. 1993; Jackson et al. 2007). Fertilization has been shown to affect production (Segelquist and Rogers 1975; Dyess et al.November 2010 | Volume 1 | Issue 2 | 64

Burning and Fertilization in Closed-Canopy Hardwood Stands

C.E. Shaw et al.

Figure 1. Location of two study areas in Tennessee: Rocky River Hunting Club is located within the Cumberland Plateau physiographic province and Ames Plantation is in the Coastal Plain physiographic province.

1994) and nutritional quality (Wood 1986; Harlow et al. 1993) of deer forages, but data evaluating effects of fertilization in closed-canopy hardwood stands are lacking. We are not aware of any published evaluations of the efficacy of prescribed fire and fertilization in closedcanopy hardwood forests. Evaluation of practices that do not alter the forest overstory is warranted because many landowners are interested in improving forage availability for deer without altering the forest canopy. We conducted this field study to evaluate the effects of prescribed fire, understory fertilization, and prescribed fire with understory fertilization on nutrient availability and browse production in closed-canopy hardwood stands in two distinct physiographic regions of Tennessee. Our objectives were to determine deer use of browse species and production and nutritional quality of browse after treatments.

Ames Plantation We selected an oakhickory stand (12.80 acre) within the Coastal Plain physiographic province at Ames Plantation in Fayette County (Figure 1). White oak, yellow-poplar Liriodendron tulipifera, southern red oak Quercus falcata, blackgum, and sweetgum Liquidambar styraciflua were common in the overstory. Midstory species included winged elm Ulmus alata, black cherry Prunus serotina, and flowering dogwood. Poison ivy Toxicodendron radicans, Japanese honeysuckle Lonicera japonica, Virginia creeper Parthenocissus quinquefolia, and supplejack Berchemia scandens were common in the understory. Soils were primarily Ruston sandy loam that are well drained and acidic, with 1230% slopes (Flowers 1964). Site index for shortleaf pine was 5070 (Flowers 1964). Deer density estimates obtained with infraredtriggered cameras (Jacobson et al. 1997) indicated a minimum density of 21 deer/mi2.

Study AreasWe selected two closed-canopy hardwood stands with no recent fire histories. Stands were chosen to ensure uniformity (e.g., similar species composition, soils) within a site. Rocky River Within the Cumberland Plateau physiographic province, we selected a shortleaf pineoak Pinus echinata Quercus spp. stand (12.80 acre [5.18 ha]) known as Rocky River, in Sequatchie County (Figure 1). Overstory species included scarlet oak Quercus coccinea, white oak Quercus alba, shortleaf pine, black oak Quercus velutina, and mockernut hickory Carya tomentosa. Midstory species included mockernut hickory, sassafras Sassafras albidum, sourwood Oxydendrum arboreum, blackgum Nyssa sylvatica, red maple Acer rubrum, pignut hickory Carya glabra, and flowering dogwood Cornus florida. Soils were primarily Lonewood silt loam and Lily loam that are well drained and acidic, with 212% slopes (Prater 2003). Site index for shortleaf pine was 70 (Prater 2003). Deer density estimates obtained with infrared-triggered cameras (Jacobson et al. 1997) indicated a minimum of 28 deer/mi2 (deer/2.59 km2).Journal of Fish and Wildlife Management | www.fwspubs.org

MethodsSampling methodology and treatment application We systematically located sixteen 100-yd (91.44-m) transects 100 ft (30.48 m) apart within each 12.80-acre (5.18-ha) stand during summer 2004. We measured woody leaf biomass (pounds per acre) and herbaceous forage within sixty-four 60-ft2 (5.57-m2) sampling plots systematically placed every 25 yd along each transect (Figure 2). We tallied woody browse plants within sample plots to species (stem count tally), and stems were noted as browsed or unbrowsed. We also noted browsing on herbaceous plants along the line transect. For woody vines, we used a measure of inches covered along the line transect in a regression equation to estimate total stem counts of these species from their coverage. We collected and sorted leaves of woody vegetation and all above-ground growth of herbaceous plants #4 ft. We placed samples in a forced-air oven at 50uC until cessation of weight loss and then weighed samples to determine dry-matter weights (grams). After pretreatment data collection, we divided stands into four 3.2-acre (1.3-ha) sections, each containing four of the established transects (Figure 2). We collected soil samples along the four transects within each section;November 2010 | Volume 1 | Issue 2 | 65


C.E. Shaw et al.

phate/acre and 131 lb potash/acre. We collected soil samples in June and August 2005 to track responses in pH, phosphate, and potash levels posttreatment. During July and August 2005, we located plots between areas sampled in 2004 to avoid previously sampled areas. Sample plots in summer 2005 were 4 ft in width 6 10 ft in length. We recorded evidence of browsing on woody plants in sample plots by using a stem tally. We collected all woody leaves and herbaceous plants #4 ft and sorted them by species or species groups (i.e., hickory, red oak, or white oak group). We placed samples in a forced-air oven at 50uC until cessation of weight loss and weighed to determine dry-matter weights. We combined samples of species or species groups within the same treatment into a composite sample and ground with a Wiley mill until particles passed through a 1-mm screen. We analyzed composite samples for nitrogen (N) with a LECO FP-2000 nitrogen analyzer (LECO Corp., St. Joseph, MI) by using the Dumas combustion method (method 990.03; AOAC 1998) to obtain estimates of crude protein (CP) for species or species groups. We conducted fiber analyses (neutral and acid detergent; Jung 1997) with an ANKOM 200 fiber analyzer (ANKOM Technology, Macedon, NY).Figure 2. Schematic illustrating sampling and treatment application (understory fertilization, prescribed burning, and prescribed burning with understory fertilization) in each of two closed-canopy hardwood stands (Rocky River Hunting Club and Ames Plantation) in Tennessee.

combined them to form a composite sample; and submitted them to the University of Tennessee Soil, Plant and Pest Laboratory for analysis of pH, phosphorus (P), and potassium (K) levels. We burned two sections in each stand during the dormant season (Rocky River, March 30, 2005; Ames Plantation, April 5, 2005) by using low-intensity fire under the following conditions: 620uC, 2040% relative humidity, wind speed of 36 mi/h (4.839.66 km/h), and a mixing height of .1,640 ft. For all controlled burns, backing fires were set initially and the remainder of the units were burned using relatively low-intensity strip-heading fires generating 618-in. (15.2445.72 cm) flame heights. We applied fertilizer in late spring 2005 (Rocky River, May 16, 2005; Ames Plantation, May 12, 2005). To avoid issues with pseudoreplication, fertilizer was applied to each individual transect (replicate) instead of across the entire burned section. Before application, we calibrated a hand spreader to ensure proper distribution for each nutrient according to pretreatment soil test results. We fertilized four transects within one burned and one unburned section with ammonium nitrate (3400 [NP K]) at 45 lb N/acre. Triple superphosphate (0460) and muriate of potash (0060) were applied to raise phosphate (P2O5) and potash (K2O) to levels where a plant response would be expected based upon soil test results. At Rocky River, we applied 72 lb (32.66 kg) phosphate/acre and 205 lb potash/acre. In the burned section at Ames Plantation, we applied 52 lb phosphate/ acre and 101 lb potash/acre. For the fertilized-only transects at Ames Plantation, we applied 72 lb phosJournal of Fish and Wildlife Management | www.fwspubs.org

Statistical analyses We collected browse and herbaceous forage in both years to compare production in control and treatment areas within the closed-canopy hardwood stands. Therefore, we used a completely randomized split-plot design for a mixed model analysis of variance. Fixed effects were treatment, year, and the treatment 6 year interaction. Random effects were transect (treatment) and sample plot (transect 6 treatment). Log or log + 0.5 transformations were used when necessary to address normality and homogeneity of variance. When the interaction term was significant (P , 0.05), we used the least significant difference method for mean separation. We chose 10 browse species or species groups for biomass comparisons after treatments based on deer selectivity (see description below) and contribution of each species or species group to total biomass at each site. We compared individual browse species or species group biomass among treatments by using a completely randomized design for the mixed model analysis of variance. Burn and fertilizer treatment were the treatment factors. Before using the log transformation for the 10 individual species or species groups, 0.5 was added to all biomass values to retain observations with 0 values. For testing treatment effects, we used a Bonferroni-corrected a level of 0.01 (0.10/10 species tested). When significant (P , 0.10) differences were found, we used the least significant difference method to detect differences among means. Using pretreatment (2004) data, we calculated a selection index (Chesson index; Chesson 1978, 1983) by dividing the ratio of use and availability for a given species or species group by the sum of ratios for all species or species groups for woody plants and browse species having stem counts $25 (Supplemental Material, Table S1, http://dx.doi.org/10.3996/102009-JFWM-016.S1). We combined species or species groups with ,25 stemsNovember 2010 | Volume 1 | Issue 2 | 66


C.E. Shaw et al.

Table 1. Soil test results (2-in. [5.1 cm] depth) before and after implementation of understory fertilization (fert), prescribed burning (burn), and prescribed burning with understory fertilization (burn/fert) in two closed-canopy hardwood stands (Rocky River Hunting Club [RR] and Ames Plantation [AP]) in Tennessee, September 2004August 2005.Treatment Site RR RR RR AP AP AP RR RR RR AP AP AP RR RR RR AP AP AP Sample date September 23, 2004 June 23, 2005 July 28, 2005 February 12, 2005 June 16, 2005 August 17, 2005 September 23, 2004 June 23, 2005 July 28, 2005 February 12, 2005 June 16, 2005 August 17, 2005 September 23, 2004 June 23, 2005 July 28, 2005 February 12, 2005 June 16, 2005 August 17, 2005 Comparison pH pH pH pH pH pH Phosphate Phosphate Phosphate Phosphate Phosphate Phosphate Potash Potash Potash Potash Potash Potash Control 4.4 4.3 4.4 6.3 6.1 6.2 4 4 4 4 4 8 48 98 51 180 120 200 Fert 4.6 4.4 4.3 6.4 6.1 6.0 4 24 20 4 16 28 71 220 160 140 200 300 Burn 4.3 4.4 4.3 6.1 6.0 6.1 8 4 12 4 4 8 49 78 53 140 100 170 Burn/fert 4.5 4.6 4.4 6.2 6.6 6.1 4 24 24 12 24 28 46 280 170 160 240 300

into an other category. We could not calculate a selection index for herbaceous forage species. Cutoff values indicating no selection depended on the number of species or species group compared at each site (Ames Plantation, 1/25 = 0.04; Rocky River, 1/11 = 0.09). Values above and below these values indicate greater and lesser use, respectively, than expected at a given site. We calculated estimates of NCC in 2005 with the explicit nutritional constraints model (Hobbs and Swift 1985). Following criteria used by Edwards et al. (2004), we estimated NCC for deer by using constraints of 12% CP and a dry matter intake of 3 lb/d. We determined nutritional values for individual browse species for each species collected within each treatment or control area. Because samples of each species within each treatment or control area were combined for nutritional analyses, we report absolute values for CP, neutral detergent fiber, and acid detergent fiber. We based browse species included in the NCC estimate upon selection indices calculated at each site. We used a completely randomized design for a mixed model analysis of variance to compare NCC estimates among treatments, with an a level of 0.05. We log-transformed data when necessary to address normality and variance problems.

Effects on forage production Effects of treatments on forage production varied among study sites. Herbaceous forage increased in all treatments as well as controls at Rocky River from 2004 to 2005 (Table 2). At Ames Plantation, herbaceous forage was increased after prescribed fire and prescribed fire with fertilization (Table 2). Browse production at Rocky River did not increase after fertilization but did increase after prescribed fire and prescribed fire with fertilization (Table 2). No treatment increased browse production at Ames Plantation (Table 2), and there was no meaningful effect on biomass of individual browse species or species groups after treatments at either site (Table 3). Crude protein and fiber content were variable among species or species groups and treatments (Table 4). Deer selection Greenbrier Smilax spp., blackgum, and blackberry Rubus spp. were used more than expected based on availability at Rocky River. Hickory, blueberry Vaccinium spp., red maple, sourwood, sassafras, white oak group, and red oak group were used less than expected. No use was recorded for species in the red oak group. At Ames Plantation, greenbrier, supplejack, blackgum, rose Rosa spp., and winged elm were browsed more than expected based on availability. Species used less than expected based on availability included slippery elm Ulmus rubra, sugar maple Acer saccharum, blackberry, red oak group, black cherry, white oak group, eastern redbud Cercis canadensis, red maple, hickory, grape Vitis spp., Japanese honeysuckle, ash Fraxinus spp., Virginia creeper, and poison ivy. No browsing was recorded for yellow-poplar,November 2010 | Volume 1 | Issue 2 | 67

ResultsEffects on soil Soil pH remained similar across all treatments and sampling periods at both study sites (Table 1). As expected, soil phosphate and potash levels increased after fertilization treatments, but they were not influenced by prescribed fire.Journal of Fish and Wildlife Management | www.fwspubs.org


C.E. Shaw et al.

Table 2. Woody leaf biomass and herbaceous forage production (dry matter pounds/acre) before and after implementation of understory fertilization (fert), prescribed burning (burn), and prescribed burning with understory fertilization (burn/fert) in two closed-canopy hardwood stands (Rocky River Hunting Club [RR] and Ames Plantation [AP]) in Tennessee, JulyAugust 2004 and 2005. Within each comparison at a site, means with the same letter are not different (P . 0.05).Treatment Control Site RR RR RR RR AP AP AP APa c b

Fert

Burn

Burn/fert

Year 2004 (pre-) 2005 (post-) 2004 (pre-) 2005 (post-) 2004 (pre-) 2005 (post-) 2004 (pre-) 2005 (post-)

Comparison Woody leafa

x (SE) 59.4 (7.1) E 71.4 (11.7) E 9.5 (2.0) 13.2 (3.0) 162.8 (18.8) 188.3 (20.5) 9.6 (4.3) BC 13.4 (7.8) B

x (SE) 135.8 (18.0) BC 106.4 (11.1) CD 13.0 (4.6) 23.4 (7.6) 139.4 (12.9) 163.6 (22.1) 28.8 (8.7) A 21.3 (4.5) A

x (SE) 108.8 (11.6) CD 195.7 (23.7) AB 8.2 (3.3) 18.0 (6.8) 140.3 (26.4) 104.3 (16.1) 7.3 (3.0) BC 15.6 (3.0) A

x (SE) 90.7 (16.9) D 232.4 (21.8) A 10.4 (2.6) 51.0 (13.3) 169.1 (19.5) 204.4 (30.3) 1.3 (0.7) C 49.1 (11.6) A

Woody leafa Herbaceousb Herbaceousb Woody leafc Woody leafc Herbaceousd Herbaceousd

Treatment effect significant (P = 0.02) for woody leaf production at RR. Treatment effect not significant (P = 0.61) for herbaceous forage production at RR; year effect significant (P , 0.01). Treatment effect not significant (P = 0.21) for woody leaf production at AP. d Treatment effect significant (P = 0.01) for herbaceous forage production at AP.

Table 3. Woody leaf biomass production (dry matter pounds/acre) of individual species or species groups after implementation of understory fertilization (fert), prescribed burning (burn), and prescribed burning with understory fertilization (burn/fert) in two closed-canopy hardwood stands (Rocky River Hunting Club [RR] and Ames Plantation [AP]) in Tennessee, JulyAugust 2005. Asterisks indicate significant burning 6 fertilization interaction at P , 0.01. Within the row, means are not different (P . 0.05) if followed by the same letter. Raw means are reported for red oak group, but analysis was conducted on transformed data.Treatment Control Speciesa RR Blueberry Sassafras Sourwood Greenbrier Blackberry Red maple Blackgum Red oak group* White oak group Hickory AP Poison ivy* Grape Virginia creeper Ash Honeysuckle Slippery elm Blackgum Winged elm Greenbrier Supplejacka

Fert

Burn

Burn/fert

Effect Burning Fertilization

x (SE) 22.1 (5.4) 7.1 (1.6) 5.2 (2.7) 6.8 (1.7) 1.6 (0.5) 7.7 (2.1) 1.6 (0.4) 4.3 (1.3) AB 12.8 (7.4) 3.7 (1.3) 84.9 (19.0) A 10.2 (4.8) 9.9 (2.3) 7.2 (2.4) 6.6 (1.9) 11.3 (7.2) 8.1 (3.4) 12.0 (4.0) 6.3 (3.7) 0.3 (0.3)

x (SE) 35.0 (5.5) 19.0 (5.2) 4.1 (2.1) 7.3 (1.4) 5.5 (1.7) 10.0 (3.9) 2.6 (1.1) 11.2 (3.6) A 7.4 (3.0) 7.7 (2.3) 29.7 (7.3) B 11.7 (3.8) 6.9 (2.5) 22.2 (6.4) 16.8 (4.6) 7.1 (2.4) 6.6 (5.1) 7.2 (1.9) 6.9 (1.9) 4.8 (1.7)

x (SE) 61.9 (13.0) 55.1 (11.0) 39.1 (20.5) 10.3 (1.5) 0.5 (0.0) 4.4 (2.0) 4.4 (1.3) 12.1 (6.0) A 7.7 (5.0) 2.0 (1.2) 23.6 (6.1) B 16.7 (8.9) 11.6 (3.6) 11.7 (5.9) 3.3 (0.9) 7.4 (3.0) 2.1 (0.9) 3.0 (0.8) 1.2 (0.5) 0.3 (0.3)

x (SE) 42.7 (9.9) 94.7 (19.5) 21.8 (8.5) 7.4 (1.4) 3.4 (1.4) 21.4 (11.3) 9.7 (2.7) 8.8 (7.9) B 8.0 (2.1) 9.7 (4.4) 96.5 (19.0) A 18.3 (7.0) 16.7 (5.2) 3.2 (1.4) 5.9 (1.5) 6.2 (2.6) 6.2 (3.9) 0.6 (0.5) 1.5 (0.7) 4.4 (1.2)

F1,12 = 0.26, P = 0.62 F1,60 = 27.12, P , 0.01 F1,60 = 2.48, P = 0.12 F1,60 = 1.60, P = 0.21 F1,12 = 2.98, P = 0.11 F1,12 = 4.57, P = 0.05 F1,12 = 8.14, P = 0.02 F1,60 = 2.88, P = 0.10 F1,12 = 0.05, P = 0.82 F1,60 = 1.55, P = 0.22 F1,12 = 0.03, P = 0.86 F1,12 = 0.22, P = 0.64 F1,60 = 2.53, P = 0.12 F1,12 = 3.26, P = 0.10 F1,12 = 2.95, P = 0.11 F1,12 = 0.11, P = 0.75 F1,60 = 0.08, P = 0.78 F1,12 = 9.18, P = 0.01 F1,60 = 8.89, P , 0.01 F1,60 = 0.12, P = 0.73

F1,12 = 0.17, P = 0.69 F1,60 = 1.19, P = 0.28 F1,60 = 0.18, P = 0.67 F1,60 = 1.21, P = 0.28 F1,12 = 7.42, P = 0.02 F1,12 = 0.16, P = 0.70 F1,12 = 2.23, P = 0.16 F1,60 = 0.04, P = 0.85 F1,12 = 1.29, P = 0.28 F1,60 = 5.88, P = 0.02 F1,12 = 0.37, P = 0.56 F1,12 = 0.84, P = 0.38 F1,60 = 0.08, P = 0.78 F1,12 = 0.09, P = 0.77 F1,12 = 2.86, P = 0.12 F1,12 = 0.08, P = 0.78 F1,60 = 0.27, P = 0.61 F1,12 = 2.89, P = 0.12 F1,60 = 1.25, P = 0.27 F1,60 = 19.60, P , 0.01

Blueberry Vaccinium spp., sassafras Sassafras albidum, sourwood Oxydendrum arboreum, greenbrier Smilax spp., blackberry Rubus spp., red maple Acer rubrum, blackgum Nyssa sylvatica, red oak Quercus falcata, white oak Quercus alba, hickory Carya spp., poison ivy Toxicodendron radicans, grape Vitis spp., Virginia creeper Parthenocissus quinquefolia, ash Fraxinus spp., honeysuckle Lonicera spp., slippery elm Ulmus rubra, winged elm Ulmus alata, supplejack Berchemia scandens.




C.E. Shaw et al.

Table 4. Nutritional content of species after implementation of understory fertilization (fert), prescribed burning (burn), and prescribed burning with understory fertilization (burn/fert) in two closed-canopy hardwood stands (Rocky River Hunting Club [RR] and Ames Plantation [AP]) in Tennessee, JulyAugust 2005. Dashes indicate composite sample of species collected in a given treatment was not large enough for nutritional analysis. Rose also not included at AP because sample was not large enough for nutritional analysis.Crude protein (%) Species RR Blackgum Red maple Blackberry Sassafras Greenbrier Sourwood Blueberry AP Ash Blackgum Japanese honeysuckle Poison ivy Greenbrier Supplejack Virginia creeper Grape Winged elma a

Neutral detergent fiber (%)

Acid detergent fiber (%) Control Fert Burn Burn/fert

Control Fert Burn

Burn/fert Control Fert Burn Burn/fert

11.9 11.3 13.5 15.2 12.6 11.5 9.4 12.4 10.8 12.6 12.2 12.6 10.4 12.4 12.8

13.7 13.3 14.8 15.1 14.9 12.2 10.6 12.8 11.6 11.7 11.8 13.2 13.0 10.4 11.4 12.7

14.3 15.2 14.9 13.3 11.7 9.7 14.5 13.5 13.3 13.4 12.7 12.6 13.6 15.8

13.4 14.7 11.9 14.0 14.3 14.4 11.0 15.2 13.1 12.1 12.4 14.4 15.0 11.8 13.2 17.7

40.5 44.0 43.9 57.3 47.1 39.3 48.3 56.6 37.9 38.8 45.1 50.2 47.7 45.4 68.2

38.3 41.2 43.0 53.4 46.8 38.3 51.0 57.3 38.5 37.0 45.0 48.8 50.8 42.9 41.1 64.0

36.7 39.2 53.5 44.1 37.3 45.6 58.7 37.7 35.9 48.5 50.4 40.9 41.5 66.9

38.8 43.0 41.6 55.7 46.6 39.7 50.1 54.5 37.0 35.8 46.4 45.2 42.6 40.5 40.3 62.0

20.4 28.3 21.9 42.1 30.7 23.8 33.2 38.7 21.9 26.7 29.7 30.4 32.9 33.1 28.2

26.2 25.4 24.3 46.1 28.4 22.2 37.6 38.4 21.2 24.5 30.8 29.9 20.7 33.7 33.2 31.0

16.3 25.3 36.0 31.6 21.8 30.0 40.3 23.4 26.1 32.8 35.0 30.4 32.5 29.8

21.9 31.7 23.8 40.2 44.0 25.1 34.4 39.8 22.2 25.9 32.9 34.0 25.0 32.7 29.2 23.8

Blackgum Nyssa sylvatica, red maple Acer rubrum, blackberry Rubus spp., sassafras Sassafras albidum, greenbrier Smilax spp., sourwood Oxydendrum arboreum, blueberry Vaccinium spp., ash Fraxinus spp., Japanese honeysuckle Lonicera japonica, poison ivy Toxicodendron radicans, supplejack Berchemia scandens, Virginia creeper Parthenocissus quinquefolia, grape Vitis spp., winged elm Ulmus alata.

sassafras, Carolina buckthorn Rhamnus caroliniana, common persimmon Diospyros virginiana, or devils walkingstick Aralia spinosa. Effect on NCC Although no treatment effects on individual species (Table 3) or nutritional quality (Table 4) were detected, prescribed fire and fertilization increased NCC estimates at Rocky River (Table 5). Conversely, although no treatment significantly affected forage production at Ames Plantation, estimates of deer days/acre were decreased after prescribed fire.

DiscussionAlthough others have noted changes in pH after prescribed fire (Binkley 1986; Blankenship and Arthur 1999), our results did not reveal an effect of fire on pH, which was consistent with Franklin et al. (2003). Although using ammonium nitrate fertilizers may lower pH if used annually, pH changes after infrequent fertilization are usually negligible (Fisher and Binkley 2000). Differences in soil potash responses at Ames Plantation and Rocky River may be a result of differences in soil

Table 5. Nutritional carrying capacity (deer days/acre, assuming 3 lb/d consumption) of selected white-tailed deer Odocoileus virginianus forages combined to average 12% crude protein after implementation of understory fertilization (fert), prescribed burning (burn), and prescribed burning with understory fertilization (burn/fert) in closed-canopy hardwood stands at Rocky River Hunting Club (RR) in Sequatchie County, Tennessee, and Ames Plantation (AP) in Fayette County, Tennessee, JulyAugust 2005. Interaction between burning and fertilization not significant (P . 0.05) at either site.Treatment Control Site RR AP Deer days per acre per acre Fert Burn Burn/fert Effect Burning F1,60 = 4.44, P = 0.04 F1,60 = 6.11, P = 0.02 Fertilization F1,60 = 4.70, P = 0.03 F1,60 = 1.96, P = 0.17

x (SE) 2.8 (0.6) 6.9 (1.7)

x (SE) 4.6 (0.8) 8.5 (2.6)

x (SE) 4.6 (0.6) 2.1 (0.5)

x (SE) 6.3 (1.2) 4.2 (1.2)




C.E. Shaw et al.

texture. The sandy loam at Ames Plantation contained less clay than the silt loam at Rocky River. As summer progressed, the greater clay content and cation exchange capacity of the silt loam probably contributed to the observed decline in potash ratings during the late summer sampling period because clay particles attract more of the free K ions. Plant response to fertilization can be expected to vary among different soil types. Past research has documented increases in browse production after fertilization, but any increase is certainly buffered by available sunlight. Segelquist and Rogers (1975) and Dyess et al. (1994) reported increased production of Japanese honeysuckle after applications of lime and N fertilization, but their plots were located in cleared openings. Production of Japanese honeysuckle did not increase after fertilization at Ames Plantation. Increases in NCC estimates at Rocky River after fertilization were significant but arguably not worth the cost. Fertilizers in our study were US$0.22/lb (3400), US$0.31/lb (0460), and US$0.28/lb (0060). Average fertilization costs for rates of N (US$28.46/acre), P (US$41.78/acre), and K (US$72.17/acre) applied totaled US$142.40/acre. The only increase in browse selected by deer after fertilization was blackberry, which increased 4.0 lb/acre at Rocky River, costing US$35.60/lb in fertilized sections. With the greatest increase of 3.5 deer days/acre (6.3 deer days/acre after prescribed burning with understory fertilization compared to 2.8 deer days/ acre in control) at Rocky River and associated fertilization rates and costs (US$173.67/acre) used in our study, it would cost US$49.62 for each additional deer day. Substantive changes in the structure and composition of understory vegetation usually necessitate several successive fires and are also influenced by season of fire and fire intensity (Brockway and Lewis 1997; Sparks et al. 1998; Hutchinson and Sutherland 2000; Peterson and Reich 2001; Glasgow and Matlack 2007; Jackson et al. 2007). Substantive changes after repeated burning are strongly correlated with increased sunlight entering the forest canopy. Our data represent the initial effect of prescribed fire in closed-canopy stands. With repeated low-intensity burning, mortality of the midstory may allow increased sunlight, which could lead to increased browse production. However, it is likely that landowners managing their property for wildlife would like to see a more timely response to their management efforts. It is important to understand the approach we used to estimate NCC is not an absolute measure of carrying capacity. However, it does allow relative comparisons among treatments by using a biologically defensible diet-quality target using species or species groups selected by deer during the growing season. This approach is important because forage quality has a tremendous influence on available nutrition. By combining selected deer forages to average a minimum of 12% CP, our data suggested NCC was actually negatively influenced by burning at Ames Plantation, although the woody leaf biomass was not significantly decreased. Although forage quality is important, browse species selected by deer influences NCC estimates more than increases in CP values. Although CP values in mostJournal of Fish and Wildlife Management | www.fwspubs.org

treatments were slightly higher than those in control areas, only control areas at Ames Plantation had reductions in NCC estimates attributable to the minimum criteria for CP (12%). Managers should only use results from diet studies as general guidelines for deer use of various species and evaluate treatment effects on browse species in relation to actual deer use on specific areas. The response of herbaceous species to treatments suggested their inclusion in NCC estimates would not have affected our results. At both study areas, sites that were burned and fertilized produced greater amounts of herbaceous forages than other treatments. However, nonpreferred species, such as American burnweed Erechtites hieraciifolia and grasses, contributed almost all of this production. On other sites with a different seedbank, a response by desirable forage species may increase NCC.

Management RecommendationsPrescribed burning and understory fertilization produced mixed effects in two physiographic regions with different soil types in Tennessee one growing season after treatment. Therefore, we caution against the use of low-intensity prescribed fire in closed-canopy stands with the objective of increasing browse for deer. Although browse production may increase during subsequent growing seasons or after additional fire prescription, we recommend some canopy reduction treatment (e.g., retention cutting and thinnings) to allow additional sunlight into the stand before burning (Healy 1997; Jackson et al. 2007), especially if a relatively quick and positive treatment effect is desired. We do not recommend understory fertilization in closed-canopy hardwood stands because plant response was minimal, and the relatively small increase makes it difficult to justify the cost of fertilization. Liming before fertilization could improve pH and nutrient availability, but application of lime in forested areas is generally not practical because of difficulty spreading lime in the woods, amount of lime needed to correct soil acidity, and associated costs. We believe money spent on liming and fertilization would be much more efficiently and effectively spent on food plot plantings.

Supplemental MaterialPlease note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author. Table S1. Selection index value data. Found at DOI: 10.3996/102009-JFWM-016.S1 (182 KB XLS).

AcknowledgmentsFunding and support for this study were provided by the Department of Forestry, Wildlife, and Fisheries at the University of Tennessee; Hobart Ames Foundation;November 2010 | Volume 1 | Issue 2 | 70


C.E. Shaw et al.

Sequatchie Forest and Wildlife; Quality Deer Management Association; and Tennessee Wildlife Resources Agency. Logistic support was provided by Benny Bowers, Carla Dilling, Jesus Gamboa, John Gruchy, Greg Julian, James McDonald, Larry Teague, and Shelton Whittington. We thank the Subject Editor and two anonymous reviewers, who helped improve the quality of this manuscript.

References[AOAC] Association of Official Analytical Chemists. 1998. Crude protein in animal feed combustion method. 16th edition, 4th revision, Chapter 4. Pages 1819 in Cunniff P, editor. Official Methods of Analysis of AOAC International. Arlington, Virginia: AOAC International. Binkley D. 1986. Soil acidity in loblolly pine stands with interval burning. Soil Science Society of America Journal 50:15901594. Blankenship BA, Arthur MA. 1999. Soil nutrient and microbial response to prescribed fire in an oak-pine ecosystem in eastern Kentucky. Pages 3947 in Stringer JW, Loftis DL, editors. Asheville, North Carolina: U.S. Forest Service Southern Research Station. Proceedings of 12th Central Hardwood Forest Conference General Technical Report SRS-24. Available: http://www.srs.fs. usda.gov/pubs/732 (August 2010). Blymyer MJ, Mosby HS. 1977. Deer utilization of clearcuts in southwestern Virginia. Southern Journal of Applied Forestry 1:1013. Brockway DG, Lewis CE. 1997. Long-term effects of dormant-season prescribed fire on plant community diversity, structure and productivity in a longleaf pine wiregrass ecosystem. Forest Ecology and Management 96:167183. Chesson J. 1978. Measuring preference in selective predation. Ecology 59:211215. Chesson J. 1983. The estimation and analysis of preference and its relationship to foraging models. Ecology 64:12971304. Dills GG. 1970. Effects of prescribed burning on deer browse. Journal of Wildlife Management 34:540 545. Dyess JG, Causey MK, Stribling HL, Lockaby BG. 1994. Effects of fertilization on production and quality of Japanese honeysuckle. Southern Journal of Applied Forestry 18:6871. Edwards SL, Demarais S, Watkins B, Strickland BK. 2004. White-tailed deer forage production in managed and unmanaged pine stands and summer food plots in Mississippi. Wildlife Society Bulletin 32:739745. Fisher RF, Binkley D. 2000. Ecology and management of forest soils. 3rd edition. New York: John Wiley and Sons. Flowers RL. 1964. Soil survey of Fayette County, Tennessee. Washington, D.C.: U.S. Department of Agriculture. Franklin SB, Robertson PA, Fralish JS. 2003. Prescribed burning effects on upland Quercus forest structureJournal of Fish and Wildlife Management | www.fwspubs.org

and function. Forest Ecology and Management 184: 315335. Glasgow LS, Matlack GR. 2007. Prescribed burning and understory composition in a temperate deciduous forest, Ohio, USA. Forest Ecology and Management 238:5464. Harlow RF, Pinkerton BW, Guynn Jr DC, Williams Jr JG. 1993. Fertilizer effects on the quality of white-tailed deer forages on utility rights-of-way. Southern Journal of Applied Forestry 17:4953. Healy WM. 1997. Thinning New England oak stands to enhance acorn production. Northern Journal of Applied Forestry 14:152156. Hobbs NT, Swift DM. 1985. Estimates of habitat carrying capacity incorporating explicit nutritional constraints. Journal of Wildlife Management 49:814822. Hutchinson TF, Sutherland S. 2000. Fire and understory vegetation, a large-scale study in Ohio and a search for general response patterns in central hardwood forests. Pages 6474 in Yaussy D, compiler. Newtown Square, Pennsylvania: U.S. Forest Service Northern Research Station. Proceedings: workshop on fire, people, and the central hardwoods landscape General Technical Report NE-274. Available: http://www.fs.fed. us/ne/newtown_square/publications/technical_reports/pdfs/ 2000/274%20papers/hutchinson274.pdf (August 2010). Jackson SW, Basinger RG, Gordon DS, Harper CA, Buckley DS, Buehler DA. 2007. Influence of silvicultural treatments on eastern wild turkey habitat characteristics in eastern Tennessee. Proceedings of the National Wild Turkey Symposium 9:199207. (Available from the lead author.) Jacobson HA, Kroll JC, Browning RW, Koerth BH, Conway MH. 1997. Infrared-triggered cameras for censusing white-tailed deer. Wildlife Society Bulletin 25:547556. Jung HJG. 1997. Analysis of forage fiber and cell walls in ruminant nutrition. Journal of Nutrition 127:S810S813. Masters RE, Lochmiller RL, Engle DM. 1993. Effects of timber harvest and prescribed fire on white-tailed deer forage production. Wildlife Society Bulletin 21: 401411. Measells MK, Grado SC, Hughes HG, Dunn MA, Idassi J, Zielinske B. 2005. Nonindustrial private forest landowner characteristics and use of forestry services in four southern states: results from a 20022003 mail survey. Southern Journal of Applied Forestry 29:194199. Measells MK, Grado SC, Hughes HG, Dunn MA, Idassi JO, Zielinske RJ. 2006. Educational needs of southern forest landowners. Journal of Extension 44(5). Article 5RIB4. Available: http://www.joe.org/joe/2006october/ rb4.php (August 2010). Miller BF, Campbell TA, Laseter BR, Ford WM, Miller KV. 2009. White-tailed deer herbivory and timber harvesting rates: implications for regeneration success. Forest Ecology and Management 258:10671072. Peterson DW, Reich PB. 2001. Prescribed fire in oak savanna: fire frequency effects on stand structure and dynamics. Ecological Applications 11:914927.November 2010 | Volume 1 | Issue 2 | 71


C.E. Shaw et al.

Prater JL. 2003. Soil survey of Sequatchie County, Tennessee. Washington, D.C.: U.S. Department of Agriculture. Schweiss BE, Dwyer J. 2008. Landowner attitudes and perceptions of forest and wildlife management in rural northern Missouri. Pages 297305 in Jacobs DF, Michler CH, editors. Newton Square, Pennsylvania: U.S. Forest Service Northern Research Station. Proceedings of 16th Central Hardwood Forest Conference General Technical Report NRS-P-24. Available: http://www.nrs. fs.fed.us/pubs/gtr/gtr-p-24%20papers/34schweiss-p-24. pdf (August 2010).

Segelquist CA, Rogers MJ. 1975. Response of Japanese honeysuckle to fertilization. Journal of Wildlife Management 39:769775. Sparks JC, Masters RE, Engle DM, Palmer MW, Bukenhofer GA. 1998. Effects of late growing-season and late dormant-season prescribed fire on herbaceous vegetation in restored pine-grassland communities. Journal of Vegetation Science 9:133142. Wood GW. 1986. Influences of forest fertilization on South Carolina deer forage quality. Southern Journal of Applied Forestry 10:203206.



Articles

MatchMismatch Regulation for Bluegill and Yellow Perch Larvae and Their Prey in Sandhill LakesJeffrey C. Jolley,* David W. Willis, Richard S. Holland J.C. Jolley, D.W. Willis Department of Wildlife and Fisheries Sciences, South Dakota State University, SNP 138, Box 2140B, Brookings, South Dakota 57007 Present address of J.C. Jolley: U.S. Fish and Wildlife Service, Columbia River Fisheries Program Office, 1211 SE Cardinal Court, Vancouver, Washington 98683 R.S. Holland Nebraska Game and Parks Commission, P.O. Box 30370, Lincoln, Nebraska 68701

AbstractFood availability may regulate fish recruitment, both directly and indirectly. The availability of zooplankton, especially to newly hatched larvae, is thought to be crucial to their early growth and survival. We examined stomach contents of larval bluegill Lepomis macrochirus and yellow perch Perca flavescens in Pelican Lake and Cameron Lake, Nebraska, in 2004 and 2005. We also determined zooplankton availability and calculated prey selection using Chessons a. In addition, we investigated potential matchmismatch regulation of recruitment from 2004 to 2008. Bluegill positively selected copepod nauplii and Bosmina spp., and yellow perch often selected copepods. Abundant zooplankton populations were available for consumption. Matches of both larval bluegill and yellow perch abundance to zooplankton abundance were detected in all years; exact matches were common. Mismatches in predator and prey production were not observed. Predation by age-0 yellow perch on age-0 bluegill was not observed, even though yellow perch hatched 2 mo prior to bluegill. Given that zooplankton were abundant and well-timed to larval fish relative abundance over the time span of this study, the matchmismatch hypothesis alone may not fully account for observed recruitment variability in these populations. Environmental conditions may also affect recruitment and warrant further investigation. Keywords: Sandhills lakes; Valentine National Wildlife Refuge; predatorprey dynamics; recruitment; phenology; survival; food habits Received: June 30, 2010; Accepted: August 25, 2010; Published Online Early: August 2010; Published: November 2010 Citation: Jolley JC, Willis DW, Holland RS. 2010. Matchmismatch regulation for bluegill and yellow perch larvae and their prey in Sandhill lakes. Journal of Fish and Wildlife Management 1(2):7385; e1944-687X. doi: 10.3996/062010JFWM-018 Copyright: All material appearing in the Journal of Fish and Wildlife Management is in the public domain and may be reproduced or copied without permission. Citation of the source, as given above, is requested. The findings and conclusions in this article are those of the author(s) and do not necessarily represent the views of the U.S. Fish and Wildlife Service. * Corresponding author: jeffrey_ [email protected]

IntroductionRecruitment dynamics of populations will ultimately structure fish communities (Diana 1995) and several factors are believed to regulate recruitment, both directly and indirectly. These include abiotic factors such as physical habitat, temperature, and weather (Beard 1982; Jackson and Noble 2000), biotic factors such as food availability and competition (Prout et al. 1990; Welker et al. 1994; Ludsin and DeVries 1997; Bunnell et al. 2003), and predation (Forney 1971; Houde 1987; Rice et al. 1987; Santucci and Wahl 2003). Critical time periods, orJournal of Fish and Wildlife Management | www.fwspubs.org

bottlenecks, of high mortality may exist for some species (Hjort 1914; May 1974) and researchers commonly incorporate this factor when describing recruitment processes (Marr 1956). This critical period is thought to occur early in age-0 yellow perch Perca flavescens and bluegill Lepomis macrochirus cohort development (Toetz 1966; Forney 1971; Clady 1976; Anderson et al. 1998). Large annual variation in yellow perch year-class strength is common (Hamley et al. 1983; Henderson 1985) although recruitment patterns may vary among water bodies within a localized region (Lucchesi 1991; Isermann et al. 2007). The early life-stages are commonlyNovember 2010 | Volume 1 | Issue 2 | 73

MatchMismatch of Bluegill and Yellow Perch and Their Prey

J.C. Jolley et al.

reported as the time when year-class strength is formed (Clady 1976; Mills and Forney 1981). Variable recruitment is often observed in bluegill populations. Previous research in South Dakota indicated that bluegill recruitment was asynchronous among four small impoundments (Edwards et al. 2007) and the authors suggested that biotic factors likely affected bluegill recruitment. Although bluegill recruitment is generally consistent (i.e., missing year-classes are rare) in Nebraska Sandhill lakes, there appears to be a moderate level of variability in relative year-class strength (Paukert et al. 2002a; Jolley 2009). The availability of prey when larval fish begin exogenous feeding has been proposed as a potential regulator of recruitment variability (matchmismatch hypothesis; Cushing 1975, 1990). Suitable growth and prey conditions for larval fish are largely determined by the physical environment (e.g., water temperature). Water temperature may indirectly affect larval fish via its influence on the food chain (e.g., zooplankton growth; Sommer et al. 1986; Taylor et al. 1987) in addition to directly mediating spawning and hatching (Beard 1982). The matchmismatch hypothesis consists of two assumptions: first, that fish at temperate latitudes spawn at approximately the same time each year and, second, that the larvae are released during the spring or autumn peaks in plankton production (Cushing 1990). A match occurs when production of fish larvae and their prey is synchronous or nearly so; conversely, a mismatch occurs when there is a large temporal difference in these two variables (Cushing 1990). Although the food habits of age-0 yellow perch have been extensively studied in Midwestern waters (e.g., Weber and Les 1982; Wahl et al. 1993; Fisher and Willis 1997), lake-specific prey-selection patterns in the face of variable prey densities is a topic of importance. Similarly, age-0 bluegill food habits are well-studied (Werner 1969; Mittelbach 1981; Werner and Hall 1988), although data specific to Sandhill lakes have not been collected. Furthermore, the panfish populations of Nebraska Sandhill lakes are high quality (i.e., relatively high abundance of large fish; Paukert et al. 2002b) and relatively unique to the Great Plains. Therefore, examinations of processes that affect recruitment of these populations are of direct interest to managers. The objectives of this study were to 1) describe prey selection for larval and juvenile yellow perch in Cameron and Pelican lakes and bluegill in Pelican Lake, Nebraska, to better understand the prey utilization in these lakes, and 2) examine potential matchmismatch regulation of yellow perch and bluegill recruitment.

2009). The fish communities of both lakes were relatively simple. Cameron Lake contained yellow perch, green sunfish Lepomis cyanellus, black bullhead Ameiurus melas, common carp Cyprinus carpio, fathead minnow Pimephales promelas, and golden shiner Notemigonus crysoleucas. Pelican Lake contained bluegill, largemouth bass Micropterus salmoides, yellow perch, northern pike Esox lucius, black bullhead, common carp, and fathead minnow. The watersheds for both lakes were primarily mixed- and tall-grass prairie and were used for limited livestock grazing (Bleed and Flowerday 1989).

MethodsWe obtained larval yellow perch from Cameron and Pelican lakes and larval bluegill from Pelican Lake using a single 1,000-mm-mesh conical ichthyoplankton net (mouth : net length ratio = 1:3) with a 0.76-mdiameter circular steel frame and 500-mL collection bucket (with 500-mm mesh). Isermann et al. (2002) found no difference in yellow perch density estimates between a 500- and 1,000-mm-mesh trawl, although the 1,000-mm size was less likely to become fouled with algae and zooplankton. We sampled larvae approximately every 10 d from late April to early September in Pelican Lake, 20042008 and from late April to mid-June in Cameron Lake, 20042005. Randomly chosen, paired locations (i.e., nearshore and offshore) were trawled in Pelican Lake (n = 5) and Cameron Lake (n = 4) on each occasion. The density of recently hatched (i.e., , 8 mm) bluegill and yellow perch larvae in the lakes was indexed using a flowmeter (Ocean Test Equipment, Inc.) in the mouth of the trawl, which allowed determination of water volume towed. We collected zooplankton at the time of each trawling sample during the daytime as two replicates at each site using a 2-mlong tube sampler (Rabeni 1996). Samples were filtered through a 65-mm-mesh net. Replicate samples were collected and processed separately. All samples were preserved in 90% ethanol and transported to the laboratory for identification and diet analysis. We tracked the same cohorts by sampling juvenile yellow perch from Cameron Lake in August and juvenile bluegill and yellow perch from Pelican Lake in August or September and the following April or May (age 1) using cloverleaf traps. Each three-lobed cloverleaf trap was constructed of galvanized 6.4-mm-bar mesh, with three 12.7-mm-wide openings between lobes to accommodate entrance of small yellow perch (Brown and St. Sauver 2002). Each lobe was 50 cm in diameter with a 41cm height. Collected fishes were preserved in 90% ethanol and returned to the laboratory. Autumn (age 0) and spring (age 1) juvenile abundance was indexed as the mean number per cloverleaf trap-night. Year-class strength was assessed in a concurrent study and information from the adult populations was used to examine recruitment; methods are described by Jolley (2009). Adult bluegill (i.e., age 2 and older) and yellow perch (i.e., age 1 and older) were collected annually from each lake using randomly located, overnight sets of double-throated trap (i.e., modified fyke) nets withNovember 2010 | Volume 1 | Issue 2 | 74

Study AreaCameron Lake (39 ha) and Pelican Lake (322 ha) are shallow (mean depth = 1.2 m and 1.3 m, respectively), windswept natural lakes in the Sandhills region of northcentral Nebraska (McCarraher 1977). Submergent and emergent vegetation coverage was low (, 17% total coverage) in Cameron Lake and was moderate (4052% total coverage) in Pelican Lake in 2004 and 2005 (JolleyJournal of Fish and Wildlife Management | www.fwspubs.org


J.C. Jolley et al.

Table 1. Number (No.) and mean total length (TL; mm) of yellow perch Perca flavescens and bluegill Lepomis macrochirus examined for food items from Cameron and Pelican lakes, Nebraska, 20042005.Lake Cameron Yellow perch 2004 13 May 18 May 27 May 7 June 2005 15 May 24 May 2 June Pelican Yellow perch 2004 8 May 17 May 26 May 6 June 16 June 26 June 7 July 17 July 27 July 6 August 16 August 26 August 2005 Pelican Bluegill 2004 26 June 7 July 17 July 27 July 6 August 16 August 27 August 5 September 2005 30 June 11 July 21 July 1 August 11 August 22 August 3 9 30 30 30 30 30 30 30 30 30 30 30 30 11 14 47 30 32 30 30 31 32 31 31 35 30 30 7.5 9.1 10.8 15.0 15.0 19.1 20.7 19.5 8.3 10.7 10.4 11.4 11.2 12.9 0.4 0.6 0.4 0.3 0.6 1.0 1.0 1.5 0.2 0.4 0.4 0.4 0.4 0.4 6.611.0 6.112.4 3.815.5 8.618.5 7.323.3 10.929.1 7.929.3 11.240.1 7.212.3 7.814.7 7.414.1 8.015.6 7.315.3 9.818.4 73 36 36 0 6 0 0 3 6 3 3 14 0 0 3 May 30 30 30 23 30 17 7 3 9 14 13 9 16 35 30 30 24 31 17 7 3 9 14 13 9 16 8.8 12.7 17.1 23.1 32.2 28.0 31.8 41.8 49.3 57.6 59.8 59.9 5.4 0.2 0.2 0.4 0.5 0.7 1.7 2.5 3.7 1.0 2.3 0.9 2.6 0.1 6.813.8 10.414.5 12.720.6 18.126.9 19.537.3 18.939.5 22.036.8 34.646.5 46.053.0 40.269.5 56.267.1 44.674.1 4.26.5 14 0 0 4 3 0 0 0 0 0 0 0 100 30 30 30 30 18 30 12 30 30 30 30 18 30 12 11.5 12.5 15.7 25.0 8.3 14.5 19.0 0.2 0.2 0.5 0.5 0.4 0.2 0.6 9.913.1 9.114.9 10.019.6 19.430.0 4.212.8 12.218.3 16.023.1 0 0 0 0 0 0 0 Date No. (with prey) No. examined x TL SE Range % Empty

16-mm-bar measure mesh, 1.1 6 1.5-m frames, and 22-m leads. Age was estimated by two independent readers viewing the sagittal otolith in whole view for fish of ages 4 and younger; older fish were aged after cracking and sanding the otoliths. Discrepancies in age estimates were reconciled by reading the otolith in concert. If agreement could not be achieved, the otolith was omitted from the analysis. Mean catch per unit effortJournal of Fish and Wildlife Management | www.fwspubs.org

values in May or June for age-1 yellow perch and age-2 bluegill were selected as life-stages recruited to the adult population. Larval fish samples were sorted and identified using identification keys (Auer 1982; Holland-Bartels et al. 1990). Larvae were counted and up to 200 fish per sample were measured (mm total length [TL]) from each site. Digestive tracts were removed from up to 30November 2010 | Volume 1 | Issue 2 | 75


J.C. Jolley et al.

Table 2. Percent occurrence and percent by number for zooplankton prey items found in larval yellow perch Perca flavescens stomachs in Cameron Lake, Nebraska, in 2004 and 2005.2004 Taxon % Occurrence Bosminidae Chydoridae Daphnidae Sididae Copepoda Nauplii Ostracoda Rotifera % by number Bosminidae Chydoridae Daphnidae Sididae Copepoda Nauplii Ostracoda Rotifera 0.2 0.5 26.8 0 61.1 0 0 11.3 2.3 1.5 37.5 1.3 33.2 0 0.2 24.1 2.5 1.1 81.7 2.5 12.1 0 0 0.1 0.3 34.4 60.2 0 5.1 0 0.1 0 0 0 24.4 0 75.6 0 0 0 0.2 0 28.8 0 71.0 0 0 0 0.3 0.5 38.6 0 60.6 0 0 0 3.3 6.7 86.7 0 100.0 0 0 33.3 33.3 20.0 100.0 23.3 96.7 0 3.3 66.7 30.0 26.7 100.0 30.0 66.7 0 0 3.3 20.0 66.7 100.0 0.0 60.0 0 3.3 0 0 0 33.3 0 94.4 0 0 0 3.3 0 90.0 0 100.0 0 0 0 16.7 16.7 100.0 0 91.7 0 0 0 13 May 18 May 27 May 7 June 15 May 2005 24 May 2 June

randomly selected larvae of each species per sampling occasion during 2004 and 2005 at Cameron and Pelican lakes. Diet items were identified using a microscope, counted, and measured (mm TL). Zooplankton were enumerated and identified to family for common cladocerans (i.e., Bosminidae, Chydoridae, and Daphnidae), and as cyclopoid or calanoid copepods, copepod nauplii (copepodites), ostracods, and rotifers. Each sample was diluted with water to a volume of 30 mL. Three subsamples were then taken with a 5-mL Hensen Stempel pipette and placed in a Ward counting wheel. Up to 20 individuals of each category were measured and all individuals were counted. The total number of zooplankton of each taxon in a sample was calculated by dividing the number of organisms counted by the proportion of the sample volume processed. Density was then calculated by dividing the number of zooplankton of each taxon by the volume of the water filtered with the tube sampler. To determine prey selectivity, mean Chessons (1983) coefficient of selectivity (a) was calculated for individual larval bluegill and yellow perch from each sampling occasion in 2004 and 2005: a~ ri =ni m X ri =nii~1

in 2004 and 2005 were examined and used to select prey items included in the matchmismatch analysis. Temporal predator and prey density curves for yellow perch and bluegill larvae and their prey were constructed for each year and the mean and standard deviation of the peak density was calculated. Prey types used in the analyses were chosen as those showing positive selection by fish during first-feeding (earlier in the season). The width and overlap of the density curves for predator and prey were calculated. Methods outlined by Mertz and Myers (1994) and Johnson (2000) were utilized for this analysis. The following parameters were first calculated: t0 ~timing between peaks of larval production and food supplydays; Dt0 ~annual differences int0 from its mean value; d~one half width of the density curve for larvae; and s~one half width of the density curve for zooplankton: These parameters were used to calculate 1) variability in peak timing from the mean for individual species (s and d), 2) variability in timing between larval abundance and peak zooplankton production (t0), and 3) year-toyear variability in peak spawning and production (Dt0). When t0 = 0, the match is exact (Mertz and Myers 1994). A mismatch occurs when one-half the width of the larval density curve (i.e., d) does not overlap one-half the width of the zooplankton density curve (i.e., s). CorrelationNovember 2010 | Volume 1 | Issue 2 | 76

where ri is the number of food type i in the predator diet, ni is the number of food type i in the environment, and m is the number of prey types available. Mean a values (6 95% CI) were compared with nonselective feeding (1/m) to determine selectivity. Patterns in prey selectionJournal of Fish and Wildlife Management | www.fwspubs.org


J.C. Jolley et al.

Figure 1. Mean (with 95% CI) prey selection (Chessons a) by yellow perch Perca flavescens in Cameron Lake, Nebraska, 2004 2005 by prey category of zooplankton. Confidence intervals above the random feeding (dashed) line indicate positive selection, values below the line indicate negative selection, and values overlapping the line indicate neutral selection.

analysis was used to examine the relationship between predatorprey overlap (t0) and indices of fish recruitment (i.e., larval abundance, juvenile abundance, and adult abundance) for yellow perch and bluegill in Pelican Lake from 2004 to 2008. Yellow perch from Cameron Lake were omitted from this analysis due to inadequate sample size (N = 2 y).

ResultsIn Cameron Lake, the most abundant zooplankters were daphnids (Supplemental Material, Table S1, http:// dx.doi.org/10.3996/062010-JFWM-018.S1). Chessons a . 0.125 indicated positive prey selection by yellow perch. Stomachs from 120 yellow perch larvae in May and June of 2004 and 60 stomachs in 2005 were examined for prey items (Table 1; Supplemental Material, Table S2, http:// dx.doi.org/10.3996/062010-JFWM-018.S2). Copepods and daphnids were consistently the most abundant prey item found in stomachs (Table 2). Temporal patterns of negative, neutral, and positive prey selection (Chessons a . 0.143 indicated positive prey selection) were found for yellow perch in Cameron Lake (Figure 1) in 2004 and 2005. In 2004, yellow perch selected copepods earliest in theJournal of Fish and Wildlife Management | www.fwspubs.org

season and later shifted to cladocerans. In 2005, copepods were exclusively positively selected (Figure 1). Empty stomachs were not encountered in either year and copepodites were not observed in stomachs in any year. In Pelican Lake in 2004, the most abundant zooplankters were generally daphnids, copepods, and copepodites. In 2005, the most abundant zooplankters were generally cladocerans (i.e., daphnids, chydorids, and bosminids). In 2004, 222 yellow perch larvae were examined from May through August for prey items (Table 1). Seven empty stomachs were encountered (3%). In 2005, 16 larval yellow perch were captured, all with empty stomachs. In 2004, copepods, daphnids, and chydorids were the most abundant prey items found in yellow perch stomachs (Table 3). Yellow perch in Pelican Lake selected copepods earliest, shifted to cladocerans, and later exhibited neutral prey selection for most prey items (Figure 2). Copepodites were not present and rotifers were uncommon in yellow perch stomachs. Temporal patterns of negative, neutral, and positive prey selection (Chessons a . 0.143 indicated positive prey selection) were also found for bluegill in Pelican Lake in 2004 and 2005 (Figure 3). Bluegill larvae had 15% and 5% empty stomachs in 2004 and 2005, respectively (Table 1). Bluegill consumed cladocerans, copepods, copepodites, ostracods, and rotifers. Copepodites and Bosmina were common in first-feeding bluegill stomachs while cladocerans and copepods became more common later in the season (Table 4). Copepodites were initially neutrally selected and bluegill eventually preferred cladocerans followed by copepods as prey items. Rotifers and ostracods were consistently neutrally or negatively selected (Figure 3) indicating opportunistic feeding or avoidance of these prey items. Copepods were elected as an important (i.e., positively selected) prey resource for first-feeding yellow perch in both lakes. The duration of larval yellow perch (TL , 13 mm) abundance varied from 1 d in 2005 (Pelican Lake) to 31 d in 2008 (Pelican Lake). The duration of peak copepod abundance varied from 9 (Cameron Lake, 2005) to 38 d (Pelican Lake, 2004). The mean annual difference in day of peak abundance for yellow perch larvae and their copepod prey (t0) was 7.1 d (Table 5). Copepods peaked after yellow perch larvae in three instances and exactly matched in four instances (Figures 4 and 5; Table 5). Copepodites and Bosmina were important (i.e., positively selected) prey item for first-feeding bluegill and the timing of their combined abundance was examined. The duration of larval bluegill abundance in Pelican Lake ranged from 21 (2004) to 63 d (2006; Figure 6). The duration of peak copepoditesBosmina abundance ranged from 42 (2006) to 72 d (2004). The mean annual difference in peak abundance date for bluegill predators and their prey (t0) was 14 d. Larval bluegill peaked in abundance before their prey in most years. In 2004, there were two peaks in larval bluegill abundance and zooplankton prey peaked between these two dates (Figure 6; Table 5). Matches between larval abundance and zooplankton prey occurred in all years (i.e., d and s overlapped; Table 5) for both species. There were exact matches (i.e.,November 2010 | Volume 1 | Issue 2 | 77


J.C. Jolley et al.

Table 3. Percent occurrence and percent by number for zooplankton prey items found in larval yellow perch Perca flavescens stomachs in Pelican Lake, Nebraska, in 2004.Taxon % Occurrence Bosminidae Chydoridae Daphnidae Sididae Copepoda Nauplii Ostracoda Rotifera % by number Bosminidae Chydoridae Daphnidae Sididae Copepoda Nauplii Ostracoda Rotifera 0 0 0 0 100.0 0 0 0 0 0.8 15.5 0 83.0 0.8 0 0 0 0 97.0 0 3.0 0 0 0 0 0 98.8 0 1.2 0 0 0 0 10.3 2.3 0 50.1 0 37.3 0 0 44.5 33.2 0 15.0 0 7.4 0 0.6 19.5 18.5 0 60.8 0 0.6 0 19.2 28.3 12.1 0 40.4 0 0 0 39.5 10.2 31.3 0.8 16.4 0 1.8 0 74.0 3.3 14.1 1.2 6.6 0 0.8 0 84.8 4.5 9.3 0.1 1.3 0 0 0 73.0 9.5 16.9 0 0.6 0 0 0 0 0 0 0 100.0 0 0 0 0 3.3 70.0 0 100.0 6.7 0 0 0 0 100.0 0 13.3 0 0 0 0 0 100.0 0 17.4 0 0 0 0 50.0 26.7 0 70.0 0 50.0 0 0 64.7 41.2 0 58.8 0 47.1 0 14.3 71.4 42.9 0 71.4 0 14.3 0 33.3 66.7 33.3 0 66.7 0 0 0 37.5 75.0 62.5 12.5 75.0 0 62.5 0 57.1 50.0 42.9 21.4 57.1 0 57.1 0 100.0 100.0 100.0 76.9 76.9 0 38.5 0 77.8 77.8 77.8 0 22.2 0 11.1 0 8 May 17 May 26 May 6 June 16 June 26 June 7 July 17 July 27 July 6 16 26 August August August

t0 = 0) in 4 y for yellow perch. No correlations between predatorprey overlap (t0) and recruitment indices were significant for yellow perch or bluegill (P . 0.05; Table 6). For both species the highest observed abundance occurred in a year where the predators temporally matched their prey (Table 5).

DiscussionLarval fish predators and their zooplankton prey were temporally well-matched in our study. Prey densities were adequate for larval yellow perch and bluegill. Although a low number of yellow perch were detected in

Figure 2. Mean (with 95% CI) prey selection (Chessons a) by yellow perch Perca flavescens in Pelican Lake, Nebraska, 2004 by prey category of zooplankton. Confidence intervals above the random feeding (dashed) line indicate positive selection, values below the line indicate negative selection, and values overlapping the line indicate neutral selection.Journal of Fish and Wildlife Management | www.fwspubs.org November 2010 | Volume 1 | Issue 2 | 78


J.C. Jolley et al.

Figure 3. Mean (with 95% CI) prey selection (Chessons a) by bluegill Lepomis macrochirus in Pelican Lake, Nebraska, 2004 and 2005 by prey category of zooplankton. Confidence intervals above the random feeding (dashed) line indicate positive selection, values below the line indicate negative selection, and values overlapping the line indicate neutral selection.

both lakes in 2005, prey was seemingly available for consumption. The observation of all larval yellow perch collected with empty stomachs in 2005 in Pelican Lake was notable. It is unknown if these fish were in a stressed condition upon their collection and were unable to adequately seek prey items or if they may not have started first-feeding (yolk-sacs were not present). Sparse prey resources for first-feeding fish larvae have been commonly implicated as a cause of high mortality (May 1974; Cushing 1975, 1990; Hart and Werner 1987) and Toetz (1966) experimentally reported mass starvation of larval bluegill between 5 and 6 mm TL, which corresponded to initiation of exogenous feeding. Isermann and Willis (2008) suggested that factors prior to, during, or immediately following hatching may play a critical role in the recruitment of yellow perch under the constraint of a narrow spawning window. Our results are contrary to multiple experimental studies in which survival of larval fishes is most influenced by zooplankton biomass (Hart and Werner 1987; Welker et al. 1994). Conversely, Houde (1994) predicted that starvation of larval fishes in freshwater environments was less likely toJournal of Fish and Wildlife Management | www.fwspubs.org

occur than in marine environments because freshwater fish larvae are generally larger, thus conferring greater energy reserves and resistance to starvation (Miller et al. 1988). Previous research examining the relation of zooplankton biomass and abundance to larval growth and survival of panfish has had mixed results (Pope and Willis 1998; Garvey et al. 2002; Bunnell et al. 2003). In 2005, yellow perch hatched over a very narrow time frame in Pelican Lake (4 d) and larvae were only collected in low numbers on one day, likely indicating a relatively weak initial year-class. Concurrently, Pelican Lake experienced a drop in water temperature from nearly 17uC to nearly 6uC over a period of 2 wk in late April (Jolley 2009). This corresponded to the time period when yellow perch eggs would have been incubating and hatching. It is unclear whether the eggs or newly hatched larvae were negatively affected by this cold front. Jansen et al. (2009) simulated the effect of a cold front on yellow perch eggs and found no decrease in egg survival, suggesting that the newly hatched larval stage may be more susceptible than eggs to these extreme weather events. Sandhill lakes are shallow and windswept; they, thus, are susceptible to erratic temperature changes and can warm and cool quickly. Jolley (2009) examined the relation of recruitment (i.e., year-class strength) of bluegill and yellow perch to climatic variables in several Sandhill lakes and found asynchronous recruitment among the study lakes. In addition, limited support for the concept of climatic influence on bluegill and yellow perch was found over the years examined. No mismatches in predator and prey abundance were detected over the years examined in our study; thus, we cannot determine if severe mismatches in the appearance of fish larvae and their zooplankton prey would lead to depressed survival of age-0 bluegill and yellow perch in Nebraska Sandhill lakes. Density of larval fish and zooplankton was variable among years but exact matches occurred frequently for bluegill and yellow perch. Although zooplankton density can vary spatially (Young et al. 2009) leading to potential spatial mismatches (Chick and Van Den Avyle 1999), differences in zooplankton densities between inshore and offshore strata were not apparent (Jolley 2009). Intrastation variability of zooplankton density was generally less than interstation variability, and density differences were rarely detected among regions of the lake (unpublished data). Young et al. (2009) reported that small-scale patchiness (i.e., , 1 m) accounted for the majority of the variation in zooplankton abundance, which could contribute to spatial mismatches considering the small search volume reported for many larval fish species (, 2.5 L; Blaxter 1986; Pepin 2004). They suggested that measures of average prey density made over larger scales may be independent of the feeding of individuals. Determination of the scale of zooplankton patchiness was beyond the scope of our study but may be a topic worth revisiting in future studies. Although copepodites and rotifers have also been reported (Whiteside et al. 1985; Schael et al. 1991; Wahl et al. 1993; Fisher and Willis 1997) as a preferred prey item of newly hatched yellow perch, we found rareNovember 2010 | Volume 1 | Issue 2 | 79


J.C. Jolley et al.

Table 4. Percent occurrence and percent by number for zooplankton prey items found in larval bluegill Lepomis macrochirus stomachs in Pelican Lake, Nebraska, in 20042005.2004 Taxon % Occurrence Bosminidae Chydoridae Daphnidae Sididae Copepoda Nauplii Ostracoda Rotifera % by number Bosminidae Chydoridae Daphnidae Sididae Copepoda Nauplii Ostracoda Rotifera 0 0 0 0 10.0 90.0 0 0 0 8.8 0 0 5.9 20.6 0 64.7 46.7 3.5 38.1 0 5.4 1.6 0.4 4.3 15.3 0.2 21.9 0.2 61.9 0.4 0 0 42.6 0.5 22.7 0.2 33.6 0.5 0 0 46.2 9.3 12.9 0.7 30.4 0.3 0 0.1 47.9 5.3 18.4 0.2 28.0 0.2 0 0 34.6 19.2 23.4 3.2 19.0 0.4 0.2 0.1 21.7 1.6 10.1 0 34.1 27.9 4.7 0 28.3 22.7 34.0 0 11.5 0.4 0.7 2.4 32.9 23.9 31.2 0.4 9.0 2.1 0.4 0 51.2 5.7 39.2 1.0 2.9 0 0 0 44.5 14.6 33.0 0.2 6.4 0.2 0 1.1 32.8 12.9 28.9 3.6 21.8 0 0 0 0 0 0 0 33.3 66.7 0 0 0 22.2 0 0 11.1 44.4 0 66.7 63.3 10.0 46.7 0 30.0 10.0 3.3 16.7 50.0 3.3 83.3 3.3 96.7 3.3 0 0 80.0 6.7 60.0 3.3 76.7 3.3 0 0 83.3 40.0 83.3 13.3 83.3 10.0 0 3.3 96.7 53.3 96.7 10.0 86.7 6.7 0 0 83.3 60.0 80.0 43.3 90.0 6.7 6.7 3.3 46.7 6.7 26.7 0 60.0 40.0 10.0 0 60.0 56.7 63.3 0.0 33.3 6.7 6.7 20.0 60.0 63.3 76.7 3.3 36.7 16.7 3.3 0 76.7 33.3 60.0 6.7 20.0 0 0 0 63.3 40.0 96.7 3.3 23.3 3.3 0 10.0 70.0 50.0 83.3 43.3 63.3 0 0 0 26 7 June July 17 July 27 July 6 16 27 August August August 5 Sept 30 11 June July 2005 21 July 1 11 22 August August August

consumption of these taxa despite their availability. Increased predation on daphnids by larger (i.e., 30 mm) yellow perch has been previously reported (Whiteside et al. 1985; Prout et al. 1990) and positive selection for these diet items has been shown (Mills et al. 1984; Schael et al. 1991). Our observations corroborate such findings

although daphnids were first consumed when yellow perch were between 15 and 17 mm TL and Wahl et al. (1993) reported consumption of daphnids by larval yellow perch as small as 9 mm TL. Yellow perch hatched 2 mo before bluegill and had the potential to either prey upon or compete with

Table 5. Mean peak larval abundance (n/100 m3), parameters used in matchmismatch examination, and result for yellow perch Perca flavescens and copepods in Cameron and Pelican Lake, Nebraska, and for bluegill Lepomis macrochirus and combined copepoditeBosmina in Pelican Lake, Nebraska, 20042008. Parameters are t0 (timing between peaks of larval production and food supply in days), Dt0 (annual differences in [t0] from its mean value), d (one-half width of the production period for larvae), and s (one-half width of the production period for zooplankton).Predatorprey combination Yellow perchcopepods Yellow perchcopepods Peak larval abundance 804 2 23 6 46 89 238 66 57 169 315 377 1,269 Parameter

Lake Cameron Pelican

Year 2004 2005 2004 2005 2006 2007 2008

t019 0 0 0 21 0 10 20 9 21 10 11 11

Dt0 11.9 7.1 7.1 7.1 13.9 7.1 2.9 8.0 3.0 9.0 2.0 1.0 1.0

d 9.5 0 0 0 4.5 4 15.5 20.5 10.5 21 31.5 21 19

s 7 4.5 19 10 16 9 26 36 36 26.5 21 26 25

Result Match Exact match Exact match Exact match Weak match Exact match Match Match Match Match Match Match Match

BluegillcopepoditeBosmina

2004a 2004b 2005 2006 2007 2008

a b

First peak in larval bluegill abundance in 2004. Second peak in larval bluegill abundance in 2004.




J.C. Jolley et al.

Figure 4. Larval yellow perch Perca flavescens density (solid line) and copepod density (broken line) in Cameron Lake, Nebraska, in 2004 and 2005 (n = number).

bluegill larvae. We did not observe any larval bluegill consumed by juvenile yellow perch in this study (up to 74 mm TL). Although Graeb et al. (2006) demonstrated experimentally that yellow perch begin a shift to fish prey at 80 mm TL, most field studies do not report consistent piscivory by yellow perch until they attain 130150 mm TL (Clady 1974; Keast 1985; Fullhart et al. 2002), a larger size than examined in our study. Total zooplankton densities in 2004 were low on the first day that bluegill larvae were collected, which may explain the lack of positive prey selection for any particular group of zooplankton by newly hatched bluegill larvae in 2004. Partridge and DeVries (1999) noted a high proportion of rotifers in larval bluegill diets, which may lead to suboptimal bluegill growth. Rotifers were rarely consumed in our study, although they were remarkably abundant in Pelican Lake in 2004. The availability of energetically profitable prey (i.e., copepods and cladocerans) likely precluded bluegill larvae from consuming rotifers. Although densities of yellow perch larvae varied considerably (. 300%) among years, it appears that recruitment of yellow perch was relatively consistent in Pelican and Cameron lakes over the years examined and the observed densities of larval yellow perch were generally higher than reported values of density in six South Dakota glacial lakes monitored for over 8 y (Jansen 2008). Many other Sandhill lakes concurrently studied exhibited relatively consistent recruitment (Jolley 2009). The life history and reproductive and spawning behavior of many species have evolved in ecosystems where environmental variability is unpredictable (Winemiller and Rose 1993). A combination of trophodynamic and physical factors may interact in complex ways, over multiple temporal and spatial scales, to affect larval fish survival, growth, and recruitment (Fitzgerald et al. 2001; Houde 2008). Availability of appropriate prey in timeJournal of Fish and Wildlife Management | www.fwspubs.org

Figure 5. Larval yellow perch Perca flavescens density (solid line) and copepod density (broken line) in Pelican Lake, Nebraska, 20042008 (n = number).

and space is no doubt an important factor to the survival and recruitment of age-0 fish and is likely a component of an integrated process acting throughout early life-stages to explain components of recruitment variability (Houde 2008). Many studies that provide support for the matchmismatch hypothesis involved marine species and systems (Cushing 1990; Fortier and Gagne 1990; Gotceitas et al. 1996; Johnson 2000); it is lesser studied in freshwater systems, with some support for landlocked striped bass Morone saxatilis (Chick and Van Den Avyle 1999), threadfin shad Dorosoma peteNovember 2010 | Volume 1 | Issue 2 | 81


J.C. Jolley et al.

Table 6. Bivariate correlations between predatorprey overlap (t0) and fish abundance indices for yellow perch Perca flavescens and bluegill Lepomis macrochirus. Abundance indices are peak larval abundance (mean n/m3), autumn and spring juvenile catch per unit effort (CPUE; mean n/cloverleaf trap-night), and age-1 and age-2 CPUE (mean n/trap-net night). Number of data pairs (N), correlation coefficient (r), and P-value are given for each bivariate correlation.Independent variable Peak larval abundance Autumn juvenile CPUE Spring juvenile CPUE Age 1 CPUE Age 2 CPUE Yellow perch Bluegill

N7 7 7 7

r20.19 0.13 0.32 20.51

P0.68 0.77 0.48 0.38

N6 6 6 5

r20.34 0.08 20.36 20.43

P0.51 0.88 0.49 0.47

Figure 6. Larval bluegill Lepomis macrochirus density (solid line) and combined copepodite and Bosmina density (broken line) in Pelican Lake, Nebraska, 20042008 (n = number).

nense (Betsill and Van Den Avyle 1997) and yellow perch (Fitzgerald et al. 2001). Houde (1994) found that marine larvae (compared to freshwater) may be more susceptible to starvation mortality due to high metabolic demand and small size at hatch. These traits may be related to the applicability of the matchmismatch hypothesis for marine fishes. Houde (1994) found that freshwater fish larvae may be more susceptible to episodic mortalities that affect recruitment, but that the juvenile stage may be equally important in regulating and controlling recruitment. In light of these complexities, and given that zooplankton were abundant and well-timed to larval fish abundance over the initial years of this study, the matchmismatch hypothesis may notJournal of Fish and Wildlife Management | www.fwspubs.org

be able to account for observed recruitment variability in the populations that we studied. Given the importance of copepods, copepodites, daphnids, and bosminids as prey for larval yellow perch and bluegill, more in-depth examination of these relationships is warranted. Abundance indices of these zooplankton taxa may be used as surrogates for prey availability. In addition, the dynamic nature of zooplankton populations may suggest important consequences via timing (e.g., matchmismatch regulation; Cushing 1975, 1990), which could be further explored. Our sampling interval of 10 d may lack the required resolution to fully understand the relation between zooplankton and larval bluegill and yellow perch recruitment. If catastrophic mortality events happen in a short time (Hjort 1914; May 1974) then more frequent sampling may be required to pinpoint the timing and explanation for it. Multiple life-stage abundance indices of yellow perch and bluegill were not correlated to zooplankton abundance indices, although those results were based on a low number of observations (Jolley 2009). Continued stage-specific investigations of the relationship of zooplankton to larval fish growth and recruitment may produce important insights into the dynamics of bluegill and yellow perch in temperate lakes. In addition, examination of later life-stages (e.g., juveniles) of bluegill and yellow perch likely is also necessary. Studies that incorporate multiple life-stages (Ludsin and DeVries 1997; Jolley 2009) and include ecosystem processes are especially valuable (Cury et al. 2008).

Supplemental MaterialPlease note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author. Table S1. Zooplankton availability. Found at DOI: 10.3996/062010-JFWM-018.S1 (292 KB XLS). Table S2. Larval fish diets. Found at DOI: 10.3996/062010-JFWM-018.S2 (1757 KB XLS).November 2010 | Volume 1 | Issue 2 | 82


J.C. Jolley et al.

AcknowledgmentsWe thank A. Husman, E. Lorenzen, J. Rydell, D. Coulter, W. Bauer, and C. Longhenry for field and laboratory assistance. M. Lindvall and Valentine National Wildlife Refuge provided access to Pelican Lake and R. Lackaff provided access to Cameron Lake. We thank D. Graham, D. Hartmann, D. Kruger, and the Valentine State Fish Hatchery for assistance. We also thank the three anonymous reviewers and the Subject Editor for valuable improvement to this manuscript. Funding for this project was provided by the Nebraska Game and Parks Commission through Federal Aid in Sport Fish Restoration Project F-118-R.

ReferencesAnderson MR, Fisher SJ, Willis DW. 1998. Relationship between larval and juvenile perch abundance in eastern South Dakota glacial lakes. North American Journal of Fisheries Management 18:989991. Auer NA. 1982. Identification of larval fishes of the Great Lakes basin with emphasis on the Lake Michigan drainage. Ann Arbor, Michigan: Great Lakes Fishery Commission. Beard TD. 1982. Population dynamics of young-of-theyear bluegill. Madison, Wisconsin: Wisconsin Department of Natural Resources. Technical Bulletin 127. Betsill RK, Van Den Avyle MJ. 1997. Effect of temperature and zooplankton abundance on growth and survival of larval threadfin shad. Transactions of the American Fisheries Society 126:9991011. Blaxter JHS. 1986. Development of sense organs and behavior of teleost larvae with special reference to feeding and predator avoidance. Transactions of the American Fisheries Society 115:98114. Bleed A, Flowerday C. 1989. Introduction. Pages 115 in Bleed A, Flowerday C, editors. An atlas of the Sand Hills. Lincoln: Univer

Documents

Journal of Fish and Wildife Management Vol 1 #2