Effects of Patch-Burn Management on Dickcissel Nest Success in a Tallgrass Prairie

Management and Conservation Article Effects of Patch-Burn Management on Dickcissel Nest Success in a Tallgrass Prairie ROY T. CHURCHWELL, 1 Department of Zoology, Oklahoma State University, 430 Life Science West, Stillwater, OK 74078, USA CRAIG A. DAVIS, 2 Department of Natural Resource Ecology and Management, 008C Agricultural Hall, Oklahoma State University, Stillwater, OK 74078, USA SAMUEL D. FUHLENDORF, Department of Natural Resource Ecology and Management, 008C Agricultural Hall, Oklahoma State University, Stillwater, OK 74078, USA DAVID M. ENGLE, Department of Natural Resource Ecology and Management, 339 Science II, Iowa State University, Ames, IA 50011, USA ABSTRACT Grassland birds have declined more than any other North American habitat-associated bird community. Because most species of grassland birds evolved within heterogeneous landscapes created by the interaction of fire and grazing, traditional rangeland management that promotes homogeneity, including annual dormant-season burning combined with early-intensive grazing, might be partly responsible for these declines, especially in some regions of the Great Plains, USA. Recently, an alternative grassland management practice known as patchburning has been promoted as a means of restoring heterogeneity to grasslands by mimicking the grazing fire interaction that once occurred on the prairie before European settlement. From 2003 to 2004, we examined effects of patch-burning and traditional management (annual burning followed by early-intensive grazing) on the reproductive success of dickcissels (Spiza americana) in tallgrass prairie in Oklahoma. We monitored 296 dickcissel nests and found that dickcissel nesting phenology differed between traditional and patch-burned pastures. Specifically, dickcissels tended to initiate their nests later in the traditional pasture. Mean number of eggs laid and fledglings produced were similar between the treatments, but nest densities were higher in traditional pastures. Predation was the predominant cause of nest failure and was higher in traditional pastures than in patch-burned pastures. Brown-headed cowbird (Molothrus ater) parasitism was higher in traditional pastures than in patch-burned pastures. Overall, dickcissel nest success was higher in patch-burned pastures than in traditional pastures. The positive response of dickcissel nest success to patch-burn management provides further evidence that this practice can be a useful tool for grassland bird conservation. By creating a mosaic of different stature vegetation, patch-burn management enhances productivity of grassland bird species by providing a refuge area in the unburned patches that affords dickcissels and other nesting grassland birds some protection from the direct (e.g., trampling) and indirect (e.g., cowbird parasitism and predation) effects of grazing, which are not available under traditional management. Patch-burn management should be encouraged as a conservation strategy for grassland birds throughout the Great Plains. (JOURNAL OF WILDLIFE MANAGEMENT 72(7):1596 1604; 2008) DOI: 10.2193/2007-365 KEY WORDS dickcissel, grazing, logistic exposure, nest success, Oklahoma, patch-burn management, prescribed fire, Spiza americana, tallgrass prairie. Grassland birds have shown a significant sustained decline, whereas populations of other North American habitatassociated bird communities have not similarly diminished (Knopf 1994, 1996; Herkert and Knopf 1998; Peterjohn and Sauer 1999; Faaborg 2002). Meanwhile, estimates of the loss of native tallgrass prairie are.80% in most American states and Canadian provinces (Samson and Knopf 1994, Vickery et al. 1999). The decline in grassland bird populations is attributed predominantly to a loss of native grasslands (Knopf 1994, Herkert and Knopf 1998, Peterjohn and Sauer 1999), but management of grasslands has also contributed to the decline. In particular, traditional rangeland management that is based on a paradigm of managing for uniformity and managing against heterogeneity has contributed to declines in grassland bird diversity and abundance (Fuhlendorf et al. 2006). The objective of traditional rangeland management techniques, such as cross-fencing of pastures, strategic placement of water, annual burning of pastures, and use of herbicides to eliminate forbs, is to increase livestock production, which produces a homogeneous grass-dominated habitat for birds (Knopf 1994, Fuhlendorf and Engle 1 Present address: Department of Biology and Wildlife, University of Alaska, Fairbanks, AK 99775, USA 2 E-mail: craig.a.davis@okstate.edu 2001). Unfortunately, these traditional rangeland management techniques counter the natural grazing and fire interaction that shaped pre-european settlement prairie and created a structurally heterogeneous habitat for birds (Knopf 1994, Fuhlendorf and Engle 2001, Fuhlendorf and Engle 2004). Moreover, the evolutionary importance of heterogeneity in prairie is evident from the variability in grassland bird habitat requirements, with some species having affinities for grasslands with specific structural characteristics (Knopf 1996). For example, horned larks (Eremophila alpestris) prefer to nest in areas composed mostly of bare ground (Beason 1995), whereas at the other end of the spectrum, Henslow s sparrows (Ammodramus henslowii) prefer nesting in areas with high amounts of litter (Herkert 1994a, Zimmerman 1997). Fuhlendorf and Engle (2001) suggested that variability in grassland bird habitat selection indicates that heterogeneous grasslands are necessary for maintaining diverse grassland bird communities. Fuhlendorf and Engle (2001) proposed an alternative, novel rangeland management approach known as patchburning, patterned after the grazing fire interaction thought to have occurred on the prairies before European settlement. Patch-burning uses disturbance created by prescribed fire and focal grazing to create a shifting mosaic in herbaceous 1596 The Journal of Wildlife Management 72(7)

plant species composition and structure within the grassland community (Fuhlendorf and Engle 2004). Specifically, patch-burning creates a heterogeneous pasture of different seral stages ranging from a patch that is recently burned and heavily grazed to a patch that has not burned for 2 years and is ungrazed to lightly grazed (Fuhlendorf and Engle 2001, 2004). Cattle behavior (i.e., cattle prefer to graze burned areas over unburned areas) is used instead of crossfencing to influence grazing intensity on the burned and unburned patches. As a result of the increased vegetation heterogeneity created by patch-burn management, grassland bird abundance and diversity are much greater in patchburn managed grasslands than traditional managed grasslands that are annually burned (Fuhlendorf et al. 2006). Although patch-burning is meant to mimic the historic patterns created by grazing fire interactions, patch burns are at a smaller scale than likely occurred on the landscape before European settlement. Although patch-burn management can maintain and enhance grassland bird biodiversity (Fuhlendorf et al. 2006), little is known about the reproductive rates of birds nesting in patch-burn managed pastures. Specifically, does patch-burn management positively affect grassland bird populations by providing source habitats (i.e., reproductive rates in patch-burn pastures exceed levels necessary for the population to increase; Pulliam 1988) or negatively affect grassland bird populations by providing sink habitats (i.e., reproductive rates are below levels that do not compensate for mortality losses resulting in a population decline; Pulliam 1988)? Therefore, an understanding of how patch-burn management affects local grassland bird populations is necessary if this management practice is to be recommended as a strategy for grassland bird conservation. Our goal was to elucidate the influence of patch-burn management on grassland bird reproductive success. Although we monitored the reproductive success of several grassland bird species, we focused on dickcissels because they were the only species that occurred in all management treatments, and samples sizes were adequate for each treatment. Because dickcissels are generalist species, they provide insight on the response of specialist species such as Henslow s sparrows, which may only occur in one of the treatments (i.e.,.1 yr postburn), to patch-burn management. That is, we argue that if nest success of a generalist species is not detrimentally affected by patch-burn management within the preferred habitat of the specialist species, then nest success of a specialist species is likely to be enhanced by patch-burn management. Consequently, our objectives were to 1) describe nesting phenology and demographics of dickcissels relative to patch-burn and traditional management and 2) compare nest success and nest survival of dickcissels in patch-burn managed and traditional managed pastures. STUDY AREA We conducted our study in 2003 and 2004 at The Nature Conservancy s 14,000-ha Tallgrass Prairie Preserve (hereafter, the Preserve) in Osage County, Oklahoma, USA (36850 0 N, 96825 0 W) located at the southern extent of the Flint Hills Region. Average total precipitation for the area was 877 mm with about 70% occurring between April and September (Coppedge et al. 1998). Dominant grasses of the Preserve are big bluestem (Andropogon gerardii), little bluestem (Schizachyrium scoparium), indiangrass (Sorghastrum nutans), and switchgrass (Panicum virgatum), and dominant forbs include ironweed (Veronia spp.), milkweed (Asclepias spp.), and ashy sunflower (Helianthus mollis). We chose 4 pastures that ranged from 400 ha to 900 ha, which were located on similar soils and had similar management histories (i.e., all pastures had been annually burned and grazed before 2001). We divided each pasture into 6 approximately equal-sized patches (average patch size, 100 ha). We selected a 100-ha patch size because it is larger than the area requirements of most area-sensitive species, thus reducing the influence of area on our results (Herkert 1994b, Johnson 2001). The exterior boundary of each pasture was fenced, but the interior patches were unfenced. We used 2 pastures for evaluation of patch-burn management and 2 pastures for evaluation of traditional management. Patch-burn management was initiated at the Preserve in 2001 and continued through 2004. In the patch-burn management pasture, one patch was burned in spring (Mar) and another patch was burned in summer or fall (Aug Dec) each year. In 2001, treatments were randomly assigned to patches and subsequent burn schedules were based on this initial assignment. Consequently, each patch was burned once (either spring or fall) every 3 years resulting at any point in time one-third of a pasture having been burned within the current year, one-third having been not burned for 1 year, and one-third having not been burned for 2 years. Cattle had free access to all patches within all pastures. Because cattle preferentially graze recently burned patches and avoid older patches (i.e.,.1 yr since burned; Fuhlendorf and Engle 2004), a mosaic of different stature vegetation is created within the patch-burn management pasture (Table 1; Fuhlendorf et al. 2006). For example, grass cover can experience a nearly 2-fold increase from the current-year burn patch to the 2-year postburn patch, and litter cover in the 2-year postburn patch can be as much as 60 times greater than in the current-year burn patch (Table 1). Traditional management was designed to mimic the prevalent burning practice in northeastern Oklahoma, so, the traditional management pastures were burned every spring (i.e., all patches within the traditional pasture were burned in one fire each yr). Cattle were stocked in each pasture from mid-april through mid-july at a rate of 1.2 ha/ 270-kg steer, which is the standard stocking rate for the region (Bourlier et al. 1979). METHODS Data Collection We randomly located a 16-ha nest plot within each springburned patch of the 2 patch-burned pastures and in one of the patches of the 2 traditional pastures, but we only selected Churchwell et al. Fire Grazing Interactions and Dickcissels 1597

Table 1. Differences in vegetation characteristics between traditional and patch-burned pastures at the Tallgrass Prairie Preserve, Oklahoma, USA, 2003 2004. We collected these data at 26 random locations within each nest search plot in association with vegetation measurements taken at actual nests (Churchwell 2005). Patch-burn Traditional Current-yr 1-yr postburn 2-yr postburn Variable x SE x SE x SE x SE Grasses (%) 51.39 2.49 33.32 2.96 60.05 2.71 74.54 2.58 Legumes (%) 3.56 0.77 3.11 1.00 2.59 0.62 3.50 1.11 Forbs (%) 43.32 2.59 41.20 3.04 35.34 2.89 23.03 2.13 Litter (%) 1.48 0.69 0.60 0.48 35.43 2.51 61.59 1.28 Bare ground (%) 2.88 0.73 16.00 3.02 0.87 0.40 0.05 0.05 Plant ht (m) 0.32 0.02 0.25 0.01 0.36 0.02 0.44 0.01 nest plots that were.50 m from the edge of each patch to reduce edge effects. We selected spring-burn patches because previous research showed little difference in bird response between spring and fall burns (Harrell 2004), and we chose a 16-ha nest plot size because it allowed us to locate a maximum number of nests without decreasing searching efficiency (R. T. Churchwell, Oklahoma State University, personal observation). All nest plots were 1 km apart. We began nest searching in mid-may and continued through July in 2003 and 2004. We conducted nest searches every other day within each nest plot. To minimize detection of nests by predators, we wore rubber boots and spent a minimal amount of time checking each nest to reduce the amount of human scent left and we approached each nest from a different path to minimize disturbance to vegetation near the nest (Martin and Guepel 1993, Winter et al. 2003). We used behavioral cues (i.e., adults approaching the nest often with nest-building material or food) and flushing of birds to locate nests (Martin and Geupel 1993, Ralph et al. 1993). We recorded Universal Transverse Mercator coordinates for each nest using a Geographic Positioning System unit and used these coordinates to locate nests on subsequent visits. After we located a nest, we monitored it every 1 5 days to assess nest outcome. We visited nests more frequently near the expected time of hatching and fledging to more accurately age nests and correctly determine nest outcome. We considered a nest successful if 1 conspecific young fledged. In almost all cases, we were able to confirm a successful nest by observing parents feeding young or hearing begging calls from nearby fledglings. For most failed nests, we were able to determine the possible cause of failure as nest predation, trampling by cattle, weather-related failure, or abandonment. We recorded predation when nest contents were removed from the nest before the expected fledging date or when there was no sign of the young or parents in the nest area on the expected fledging date. Trampling by cattle was evident when the nest contents were destroyed, and we found cattle tracks in the nest (Churchwell et al. 2005). We recorded weatherrelated failures when the nest contents were on the ground after a storm, and there were no signs of disturbance from passing cattle or ground nests were filled with soil and debris from flowing water. We determined nest loss from abandonment when eggs or young were still present, but the female was absent from the nest after several nest checks. We also recorded the extent of brown-headed cowbird (Molothrus ater) parasitism. Statistical Analyses We used a 2-way analysis of variance (ANOVA) to compare clutch sizes, number of fledglings produced, and nest densities among treatments (i.e., traditional pasture, current-yr burn patch in patch-burned pasture, 1-yr postburn patch in patch-burned pasture, and 2-yr postburn patch in patch-burned pasture) and between years (SYSTAT version 9, SPSS, Chicago, IL). We considered the patch within a pasture as the experimental treatment unit because each patch has a unique treatment imposed by the fire grazing interaction. Hence, each patch (i.e., current-yr burn patch, 1-yr postburn patch, and 2-yr postburn patch) is an independent experimental unit. We used Fisher s least significant difference for mean separation following a significant ANOVA for treatment effects. We calculated 2 estimates of nest survival using the Mayfield technique (Mayfield 1961, 1975) and the logisticexposure method (Shaffer 2004). We calculated standard errors for Mayfield daily nest survival following Johnson (1979). Although we believe the logistic-exposure method is more appropriate for investigating factors that may influence nest success, results from logistic-exposure method are not comparable with studies that report apparent nest success or Mayfield nest success. Thus, for comparative purposes, we also report apparent nest success and Mayfield nest success. For the logistic-exposure method, we used a generallinear-model approach to estimate daily nest success and evaluate factors that may affect nest success (Shaffer 2004). The logistic exposure method is similar to logistic regression except that it allows the time between visits (t) to vary in the logit function g(h) ¼ log e [h 1/t /(1 h 1/t )], where h is the probability the nest survives between nest checks. The logistic exposure method also allows for the modeling of time-dependent explanatory variables by assuming that the variable is constant within a nest-check interval, but the variable can be varied between nest-check intervals (Peak et 1598 The Journal of Wildlife Management 72(7)

Table 2. A priori models explaining effects of management treatments, nest age, and year on nesting success of dickcissels at the Tallgrass Prairie Preserve, Oklahoma, USA, 2003 2004. The number of parameters (K) in each model included intercept and each explanatory variable. The model with the lowest Akaike s Information Criterion difference (DAIC) and the largest Akaike weight (w i ) was most supported. Sample size is 2,022 nestobservation intervals. Model K D AIC w i Nest age effects 2 0.00 0.420 Treatment and nest age effects 5 1.41 0.207 Nest age and yr effects 3 1.91 0.162 Treatment, yr, and nest age effects 6 3.41 0.076 Treatment, yr, nest age, and treatment 3 yr effects 9 4.18 0.052 Treatment, nest age, and treatment 3 nest age effects 8 5.29 0.030 Constant survival 1 6.27 0.018 Treatment, yr, nest age, and treatment 3 nest age effects 9 7.31 0.011 Treatment, yr, nest age, treatment 3 nest age, and treatment 3 yr effects 8 7.47 0.010 Yr effects 2 8.25 0.007 Treatment effects 4 9.03 0.004 Treatment and yr effects 5 10.83 0.002 Treatment, yr, and treatment 3 yr effects 7 12.76 0.001 al. 2004, Shaffer 2004). We used PROC GENMOD (SAS Institute Inc., Cary, NC) to fit models. We used an information-theoretic approach (Burnham and Anderson 2002) to evaluate a priori models (Table 2) concerning effects of treatment, year, nest age, and secondorder interactions (i.e., treatment 3 yr and treatment 3 nest age) on nest success. We coded treatment categorical variables as dummy variables in the model (Agresti 1996), and we modeled all variables as fixed variables. We tested suitability of using logistic regression with these data using goodness-of-fit tests (Hosmer and Lemeshow 2000) on the global model for each group of candidate models. We considered each candidate model a well-fitting model (i.e., we fail to reject the null hypothesis that there is no difference between observed and model-predicted values) at P. 0.05 (Hosmer and Lemeshow 2000). To calculate the maximum likelihood probability of daily nest success, we used model-averaged coefficients [R (coeff. 3 Akaike wt)]. We applied Akaike s Information Criterion (AIC) to rank models by comparing the DAIC value and Akaike weights (Burnham and Anderson 2002). By averaging all models with DAIC, 2, we were able to account for the fact that all models with DAIC, 2 are Figure 1. Nesting phenology of dickcissels in traditional pastures and patches of patch-burned pastures at the Tallgrass Prairie Preserve, Oklahoma, USA, 2003 2004. Sample sizes were n ¼ 67 for traditional pastures, n ¼ 11 for current-year burn patches, n ¼ 30 for 1-year postburn patches, and n ¼ 59 for 2-year postburn patches. likely candidate models (Burnham and Anderson 2002). We then calculated maximum likelihood probability of daily nest success by inserting averaged coefficients into the selected model and calculating the logistic function fs(x) ¼ [e boþb1x ]/ [1 þ e boþb1x ]; Shaffer 2004g, where s(x) is daily nest survival probability and b is averaged model coefficients. RESULTS Dickcissel nesting phenology differed between traditional and patch-burned pastures. On average, dickcissel nest initiation dates in traditional pastures and current-year burn patches in patch-burned pastures were 5 7 days later than in 1-year postburn and 2-year postburn patches in patchburned pastures. Dickcissels initiated most nests (.80%)by 14 June in patch-burned pastures and by 28 June in traditional pastures (Fig. 1). Mean number of eggs laid (F 3,254 ¼ 1.67, P ¼ 0.175) and fledglings produced (F 3,272 ¼ 1.59, P ¼ 0.192) were similar among traditional and patchburned pastures (Table 3). Nest densities differed among traditional and patch-burned pastures (F 3,8 ¼ 5.11, P ¼ 0.029); the highest densities of nests occurred in the traditional pasture and the 2-year postburn patch in the patch-burned pasture and the lowest densities occurred in the current-year burn patch in the patch-burned pasture (Table 3). None of the year 3 treatment interactions were significant (F, 2.28, P. 0.079). We monitored 296 dickcissel nests (141 nests in 2003, 155 nests in 2004). Nests failed because of predation, cattle Table 3. Number of eggs and fledglings produced per nest and nest density of dickcissels in traditional and patch-burned (current-yr burn patch, 1-yr postburn patch, and 2-yr postburn patch) pastures at the Tallgrass Prairie Preserve, Oklahoma, USA, 2003 2004. Patch-burn Traditional Current-yr 1-yr postburn 2-yr postburn Variable x SE x SE x SE x SE No. eggs/nest 4.24 0.07 4.24 0.14 4.09 0.09 4.01 0.07 No. fledglings/nest 1.01 0.17 1.10 0.40 1.40 0.22 1.51 0.20 No. nests/ha 2.00 0.33 0.41 0.16 1.03 0.26 1.64 0.32 Churchwell et al. Fire Grazing Interactions and Dickcissels 1599

Figure 2. Causes of nest failure for all failed dickcissel nests in traditional pastures and patches of patch-burned pastures at the Tallgrass Prairie Preserve, Oklahoma, USA, 2003 2004. Sample sizes were n ¼ 84 for traditional pastures, n ¼ 17 for current-year burn patches, n ¼ 37 for 1-year postburn patches, and n ¼ 57 for 2-year postburn patches. trampling, abandonment, and weather. Predation was the predominant cause of nest failure in traditional and patchburned pastures (Fig. 2). Predation was higher in traditional pastures and current-year burn patches in the patch-burned pasture (89 94%) than in the 1-year postburn and 2-year postburn patches in the patch-burned pasture (76 78%). We did not observe nest predation events, but we suspect reptiles, which have been reported as a major predator of dickcissel nests in Missouri, USA (Suedkamp Wells 2005) and at the Preserve (Shochat et al. 2005), were the principal predator because we found many of the depredated nests intact but empty with the nest slightly tipped to one side. Additionally, Shochat et al. (2005) suggested that box turtles (Terrapene spp.), which were abundant in burned pastures, could be a major nest predator at the Preserve. Overall, brood parasitism was higher in traditional pastures (19%) than patch-burned pastures (5%). Within patch-burned pastures, brood parasitism ranged from 69% in the current-year burn patch to 0% in the 1-year postburn and 2-year postburn patches. Of the nests that were parasitized, most were later depredated, but one parasitized nest did produce young (2 dickcissels and 1 cowbird). Apparent nest success for traditional pastures was 26%, whereas apparent nest success for patch-burned pastures was 41%. Within patch-burned pastures, apparent nest success ranged from 29% for the current-year burn patch to 42% for the 1-year postburn and 2-year postburn patches. Overall, Mayfield daily nest survival (0.938 þ 0.006 [SE]) and nest success (26.3%) for patch-burned pastures were higher than in traditional pastures (Table 4). Within patchburned pastures, the lowest Mayfield daily nest survival and nest success occurred in current-year burn patches and the highest Mayfield daily nest survival and nest success occurred in 2-year postburn patches (Table 4). We used 166 nests of known age for logistic exposure analysis. Results from Hosmer and Lemeshow (2000) goodness-of-fit test indicated the global model fit the data (v 2 ¼ 12.0, df ¼ 8, P ¼ 0.15). We identified 3 candidate models with DAIC scores,2: nest age effects model, treatment and nest age effects model, and nest age and year effects model (Table 2). The nest age effects model ranked highest, with 2.03 times more weight of support than the next-best ranked model (treatment and nest age effects model). Daily nest survival of dickcissels declined at a rate of 0.05 over the entire nesting stage. Further examination of the effects of treatment showed that daily nest survival was influenced by the frequency of burning, with patches that had not burned during the current year (1-yr postburn and 2-yr postburn patches in the patch-burned pastures) having higher daily nest survival than those that had burned during the current year (traditional pastures and current-yr burn patches in the patch-burned pastures; Fig. 3). Moreover, the traditional managed pasture and current-year burn patch had coefficient estimates near 0.20, indicating a negative effect on nest survival, whereas 1-year postburn and 2-year postburn patches had coefficient estimates ranging from 0.01 to 0.02, indicating a positive effect on nest survival. Daily nest survival was 0.07 times higher in 2003 compared with 2004. DISCUSSION Our results suggest that dickcissels will be more successful nesting in patch-burned pastures than traditional pastures. Overall, dickcissels nesting in patch-burned pastures had higher nest success and lower cowbird parasitism than dickcissels in traditional pastures. The higher nest success in patch-burned pastures is likely related to the increased spatial heterogeneity of vegetation structure that occurs in patch-burned pastures (Fuhlendorf and Engle 2001, 2004; Fuhlendorf et al. 2006). Dickcissels rarely nest on the ground and rely on dense vegetation with nearly complete overhead cover to conceal the nest (Temple 2002). Moreover, dickcissel daily nest-survival rates have been shown to be positively related to increased litter cover (Hughes et al. Table 4. Mayfield daily nest survival and nest success for dickcissels nesting in traditional and patch-burned (current-year burn patch, 1-year postburn patch, and 2-year postburn patch) pastures at the Tallgrass Prairie Preserve, Oklahoma, USA, 2003 2004. Patch-burn Variable Traditional Current-yr 1-yr postburn 2-yr postburn No. of nests 110 27 63 90 No. of unsuccessful nests 79 16 36 51 Exposure days 1,063 185 568.5 837 Daily nest survival 0.926 0.914 0.937 0.945 SE for daily nest survival 0.008 0.021 0.010 0.008 Nest success 0.1989 0.1513 0.2550 0.3051 1600 The Journal of Wildlife Management 72(7)

Figure 3. Daily nest survival rate of dickcissel nests in (a) traditional pastures (n ¼ 63 nests), (b) current-year burn patches (n ¼ 13 nests), (c) 1- year postburn patches (n ¼ 29 nests), and (d) 2-year postburn patches (n ¼ 58 nests) at Tallgrass Prairie Preserve, Oklahoma, USA, 2003 2004. Dashed lines represent 95% confidence limits for the logistic-exposure model (Shaffer 2004). 1999). Unlike the traditional treatment in which annual spring fire eliminated nearly all litter cover, 2 (1-yr postburn and 2-yr postburn patches) of the 3 patches in patch-burned pastures had high litter accumulation (Table 1; Fuhlendorf et al. 2006). Consequently, dickcissel nests were probably better concealed in the 1-yr postburn and 2-yr postburn patches because of the high litter accumulations as well as the taller vegetation and less bare ground that occurred in those patches. Grazing in combination with other management treatments can negatively impact dickcissel productivity (Dechant et al. 2002). Zimmerman (1997) reported that combined grazing and burning in Kansas, USA, prairie resulted in significantly lower dickcissel productivity compared with prairies that were undisturbed or separately managed by grazing and burning. In Oklahoma, dickcissel nest success was significantly lower in disturbed prairie (burned and grazed prairie and grazed prairie) compared with undisturbed prairie (Rohrbaugh et al. 1999). In our study, dickcissel nest success and daily nest survival were lowest in habitats that received a grazing and prescribed burning treatment (traditional pasture and current-yr burn patch). Although it is difficult to isolate the effects of grazing from the effects of burning, the combination of these treatments affect dickcissel productivity much more than when these treatments are applied separately. The combination of grazing and burning impacts both dense cover (includes litter and live vegetation) and tall vegetation, which are important habitat characteristics for dickcissel nest success (Winter 1999). Specifically, fire eliminates residual litter cover but also stimulates high-quality vegetation regrowth that results in intense grazing pressure in burned areas and a subsequent reduction in vegetation cover and height (Fuhlendorf and Engle 2001, 2004). Moreover, the intense grazing pressure can result in higher nest losses due to trampling of nests by cattle (Rohrbaugh et al. 1999, Churchwell et al. 2005). In our study, the higher predation rates in grazed and burned treatments is likely due to the reduced vegetation cover, which made dickcissel nests more vulnerable to predation. Rohrbaugh et al. (1999) also reported higher rates of nest predation in disturbed prairies than in undisturbed prairies (75% in disturbed prairies vs. 64% in undisturbed prairie) and attributed the higher predation rates to reduced ground cover caused by the fire grazing interaction. Ecological traps are poor-quality habitats that are perceived as good-quality habitats by animals because the habitats retain certain environmental or structural cues that are indicative of good-quality habitat, resulting in a decoupling of habitat selection from habitat quality (Dwernychuk and Boag 1972, Schlaepfer et al. 2002). This decoupling may have detrimental effects on individual fitness and long-term population persistence. In a study evaluating the role management plays in creating ecological traps in the tallgrass prairie, Shochat et al. (2005) determined that annual spring fires followed by highintensity grazing increased arthropod and predator abun- Churchwell et al. Fire Grazing Interactions and Dickcissels 1601

dances, resulting in the habitat quality in these managed pastures being perceived by grassland birds as higher quality than pastures that were undisturbed. Furthermore, Shochat et al. (2005) reported substantially lower nest success and significantly higher nest densities in pastures managed by annual spring burning followed by season-long grazing (i.e., our traditional treatment) compared with undisturbed pastures, suggesting that these pastures are ecological traps. Similarly, our results seem to indicate that traditional managed pastures are ecological traps for dickcissels. In comparison with patch-burned pastures, overall nest success was lower and nest densities and predation were higher in traditional managed pastures. Zimmerman (1997) found dickcissel nesting activity in annually burned and grazed treatments lagged 2 3 weeks behind unburned and ungrazed, annually burned and ungrazed, and unburned and grazed treatments at tallgrass prairie in Kansas. Similarly, we found dickcissels tended to nest later in the traditional managed pasture and currentyear burn patch in the patch-burned pastures than in the 1- year postburn and 2-year postburn patches in the patchburned pastures. Early in the nesting season, standing dead vegetation provides perches for territorial males as well as provides nesting sites for females (Zimmerman 1971, 1982). The delay in nesting activity in the traditional and currentyear patch was likely due to the lack of nesting cover and elimination of standing dead vegetation for territorial males early in the nesting season in these treatments (Robel et al. 1998). Brood parasitism by brown-headed cowbirds was higher in traditional pastures than in patch-burned pastures. Moreover, within patch-burned pastures, only nests in the current-year burn patch were parasitized. Several studies have shown that cowbirds prefer short-grazed habitats over other grassland habitats, presumably for the higher invertebrate densities that occur in grazed habitats (Morris and Thompson 1998; Goguen and Mathews 1999, 2000; Patten et al. 2006). In our study, the combination of grazing and burning may have increased cowbird parasitism rates because burning may have further enhanced the abundance of herbivorous invertebrates (e.g., grasshoppers) and may have facilitated cowbird foraging opportunities through eliminating thick litter accumulation and dense vegetation (Patten et al. 2006). Interestingly, brood parasitism rates in the current-year burn patch (69%) were much higher than in the traditional pasture (19%). Given the low nest density in the currentyear burn patch (0.41 þ 0.16 nests/ha), the higher parasitism rates were not unexpected. Cowbird parasitism rates of dickcissels have been shown to be inversely correlated with dickcissel nest densities (Fretwell 1977, Zimmerman 1983). In Kansas, Zimmerman (1983) reported cowbird parasitism rates of 100% and 60% for dickcissel nest densities of,0.20 nests/ha and 0.25 0.75 nests/ha, respectively. At low nest densities, female cowbirds are more likely to locate the few nests that occurred in these habitats. Moreover, the lack of cover in the current-year burn patch was much greater than in the traditional pasture, possibly making the nests even more vulnerable to parasitism in the current-year burn patch. In particular, current-year patches contained greater amounts of bare ground and shorter vegetation than the traditional pasture (Table 1; Fuhlendorf et al. 2006). We found the overall parasitism rate at the Preserve for dickcissels was 7%, which is similar to parasitism rates recorded by Jensen and Cully (2005b) for dickcissels at the Preserve (5%) but lower than rates reported by Patten et al. (2006) at the Preserve (19%). Overall, our parasitism rates were much lower than rates reported for most other studies in the Great Plains (Shaffer et al. 2003). Jensen and Cully (2005a) suggested that the low parasitism rates in some areas of the Great Plains are related to the local density of female cowbirds. Moreover, cowbird parasitism rates seem to correspond with the latitudinal pattern of cowbird abundance observed in the Great Plains; brown-headed cowbird abundances increase with latitude in the Great Plains (Sauer et al. 2005). At the Preserve, cowbird abundance was quite low (0.32 birds/16 ha; Coppedge et al. 2008), which suggests that the lower parasitism rates in our study may be attributed to low cowbird abundance. For comparison, cowbird abundance and parasitism rates at tallgrass prairie in Kansas were 7.70 birds/16 ha (Powell 2006) and 80% (Jensen and Cully 2005b), respectively. Furthermore, other factors such as proximity to perches and edge habitats and size of the pastures may have influenced cowbird parasitism at the Preserve. Several studies have suggested that nests located close to perches and edge habitats are more vulnerable to parasitism (Jensen and Finck 2004, Jensen and Cully 2005a, Patten et al. 2006). Because we conducted our study in large pastures (400 900 ha) embedded in a larger intact grassland (i.e., the entire preserve), the greater distance to perches and edge habitats likely played a role in reducing the frequency of cowbird parasitism. Daily nest survival of dickcissels decreased during the breeding season. In fact, nest age was the most supported model among our candidate models. Several studies of grassland birds in the Great Plains have reported declines in daily nest survival as nest age increases (Winter 1999, Shochat et al. 2005, Suedkamp Wells 2005; but see Grant et al. 2005). Several studies have suggested the increased visits to the nest by parents may make the nest more vulnerable to predators (Conway and Martin 2000, Martin et al. 2000). Additionally, the decline in daily nest survival may also reflect an additive exposure to risk as the incubation and brood rearing periods extend (Grant et al. 2005). Specifically, the longer a nest remains active, there is a greater likelihood that the eggs and nestlings will be lost to predation, weather, or other factors (Grant et al. 2005). Finally, daily nest survival was higher in 2003 than 2004. Weather conditions may have played a role in these differences in nest survival estimates. In 2003, temperature and precipitation slightly deviated from the long-term average for the region, whereas in 2004, temperature was below the long-term average and precipitation was above the 1602 The Journal of Wildlife Management 72(7)

long-term average for the region (National Oceanic and Atmospheric Administration 2005). MANAGEMENT IMPLICATIONS Our results suggest that patch-burn management can be a viable option for enhancing grassland bird productivity. By creating a mosaic of different stature vegetation, patch-burn management enhances productivity of grassland bird species by providing a refuge area in the unburned patches that affords dickcissels and other nesting grassland birds some protection from the direct (e.g., trampling) and indirect (e.g., cowbird parasitism and predation) effects of grazing, which are not available under traditional management. Moreover, patch-burn management does provide habitat for a greater suite of grassland bird species than does traditional management (Fuhlendorf et al. 2006, Coppedge et al. 2008). ACKNOWLEDGMENTS We thank the United States Department of Agriculture for providing funding for this project through their National Research Initiative Competitive Grants Program, grant 2003-35101-12928. The Nebraska Chapter of The Nature Conservancy provided funding for field equipment and travel through J. E. Weaver Competitive Grant Program. The Nature Conservancy s Tallgrass Prairie Preserve provided housing and allowed us access to their property. The Oklahoma Cooperative Fish and Wildlife Research Unit (Oklahoma State University, Oklahoma Department of Wildlife Conservation, United States Geological Survey, Wildlife Management Institute, and United States Fish and Wildlife Service cooperating) provided administrative support for the project, and the Oklahoma State University Zoology Department and the Oklahoma Agricultural Experiment Station provided additional support. We thank T. Shaffer for guidance in analyzing nest success data. T. J. O Connell, D. M. Leslie, Jr., S. S. Rosenstock, J. P. McCarty, and D. Perkins provided helpful comments on earlier drafts of the manuscript. Finally, we are grateful to our dedicated field technicians: C. Picinich, A. Thomas, J. Bryant, and J. Sterling Graber. LITERATURE CITED Agresti, A. 1996. An introduction to categorical data analysis. John Wiley and Sons, Inc., New York, New York, USA. Beason, R. C. 1995. Horned Lark (Eremophila alpestris). Account 195 in A. Poole and F. Gill, editors. The birds of North America, The Academy of Natural Sciences, Philadelphia, Pennsylvania, and The American Ornithologists Union, Washington, D.C., USA. Bourlier, B. G., J. D. Nichols, and W. J. Ringwald. 1979. Soil survey of Osage County, Oklahoma. U.S. Department of Agriculture Soil Conservation Service and U.S. Department of Interior Bureau of Indian Affairs, Washington, D.C., USA. Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: a practical information-theoretic approach. Second edition. Springer-Verlag, New York, New York, USA. Churchwell, R. C. 2005. The influence of patch-burn management on the nesting ecology of grassland birds at the Tallgrass Prairie Preserve, Oklahoma. Thesis, Oklahoma State University, Stillwater, USA. Churchwell, R. C., C. A. Davis, S. D. Fuhlendorf, and D. M. Engle. 2005. Impacts of cattle grazing on grassland nesting birds. Bulletin of the Oklahoma Ornithological Society 38:25 30. Conway, C. J., and T. E. Martin. 2000. Evolution of passerine incubation behavior: influence of food, temperature, and predation. Evolution 54: 670 685. Coppedge, B. R., D. M. Engle, C. S. Toepfer, and J. H. Shaw. 1998. Effects of seasonal fire, bison grazing, and climatic variation on tallgrass prairie vegetation. Plant Ecology 139:235 246. Coppedge, B. R., S. D. Fuhlendorf, W. C. Harrell, and D. M. Engle. 2008. Avian community response to vegetation and structural features in grasslands managed with fire and grazing. Biological Conservation 141: 1196 1203. Dechant, J. A., M. L. Sondreal, D. H. Johnson, L. D. Igl, C. M. Goldade, A. L. Zimmerman, and B. R. Euliss. 2002. Effects of management practices on grassland birds: dickcissel. Northern Prairie Wildlife Research Center, Jamestown, North Dakota, USA. Dwernychuk, L. W., and D. A. Boag. 1972. Ducks nesting in association with gulls: an ecological trap? Canadian Journal of Zoology 50:559 563. Faaborg, J. 2002. Saving migrant birds developing strategies for the future. University of Texas Press, Austin, USA. Fretwell, S. D. 1977. Is the dickcissel a threatened species? American Birds 31:923 932. Fuhlendorf, S. D., and D. M. Engle. 2001. Restoring heterogeneity on rangelands: ecosystem management based on evolutionary grazing patterns. Bioscience 51:625 633. Fuhlendorf, S. D., and D. M. Engle. 2004. Application of the fire-grazing interaction to restore a shifting mosaic on tallgrass prairie. Journal of Applied Ecology 41:604 614. Fuhlendorf, S. D., W. C. Harrell, D. M. Engle, R. G. Hamilton, C. A. Davis, and D. M. Leslie, Jr. 2006. Should heterogeneity be the basis for grassland conservation? Response of grassland birds to the fire-grazing interaction. Ecological Applications 16:1706 1716. Goguen, C. B., and N. E. Mathews. 1999. Review of the causes and implications of the association between cowbirds and livestock. Studies in Avian Biology 18:10 17. Goguen, C. B., and N. E. Mathews. 2000. Local gradients of cowbird abundance and parasitism relative to livestock grazing in a western landscape. Conservation Biology 14:1862 1869. Grant, T. A., T. L. Shaffer, E. M. Madden, and P. J. Pietz. 2005. Timespecific variation in passerine nest survival: new insights into old questions. Auk 122:661 672. Harrell, W. C. 2004. Importance of heterogeneity in the grassland ecosystem. Dissertation, Oklahoma State University, Stillwater, USA. Herkert, J. R. 1994a. Status and habitat selection of the Henslow s sparrow in Illinois. Wilson Bulletin 106:35 45. Herkert, J. R. 1994b. The effects of habitat fragmentation on Midwestern grassland bird communities. Ecological Applications 4:461 471. Herkert, J. R., and F. L. Knopf. 1998. Research needs for grassland bird conservation. Pages 273 282 in J. M. Marzluff and R. Sallabanks, editors. Avian conservation: research and management. Island Press, Washington D.C., USA. Hosmer, D. W., Jr., and S. Lemeshow. 2000. Applied logistic regression. John Wiley and Sons, New York, New York, USA. Hughes, J. P., R. J. Robel, K. E. Kemp, and J. L. Zimmerman. 1999. Effects of habitat on dickcissel abundance and nest success in Conservation Reserve Program fields in Kansas. Journal of Wildlife Management 63:523 529. Jensen, W. E., and J. F. Cully, Jr. 2005a. Density-dependent habitat selection by brown-headed cowbirds (Molothrus ater) in tallgrass prairie. Oecologia 142:136 149. Jensen, W. E., and J. F. Cully, Jr. 2005b. Geographic variation in brownheaded cowbird (Molothrus ater) parasitism on dickcissels (Spiza americana) in Great Plains tallgrass prairie. Auk 122:648 660. Jensen, W. E., and E. J. Finck. 2004. Edge effects on nesting dickcissels (Spiza americana) in relation to edge type of remnant tallgrass prairie in Kansas. American Midland Naturalist 151:192 199. Johnson, D. H. 1979. Estimating nest success: the Mayfield method and an alternative. Auk 96:651 661. Johnson, D. H. 2001. Habitat fragmentation effects on birds in grasslands and wetlands: a critique of our knowledge. Great Plains Research 11:211 231. Churchwell et al. Fire Grazing Interactions and Dickcissels 1603

Knopf, F. L. 1994. Avian assemblages on altered grasslands. Studies in Avian Biology 15:247 257. Knopf, F. L. 1996. Prairie legacies birds. Pages 135 147 in F. B. Samson and F. L. Knopf, editors. Prairie conservation: preserving North America s most endangered ecosystem. Island Press, Washington D.C., USA. Martin, T. E., and G. R. Geupel. 1993. Nest-monitoring plots: methods for locating nests and monitoring success. Journal of Field Ornithology 64:507 519. Martin, T. E., J. Scott, and C. Menge. 2000. Nest predation increases with parental activity: separating nest site and parental activity effects. Proceedings of the Royal Society of London, Series B 267:2287 2293. Mayfield, H. E. 1961. Nesting success calculated from exposure. Wilson Bulletin 73:255 261. Mayfield, H. E. 1975. Suggestions for calculating nest success. Wilson Bulletin 87:456 466. Morris, D. L., and F. R. Thompson, III. 1998. Effects of habitat and invertebrate density on abundance and foraging behavior of brownheaded cowbirds. Auk 115:376 385. National Oceanic and Atmospheric Administration. 2005. National Weather Service internet services team.,http://cdo.ncdc.noaa.gov/dly/ DLY.. Accessed 1 Dec 2006. Patten, M. A., E. Shochat, D. L. Reinking, D. H. Wolfe, and S. K. Sherrod. 2006. Habitat edge, land management, and rates of brood parasitism in tallgrass prairie. Ecological Applications 16:687 695. Peak, R. G., F. R. Thompson, III, and T. L. Shaffer. 2004. Factors affecting songbird survival in riparian forests in a midwestern agricultural landscape. Auk 121:726 737. Peterjohn, B. G., and J. R. Sauer. 1999. Populations status of North American grassland birds from the North American breeding bird survey 1966 1996. Studies in Avian Biology 19:27 44. Powell, A. F. L. A. 2006. Effects of prescribed burns and bison (Bos bison) grazing on breeding bird abundances in tallgrass prairie. Auk 123:183 197. Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 132:652 661. Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of field methods for monitoring landbirds. General Technical Report PSW-GTR-144. U.S. Forest Service, Pacific Southwest Research Station, Albany, California, USA. Robel, R. J., J. P. Hughes, S. D. Hull, K. E. Kemp, and D. S. Klute. 1998. Spring burning: resulting avian abundance and nesting in Kansas CRP. Journal of Range Management 51:132 138. Rohrbaugh, R. W., D. L. Reinking, D. H. Wolfe, S. K. Sherrod, and M. A. Jenkins. 1999. Effects of prescribed burning and grazing on nesting and reproductive success of three grassland passerine species in tallgrass prairie. Studies in Avian Biology 19:165 170. Samson, F. B., and F. L. Knopf. 1994. Prairie conservation in North America. Bioscience 44:418 421. Sauer, J. R., J. E. Hines, and J. Fallon. 2005. The North American breeding bird survey, results and analysis 1966 2004. Version 2005.2. U.S. Geological Survey, Patuxent Wildlife Research Center, Laurel, Maryland, USA. Schlaepfer, M. A., M. C. Runge, and P. W. Sherman. 2002. Ecological and evolutionary traps. Trends in Ecology and Evolution 17:474 480. Shaffer, J. A., C. M. Goldade, M. F. Dinkins, D. H. Johnson, L. D. Igl, and B. R. Euliss. 2003. Brown-headed cowbirds in grasslands: their habitats, hosts, and response to management. Prairie Naturalist 35:145 186. Shaffer, T. L. 2004. A unified approach to analyzing nest success. Auk 121: 526 540. Shochat, E., M. A. Patten, D. W. Morris, D. L. Reinking, D. H. Wolfe, and S. K. Sherrod. 2005. Ecological traps in isodars: effects of tallgrass prairie management on bird nest success. Oikos 111:159 169. Suedkamp Wells, K. M. 2005. Resource selection, movement patterns, and survival of post-fledging grassland birds in Missouri. Dissertation, University of Missouri, Columbia, USA. Temple, S. A. 2002. Dickcissel (Spiza americana). Account 703 in A. Poole and F. Gill, editors. The birds of North America, The Birds of North America, Incorporated, Philadelphia, Pennsylvania, USA. Vickery, P. D., P. L. Tubaro, J. M. C. Silva, B. G. Peterjohn, J. R. Herkert, and R. B. Cavalcanti. 1999. Conservation of grassland birds in the Western Hemisphere. Studies in Avian Biology 19:2 26. Winter, M. 1999. Nesting biology of dickcissels and Henslow s sparrows in southwestern Missouri prairie fragments. Wilson Bulletin 111:515 527. Winter, M., S. E. Hawks, J. A. Shaffer, and D. H. Johnson. 2003. Guidelines for finding nests of passerine birds in tallgrass prairie. Prairie Naturalist 35:197 211. Zimmerman, J. L. 1971. The territory and its density dependent effect in Spiza americana. Auk 88:591 612. Zimmerman, J. L. 1982. Nesting success of dickcissels in preferred and less preferred habitats. Auk 99:292 298. Zimmerman, J. L. 1983. Cowbird parasitism of dickcissels in different habitats and at different nest densities. Wilson Bulletin 95:7 22. Zimmerman, J. L. 1997. Avian community responses to fire, grazing, and drought in the tallgrass prairie. Pages 167 179 in F. L. Knopf and F. B. Samson, editors. Ecology and conservation of Great Plains vertebrates. Springer-Verlag, New York, New York, USA. Associate Editor: Rosenstock. 1604 The Journal of Wildlife Management 72(7)