Feeding Habitat Selection by Great Blue Herons and Great Egrets Nesting in East Central Minnesota

Transcription

1 Feeding Habitat Selection by Great Blue Herons and Great Egrets Nesting in East Central Minnesota CHRISTINE M. CUSTER 1 AND JOAN GALLI 2 1 USGS, Upper Midwest Environmental Sciences Center, 2630 Fanta Reed Road, La Crosse, WI USA Internet: Christine_custer@usgs.gov 2 Minnesota Department of Natural Resources, Nongame Wildlife Program 500 Lafayette Road, St. Paul, MN USA Abstract. Great Blue Herons (Ardea herodias) and Great Egrets (Casmerodius albus) partitioned feeding habitat based on wetland size at Peltier Lake rookery in east central Minnesota. Great Blue Herons preferred large waterbodies ( 350 ha), whereas Great Egrets fed most often at small ponds (<25 ha). Forty-nine percent of Great Blue Herons used wetlands hectares in size and 83% of Great Egrets fed in wetlands <100 ha in size. Great Blue Herons selected large wetlands more often than expected both at the regional (30-km radius) and local (4-km radius) scales. Habitat use by Great Egrets was in proportion to availability at the regional scale, but they selected smaller wetlands for feeding more often than expected at a local scale. The median flight distance of Great Blue Herons was 2.7 km, similar to distances reported elsewhere. Great Egrets flew farther to feeding sites than Great Blue Herons, and flew farther (median = 13.5 km) than reported in other geographic areas. Received 22 September 2001, accepted 5 November Key words. Great Blue Herons, Ardea herodias, Great Egrets, Casmerodius albus, feeding habitat, Peltier Lake, Minnesota. Waterbirds 25(1): , 2002 Numbers of Great Blue Herons (Ardea herodias) and Great Egrets (Casmerodius albus) have declined in the Upper Midwest of the USA (Thompson 1977, 1978a). The declines have been most pronounced where habitats have been degraded or lost. For example, numbers of Great Blue Herons and Great Egrets are larger in the northern portion of the Upper Mississippi River, above Rock Island, Illinois where the floodplain of the Mississippi River has not been as extensively modified as in the southern portion (Thompson 1978a). In the northern portion of the Upper Mississippi River, increasing, stable, or new wading bird rookeries compensate for declining or extinct rookeries; this is not the case in the southern portion of the river. Thompson (1978a) noted an inverse relationship between human numbers and the amount of land planted with agricultural crops on the one hand, and wading bird numbers in the Upper Midwest. It is important to characterize habitat use of herons and egrets so that appropriate habitat conservation measures can be implemented where needed. Quality feeding habitat in proximity to breeding rookeries is an important aspect that governs the general health of colonial waterbird colonies. Quality habitat is defined by physical, biological, and anthropogenic parameters, such as water depth (Custer and Osborn 1978a), turbidity (Krebs 1974), vegetation characteristics (Thompson 1978b), food abundance (Butler 1993; Gibbs and Kinkel 1997), distance from breeding site (Gibbs 1991), and distance from human disturbance (Miller 1943). Most feeding habitat selection studies of nesting Great Blue Herons and Great Egrets have been made in coastal habitats (Custer and Osborn 1978a 1978b; Simpson et al. 1987), along rivers (Thompson 1978a; Dowd and Flake 1985; Maccarone and Parsons 1994) or in large lake situations (Smith 1995a; 1995b). In coastal situations, habitat use is driven mainly by tidal fluctuations (Custer and Osborn 1978a) and hydrologic changes (Smith 1995a). In riverine habitats, usually only one aquatic habitat type (riverine) is available. Few studies have quantified habitat use in areas where a mosaic of different-sized habitats is available, nor have they examined foraging habitat selection at differing scales. 115

2 116 WATERBIRDS Peltier Lake rookery (45 10 N W) is located northeast of the Minneapolis/St. Paul Twin Cities metropolitan area (Fig. 1). This island rookery was established in 1989 (N = 12 Great Blue Heron nests) and had expanded to 500 nests by Great Egrets were first noted in the Peltier Lake rookery in Except for 1996 when some Blackcrowned Night Herons (Nycticorax nycticorax) nested at Peltier Lake, Great Blue Herons and Great Egrets are the only two nesting species in the rookery. The habitat in the vicinity of Peltier Lake contains a mosaic of wetland sizes ranging from small ponds (<0.1 ha in size) to large lakes ( ha) (See Fig. 1, lower panel for the habitat mosaic). Peltier Lake heron rookery is increasingly threatened by urban sprawl from the Twin Cities metropolitan area. Although Peltier Lake is afforded protection, surrounding feeding sites need to be identified and protected. The objectives of this study were to identify the geographic distribution of feeding sites for both wading bird species, to determine what attributes were associated with feeding sites, to determine whether habitats were being selected or used in proportion to availability, and to assess habitat partitioning. METHODS Individual Great Blue Herons (N = 63) and Great Egrets (N = 29) were followed from their nesting colony to their first feeding location using a fixed-winged aircraft, (Custer and Osborn 1978a, Custer and Bunck 1992). Between 1 May and 15 June, 1998 flights were made approximately once per week, with a total of seven days of observations. The breeding rookery was circled in a counter-clockwise direction at an altitude of approximately 300 m above the ground. A bird was selected for following as it left the rookery. The location where the bird landed to feed was mapped using global position system equipment and/or landmarks. General feeding habitat categories included: ponds = small bodies of water (<10 ha); lakes = large bodies of water (>10 ha); marshes = shallow bodies of water with emergent vegetation prominent; and rivers or creeks. Landing substrates were classified as bare shoreline; dead or live vegetation whether at shoreline or not; man-made structure, log, snag, or rock; or in open water. Feeding habitat categories and landing substrate were assigned while circling the bird s landing site. The presence or absence of other herons and egrets on the same lake or pond was noted and the elapsed time since the bird left the nesting colony was recorded, using a stopwatch. After the above data were recorded, the observer returned to the nesting rookery and another bird was followed. As many birds as possible were followed during a 2- to 3-hr period in the morning and a similar period in the afternoon. Only one species was followed during a specific time period (i.e. either morning or afternoon) and the species followed was alternated so that birds of one species were followed in the morning and the other in the afternoon. The time of day when a flight was made, however, was probably unimportant because Great Blue Herons feed throughout the day (McNeil et al. 1993). In Belgium, radio-marked Grey Herons (Ardea cinerea) flew directly from the colony to a feeding site and remained there until they returned to the colony (Van Vessem et al. 1984). Grey herons were not observed to move between feeding sites without first returning to the colony. Consequently, feeding locations identified in this study should be representative of feeding locations of birds from this colony. Six Great Egrets flew to an adjacent roost site and were not included in any of the analyses that follow. The actual size of water bodies (ha) in the study area was derived from National Wetland Inventory (NWI) maps using ArcView software (Environmental Systems Research Institute, Inc.). The area used for feeding by Great Blue Herons and Great Egrets from Peltier Lake rookery encompassed seventeen 7.5-min. quadrangle sheets (quads); this was the basic unit used for display and calculation purposes. These precise wetland size data were used to corroborate categorical assessments made from the airplane, to test for differences in the size of wetlands selected for feeding (median test, Siegel 1956:111), and to test whether habitats used were in proportion to those available (Neu et al. 1974). The total area of surface water available was quantified and put into six 100-ha wetland size categories (0-100 ha, ha, etc.). Water bodies on the outside boundaries of the seventeen quad sheets area were truncated at the boundary. The small proportion of truncated wetlands did not affect the surface water area calculations because of the small surface area involved relative to the total surface area in the 17 quads. Availability of wetland size classes were summarized at two scales, the maximum distance either species flew from Peltier Lake and the 17 NWI quads associated with those feeding sites (regional scale), and a 4-km radius from the colony (local scale). The 4-km radius was chosen because this was the average or median feeding distance reported by others for these two species (Custer and Osborn 1978a; Thompson 1978b). To test whether habitats were used in proportion to availability, the expected numbers of herons or egrets were calculated based on the proportion of differentsized wetlands available at the two scales. The expected number of birds was then compared to the actual number of birds using Chi-square for Great Blue Herons and Fisher s exact test for Great Egrets. Fisher s exact test was used because cell size was <5 in several categories. If the Chi-square or Fisher Exact was significant, then which size categories were preferred, avoided, or used in proportion to availability were determined using the confidence interval procedure of Neu et al. (1974). The calculated confidence intervals took into account the number of simultaneous estimates being made (N = 5) (Neu et al. 1974). If the expected proportion of usage fell outside the 95% confidence intervals for actual usage then that size category was used significantly more or less than expected.

3 HERON AND EGRET FEEDING HABITAT 117 Figure 1. Map of Peltier study area near Minneapolis/St. Paul, Minnesota with an example of the wetland sizes available near the nesting lake.

4 118 WATERBIRDS The median test was used to test for species differences in time and distance flown. Median tests were used because of non-normal distribution of the data and to make it possible to use more of the data. Because we lost nine birds enroute to feeding sites, we did not have the exact time or distance flown for those birds, and only the minimum time and distance flown. Birds lost enroute were generally those that had flown for a long period of time; the median test was appropriate because it did not require the exact time or distance, but only that birds lost in flight had gone farther than the median time or distance. One Great Egret was lost at less than the median distance. The median distance remained the same, 13.5 km, whether this bird was included or excluded from the data, so it was included. Fisher s exact tests were used to test whether Great Blue Herons and Great Egrets used similar feeding habitat types and landing substrates. Fisher Exact was used because some cells contained fewer than 5 observations. If the overall test was significant, then pairwise Fisher Exacts were done to determine differences. The alpha level was reduced to accommodate the increased number of comparisons (P = 0.05 divided by the number of pairwise comparisons). Fisher s exact tests used the actual number of individuals; proportions, however, are presented in the tables for easier comparisons. River and Other habitats were combined so that no cell contained a zero. Sample sizes for the various endpoints are not always equal. Some birds were lost enroute, so the landing habitat was not available for those birds. It also was not always possible to see the landing substrate used by an individual, and occasionally, the stopwatch was not set. RESULTS Great Blue Heron feeding sites were spread across the landscape in a widely dispersed pattern (Fig. 2). Many Great Blue Herons (N = 27 of 63) fed on the lake where the nesting colony was located and the modal distance flown to feeding sites was < 1 km from the colony (Fig. 3). Great Egrets also fed at widely scattered locations (Fig. 4). No Great Egrets, however, fed on the lake where the nesting colony was located or within 4 km of it (Fig. 3). Only four of 23 Great Egrets fed at locations south of the rookery, whereas about half of the Great Blue Herons fed south of the rookery. The maximum distance flown south was 6.3 km for Great Blue Herons and 7.9 km for Great Egrets. The maximum distance flown to the north was 15.3 and 30.3 km. The median time Great Blue Herons flew (4.9 min) to reach a feeding site was significantly shorter than that for Great Egrets (20.0 min) (Table 1, P < 0.01). The median distance flown was also significantly shorter Figure 2. Location of Great Blue Heron feeding sites for birds nesting at Peltier Lake rookery, Minnesota in for Great Blue Herons (2.7 km) than for Great Egrets (13.5 km, P < 0.001). The maximum time, 43 minutes, and distance flown (27 and 30 km) were similar for both species. At Peltier Lake rookery, Great Blue Herons selected lakes for feeding (76% of the cases) more often than Great Egrets did (13%), whereas, Great Egrets selected ponds (65% of cases) more often than Great Blue Herons (13%, Table 2, P < 0.001). Rivers and creeks were used in less than 5% of the cases by both species. Only 7% of the feeding sites used by Great Blue Herons were in marshes, compared to 17% of the feeding sites for Great Egrets. There was no difference in marsh, river, or other habitat use by the two species. The median size of the water body used for feeding was significantly larger for Great Blue Herons (372 ha) than for Great Egrets (4.4 ha) (P < 0.001, Table 1, Fig. 5). Most (>50%) feeding sites were over 350 ha for Great Blue Herons, while most feeding sites for Great Egrets were under 25 ha in size. Eighty-three percent of feeding sites used by Great Egrets were in wetlands ha in size (Table 3), but only 20% of feeding sites were in that size category for Great Blue Her-

5 HERON AND EGRET FEEDING HABITAT 119 Figure 3. Distances that Great Egrets (top panel) and Great Blue Herons (bottom panel) flew to feeding sites while nesting at Peltier Lake rookery, Minnesota in Each bar represents a 1-km interval and the x-axis numbers denotes the end of the interval, i.e. the 5 represents the 4-5 km interval. ons (Table 4). Forty-nine percent of Great Blue Heron feeding sites were in the ha size class (Table 4). Great Blue Herons used the ha size wetlands less than expected and the ha size classes more than expected at the regional scale (P < 0.001, Table 4). At the local scale, Great Blue Herons used the ha size more than expected based on availability (P < 0.001, Table 4). On the regional scale, Great Egrets used the six-wetland size categories in proportion to their availability (χ 2 5 = 0.28, n.s., Table 3). On the local scale, however, Great Egrets used the ha size more than expected and the and >500 ha size less than expected (P < 0.001). The other wetland sizes were used in proportion to their availability. Great Blue Herons tended to land most frequently on vegetation mats (27%) and non-vegetated shorelines (36%) (Table 5). The most frequent landing substrates for Great Egrets were vegetation mats (38%) or directly in the water (33%). There were no differences between the two species in landing substrates (χ 2 3 = 2.28, n.s.). Great Blue Herons landed at a feeding site with no other herons or egrets present 39 times (87%) and with other herons or egrets present only six times. In these six instances, one Great Egret was present once, one other Great Blue Heron was present three times, and two and three other Great Blue Herons were present on one occasion each. In five of these six instances there was no interaction among the birds. In one instance, the incoming heron displaced the Great Blue Herons already present at the feeding site. Great Egrets landed where no other birds were present 20 times (74%) and where other egrets were already present seven times. During these seven instances there were six Great Egrets present once, three egrets present three times, two egrets present twice and one egret present once. In five of these instances, there was no interaction among the Great Egrets. In one instance, the incoming bird chased others off the feeding site and in the other instance the incoming bird gave way to the individual already present. We did not observe any Great Egrets landing where Great Blue Herons were already present. Figure 4. Location of Great Egret feeding sites for birds nesting at Peltier Lake rookery, Minnesota in 1998.

6 120 WATERBIRDS Table 1. Feeding flight information for Great Blue Herons and Great Egrets nesting at Peltier Lake rookery near Minneapolis/St. Paul, Minnesota in Great Blue Heron Great Egret N Median Maximum N Median Maximum P-values* Time (min) <0.01 Distance (km) <0.001 Wetland size (ha) <0.001 *P-values for median tests comparing between species. During the flights, 13 Great Blue Herons were observed to feed by plunge diving. When feeding in this manner, each heron dropped into the water as it slowly flew just above the surface of the lake. This feeding behavior was observed on four dates. On these four occasions, Great Blue Herons other than those being followed, were also observed feeding in this manner. DISCUSSION Distances flown by Great Blue Herons nesting at Peltier Lake to feeding sites were similar to distances reported in other studies. Dowd and Flake (1985) found that the average distance flown by Great Blue Herons was 3.1 km (max km). Eighty-five percent of Great Blue Herons in that study fed within 4 km of the colony. Along the Mississippi River in central Minnesota, the average distance flown by 17 Great Blue Herons was 6.5 km (max 20.4 km); 53% fed within 4 km of the colony (Thompson 1978a). This compares to a median distance of 2.7 km, and 59% of Great Blue Herons feeding within 4 km of the rookery at Peltier Lake. Great Egrets in the southeastern U.S. had mean flight distances of km (max. = 27.8 km; Custer and Osborn 1978a) and 3.7 km (max. = 33.3 km; Smith 1995a). Along the Mississippi River in Minnesota, the mean distance was 8.6 km (max = 35.2 km), and 54% of Great Egrets fed within 4 km of the colony (Thompson 1978b). Data from these three studies had mean distances less than at Peltier Lake, where we found a median distance of 13.5 km. The longer distances that Great Egrets flew to feed may have negative repercussions on nesting success. Simpson et al. (1987) found that locally feeding Great Blue Herons had higher rates of nest success than herons that flew to distant feeding locations, although this premise was not supported by Smith s (1995) data for Great Egrets in Florida. Extremely good feeding opportunities at distance sites might compensate for the additional energetic requirements of flying to those distant sites. Herons and egrets from Peltier rookery did not fly more than 6-8 km south of the colony to feed, but flew up to 30 km north to feed. This may have been due to the increasingly urbanized landscape to the south or because feeding areas to the south were used by herons and egrets from other rookeries in the Twin Cities area. In contrast to Peltier Lake, where no Great Egrets fed on the lake where the nest- Table 2. Percentage of feeding habitats used by Great Blue Herons and Great Egrets nesting at Peltier Lake rookery near Minneapolis/St. Paul, Minnesota in 1998, as determined from the air Lake Pond Marsh River Other 1 Great Blue Heron 76 a 2 13 b (N = 55) Great Egret 13 b 65 a (N = 23) 1 Small drainage ditch. 2 Overall χ 2 3 = 28.7, P < 0.001; river and other were combined. Percentages with different letters are significantly different between species. Percentages not followed by a letter do not differ between species.

7 HERON AND EGRET FEEDING HABITAT 121 Figure 5. Size of wetlands used, in 25-ha units, for feeding by Great Egrets and Great Blue Herons nesting at Peltier Lake rookery, Minnesota in ing colony was located nor within 4 km of the rookery, 55% of Great Egrets on Lake Okeechobee, Florida fed on the lake were the rookery was located (Smith 1995a). Lake Okeechobee, a large 1,732 km 2 lake, offered a great diversity of foraging opportunities, so Great Egrets only had to shift to other parts of the lake or to other nearby aquatic habitats to find preferred feeding habitat as the hydrologic regime changed on the lake. It is not known why Great Egrets in this study did not feed on Peltier Lake or within the first 4 km of the colony except that preferred habitat (small ponds) was in short supply (<20% of total available) close to the colony. Others have not reported this donut-effect. The Great Egret s strong preference, on the local scale, for small water bodies might imply that they preferred shallower water for feeding. Willard (1977) documented that Great Egrets fed by slow wading. Peltier Lake and the surrounding large lakes probably only had a very narrow strip of shallow water suitable for wading around the perimeter, effectively reducing the area of preferred feeding habitat for Great Egrets. Great Blue Herons, however, that stand-and-wait (Willard 1997) or plunge-dive had much more feeding habitat available on Peltier and nearby large lakes. Rogers and Nesbitt (1979) suggested that White Ibis (Eudocimus albus) favored distant feeding locations to prevent depletion of the food resource nearer the colony that was important to newly fledged young. This hypothesis is consistent with the data we presented from Peltier Lake for Great Egrets, but not for Great Blue Herons. It would seem disadvantageous for Great Egrets to fly to distance feeding locations to preserve local feeding opportunities for their young when Great Blue Herons are feeding locally. If, however, Great Egrets and Table 3. Numbers and proportions of Great Egrets using each wetland size category at Peltier Lake rookery near Minneapolis/St. Paul, Minnesota in Proportion of wetlands available and preference for each wetland size presented at two different scales. Wetland size (ha) Number and (proportion) using 95% CI for proportion used 3 Proportion of habitat available, expected usage 1, and preference category 2 Regional scale Local scale (0.83) more (0.04) less (0.04) (0.04) (0.04) less Fisher Exact 5 observed vs expected P = 0.17 P < Number of birds expected to use each habitat-size category based on proportion of habitat available. 2 birds using that wetland size in proportion to its availability, more = birds using that wetland size significantly more than expected based on availability, and less = birds using that wetland size less than expected based on availability. 3 If proportion of habitats available falls outside the 95% CI, then birds are using that size category differently than expected based on availability.

8 122 WATERBIRDS Table 4. Numbers and proportions of Great Blue Herons using each wetland size category at Peltier Lake rookery near Minneapolis/St. Paul, Minnesota in Proportion of wetlands available and preference for each wetland size at two different scales. Wetland size (ha) Number and (proportion) using 95% CI for proportion used 3 Proportion of habitat available, expected usage 1, and preference category 2 Regional scale Local scale (0.20) less (0.09) (0.18) more (0.49) more more (0.04) less less χ 2 observed vs expected χ 2 4 = 42.2, P < χ 2 4 = 31.2, P < Number of birds expected to use each habitat-size category based on proportion of habitat available. 2 birds using that wetland size in proportion to its availability, more = birds using that wetland size more than expected base on availability, and less = birds using that wetland size less than expected based on availability. 3 If proportion of habitats available falls outside the 95% CI, then birds are using that size category significantly more or less than expected based on availability. Great Blue Herons are feeding on different sizes of prey, Rogers and Nesbitt s hypothesis could be supported. The low level of aggressive encounters between Great Egrets at Peltier Lake (5%) was similar to that reported by Custer and Osborn (1978a) in South Carolina (6%). The gregariousness of Great Egrets nesting at Peltier Lake seemed to be lower than that reported elsewhere. Smith (1995b) reported that only 29% of Great Egrets landed alone in Florida. This contrasts with 74% of Great Egrets landing alone at Peltier Lake. A more abundant food resource (see Kushlan [1978] for general summary) or more nesting birds in Lake Okeechobee might account for this higher level of gregariousness. Additionally, small wetland size, such as are present in the Peltier Lake area, should preclude large feeding aggregations. The combination of a large number of small wetlands available near Peltier Lake, the small number of egrets nesting in the colony, and wading birds tendency to return to the same feeding area repeatedly (Dowd and Flake 1985; Van Vessem et al. 1984) may have contributed to this widely dispersed, but mostly solitary feeding pattern. Thompson (1978b) noted that two of 17 Great Blue Herons in his study plunged into deep water; he was not sure if they obtained food. In both cases, the birds continued on their flight path and plunge diving was not their preferred feeding method. Plunge diving was observed for Great Egrets, but not for Great Blue Herons, in New Jersey (Willard 1977). Thirteen of the Great Blue Herons that we followed, plus additional herons that we casually observed, used plunged-diving to feed on Peltier Lake; no Great Egrets were observed feeding in this manner. This unusual feeding method may have been energetically feasible on Peltier Lake if a large number of forage fish were present near the surface (Kushlan 1978). Rogers Jr. (1974) noted plunge diving by Great Egrets in Lou- Table 5. Types of landing substrates used (percentages) by Great Blue Herons and Great Egrets nesting at Peltier Lake rookery near Minneapolis/St. Paul, Minnesota in Vegetation Shoreline Water Stuctures Great Blue Heron (N = 55) Great Egret (N = 21) Overall χ 2 3 = 2.28, n.s.

9 HERON AND EGRET FEEDING HABITAT 123 isiana; these egrets were feeding on dead fish floating at the surface. It is unknown on Peltier Lake whether the plunge-diving herons were feeding on live or dead fish or whether they even caught fish. The use of this foraging method on 4 of 5 days, along with multiple birds feeding in this manner, would argue for them catching fish, even though this could not be confirmed from the airplane. Heron and egret species minimize competition among themselves by utilizing different-sized prey (Willard 1977; Hoffman 1978) or by feeding in different water depths (Willard 1977; Custer and Osborn 1978b). On the Mississippi River, Thompson (1978b) suggested that Great Blue Herons and Great Egrets reduced interspecific competition by utilizing different habitats; Great Blue Herons fed while standing from snags and lodges or nearby trees while egrets foraged by wading in large open marshes. At Peltier Lake, Great Egrets landed and fed from a snag once, whereas Great Blue Herons used structures six times (5 logs/snags and 1 boathouse). At Peltier Lake rookery, Great Blue Herons and Great Egrets partitioned the habitat by feeding in different-sized wetlands. Great Blue Herons fed in large bodies of water (median size 370 ha) compared with Great Egrets (4.4 ha). This is the first time that Great Blue Herons and Great Egrets have been shown to partition habitat during the breeding season based on wetland size. On wintering areas in coastal Texas, however, Great Egrets use mainly small pools (<4-m 2 ) for feeding (Chavez-Ramirez and Slack 1995). Great Blue Herons use bays and lakes (>100 m 2 ) significantly more than expected and small water bodies (pools and ponds m 2 ) significantly less than expected (Chavez-Ramirez and Slack 1995). This is consistent with our data from Peltier Lake. Great Blue Herons and Great Egrets seem to exhibit great variability in how they reduce competition between themselves. In other areas where Great Blue Heron and Great Egrets have been studied, high variability in available habitat-size classes may not have been present, and therefore could not be used as a prominent selection factor. In these other study locations, other factors, such as water depth and tidal fluctuations may be more important than wetland size. Because Great Blue Herons and Great Egrets usually occur together in breeding colonies in the Upper Midwest, this partitioning by size of feeding habitat could be suggested as a means to avoid competition. To answer this question explicitly, however, data from additional colonies would be needed where only great egrets nested or where the mosaic of habitats surrounding the rookery differed from Peltier Lake. Preserving and protecting Great Egret and Great Blue Heron feeding habitats at the Peltier Lake rookery may be difficult because of their widely dispersed feeding sites. Herons and egrets fed solitarily over a 30-km radius. Great Egrets fed in small wetlands, whereas Great Blue Herons fed on the rookery lake, as well as adjacent large lakes. Small wetlands may be more vulnerable to modification than larger wetlands and offer the greater management challenge. ACKNOWLEDGMENTS We thank Dwyer Aviation, Mason City, Iowa for many safe hours of flying; Minneapolis Air Traffic control for allowing us to enter restricted air space as we followed herons and egrets; Jason Rohweder for GIS software support; and Thomas Custer, Eileen Kirsch, Fred Meyer, one anonymous reviewer, and the Editor for comments on earlier drafts of this manuscript. LITERATURE CITED Butler, R. W Time of breeding in relation to food availability of female Great Blue Herons (Ardea herodias). Auk 110: Chavez-Ramirez, F. and R. D. Slack Differential use of coastal marsh habitats by nonbreeding wading birds. Colonial Waterbirds 18: Custer, T. W. and C. Bunck Feeding flights of breeding double-crested cormorants at two Wisconsin colonies. Journal of Field Ornithology 63: Custer, T. W. and R. G. Osborn. 1978a. Feeding habitat used by colonially-breeding herons, egrets and ibises in North Carolina. Auk 95: Custer, T. W. and R. G. Osborn. 1978b. Feeding-site description of three heron species near Beaufort, North Carolina. Pages in Wading Birds (A. Sprunt IV, J. C. Ogden and S. Winckler Eds.). National Audubon Society Research Report No. 7, New York. Dowd, E. M. and L. D Flake Foraging habitats and movements of nesting Great Blue Herons in a prairie river ecosystem, South Dakota. Journal of Field Ornithology 56:

10 124 WATERBIRDS Gibbs, J. P Spatial relationships between nesting colonies and foraging areas of Great Blue Herons. Auk 108: Gibbs, J. P. and L. K. Kinkel Determinants of the size and location of Great Blue Heron colonies. Colonial Waterbirds 20: 1-7. Hoffman, R. D The diets of herons and egrets in southwestern Lake Erie. Pages in Wading Birds (A. Sprunt IV, J. C. Ogden and S. Winckler Eds.). National Audubon Society Research Report No. 7, New York. Krebs, J. R Colonial nesting and social feeding as strategies for exploiting food resources in the Great Blue Heron (Ardea herodias). Behaviour 51: Kushlan, J. A Feeding ecology of wading birds. Pages in Wading Birds (A. Sprunt IV, J. C. Ogden and S. Winckler Eds.). National Audubon Society Research Report No. 7, New York. Maccarone, A. D. and K. C. Parsons Factors affecting the use of a freshwater and an estuarine foraging site by egrets and ibises during the breeding season in New York City. Colonial Waterbirds 17: Miller, R. F The Great Blue Heron. The breeding birds of the Philadelphia region (Part II). Cassinia 33: McNeil, R., R. Benoît and J-L. Desgranges Daytime and nighttime activity at a breeding colony of Great Blue Herons in a nontidal environment. Canadian Journal of Zoology 71: Neu, C. W., C. R. Byers and J. M. Peek A technique for analysis of utilization-availability data. Journal of Wildlife Management 38: Rogers, Jr., J. A., Aerial feeding by Snowy and Great Egrets in Louisiana waters. Wilson Bulletin 86: Rogers, Jr., J. A. and S. A. Nesbitt Feeding energetics of herons and ibises at breeding colonies. Proceedings of the Colonial Waterbird Group 3: Siegel, S Nonparametric Statistics for the Behavioral Sciences, McGraw-Hill, New York. Simpson, K., J. N. M. Smith and J. P. Kelsall Correlates and consequences of coloniality in Great Blue Herons. Canadian Journal of Zoology 65: Smith, J. P. 1995a. Foraging flights and habitat use of nesting wading birds (Ciconiiformes) at Lake Okeechobee, Florida. Colonial Waterbirds 18: Smith, J. P. 1995b. Foraging sociability of nesting wading birds (Ciconiiformes) at Lake Okeechobee, Florida. Wilson Bulletin 107: Thompson, D. H Declines in populations of colonial waterbirds nesting within the floodplain of the upper Mississippi River. Proceedings of the Colonial Waterbird Group 1: Thompson, D. H. 1978a. Declines in populations of Great Blue Herons and Great Egrets in five midwestern states. Proceedings of the Colonial Waterbird Group 2: Thompson, D. H. 1978b. Feeding areas of Great Blue Herons and Great Egrets nesting within the flood plain of the upper Mississippi River. Proceedings of the Colonial Waterbird Group 2: Van Vessem, J., D. Draulans and A. F. De Bont Movements of radio-tagged grey herons Ardea cinerea during the breeding season in a large pond area. Ibis 126: Willard, D. E The feeding ecology and behavior of five species of herons in southeastern New Jersey. Condor 79: