Foraging habits of spring migrating waterfowl in the upper mississippi river and great lakes region
Yüklə 0,59 Mb.
|
- Bu səhifədəki naviqasiya:
- ________________________________________________________________________
- Temporal and Latitudinal Variation in Spring Diet Within a Species
- Longitudinal Variation in Spring Diet Within a Species
Transect 
a a b b c c 
Figure 1.11. Percent of seeds, vegetation and invertebrates consumed by gadwalls at western/eastern transect in the Upper MS River and Great Lakes Region during spring 2007 (least-squares means ± standard error). Different letters indicate significantly different means. Table 1.21. Date of first and last diving duck collected at each study site in 2006 and 2007. ________________________________________________________________________ 2006 2007 SPECIES FIRST LAST FIRST LAST ______________________________________________________________________________________
Cache River N/A N/A 18 March 31 March Illinois River 15 March 10 April 03 March 30 April
Table 1.22. Results from a MANCOVA model evaluating the effects of site, date (Jul), habitat (Hab), and year (Yr) on proportions of invertebrates and seeds consumed by lesser scaup in the Upper MS River and Great Lakes Region during spring 2006 and 2007. ________________________________________________________________________ Source DF Type III SS Mean Square F value Pr > F ________________________________________________________________________
scaup at the Scioto River study site than at the Wisconsin and Illinois River study sites (Figure 1.12). Scaup consumed moderately more invertebrates (P = 0.069) in permanent emergent habitat (73.1% ± 11.1%) than in agricultural habitat (16.4% ± 21.1%) and less seeds (P = 0.045) in permanent emergent habitat (21.2% ± 10.4%) than agricultural habitat (81.3% ± 19.7%) and open-water habitat (50.1% ± 6.9%). The final reduced MANCOVA model evaluating longitudinal variation in diets of scaup included the main effects of transect, date, habitat, and year. The percentage of invertebrates in the diet varied significantly among transects, whereas the percentage of seeds in the diet varied significantly among transects and years (Table 1.23). More seeds and, thus, less invertebrates were consumed by scaup in the western transect (Figure 1.13). Scaup consumed 13.78% ± 7.19% more seeds in year 1 than year 2 (P = 0.058).
Two-hundred and five ring-necked ducks (n = 86 in 2006 and n = 119 in 2007) were included in analyses evaluating diet at the scale of study site (Table 1.1). Likewise, 205 ring-necked ducks (n = 132 from the eastern transect and n = 73 from the western transect) were included in the analysis evaluating diet at the scale of transect (n = 59 in 2006 and n = 73 in 2007 form the eastern transect and n = 27 in 2006 and n = 46 in 2007 from the western transect). In 2006, the first ring-necked duck was collected on 23 February and the last on 25 April and in 2007, the first was collected on 25 February and the last on 2 May (Table 1.21). 
a/c d/e a/b d c e a/b/c d/e b d b d 
East Transect Figure 1.12. Percent of seeds and invertebrates consumed by lesser scaup at 6 locations (SR = Scioto River, LE = Lake Erie, SAG = Saginaw Bay, CA = Cache River, IR = Illinois River, WI = Wisconsin) in the Upper MS River and Great Lakes Region during spring 2006 and 2007 (least-squares means ± standard error). Different letters indicate significantly different means. Table 1.23. Results from a MANCOVA model evaluating the effects of transect (Tran), date (Jul), habitat (Hab), and year (Yr) on proportions of invertebrates and seeds consumed by lesser scaup in the Upper MS River and Great Lakes Region during spring 2006 and 2007.
Source DF Type III SS Mean Square F value Pr > F ________________________________________________________________________
Figure 1.13. Percent of seeds and invertebrates consumed by lesser scaup at western/eastern transect in the Upper MS River and Great Lakes Region during spring 2006 and 2007 (least-squares means ± standard error). Different letters indicate significantly different means. The final reduced MANCOVA model evaluating latitudinal variation in diet (invertebrates and seeds) of ring-necked ducks included the main effects of site, date, habitat, and year. The percentage of invertebrates and seeds in the diet varied significantly among study sites (Table 1.24). Ring-necked ducks collected at the Saginaw Bay study site contained more invertebrates and fewer seeds than those collected at the other study sites (P = 0.022). Also, ring-necked ducks collected at the Scioto River study site contained moderately more invertebrates and less seeds than those collected at Wisconsin (P = 0.060) (Figure 1.14). The final reduced MANCOVA model evaluating longitudinal variation in diets of ring-necked ducks included the main effects of transect, date, habitat, and year. The percentage of invertebrates and seeds in the diet varied significantly by transect and date. The percentage of invertebrates in the diet varied significantly between years (Table 1.25). Ring-necked ducks collected in 2006 consumed less invertebrates than those collected in 2007 (P = 0.050). Also, ring-necked ducks on the eastern transect consumed significantly more invertebrates than ring-necked ducks collected on the western transect (P = 0.0001) (Figure 1.15). DISCUSSION Many factors influence diet of waterfowl. Some of which may be foraging behaviors, morphological adaptations, time of year, and food availability (Poysa 1983). An example of a morphological adaptation is bill morphology, with some being narrow (e.g., favoring grazers such as gadwall) and some being broad (e.g., favoring straining as seen in Northern shovelers and teal sp.). Although diets tend to be similar during parts of Table 1.24. Results from a MANCOVA model evaluating the effects of site, date (Jul), habitat (Hab), and year (Yr) on proportions of invertebrates and seeds consumed by ring-necked ducks in the Upper MS River and Great Lakes Region during spring 2006 and 2007. ________________________________________________________________________ Source DF Type III SS Mean Square F value Pr > F ________________________________________________________________________
Figure 1.14. Percent of seeds and invertebrates consumed by ring-necked ducks at 6 locations (SR = Scioto River, LE = Lake Erie, SAG = Saginaw Bay, CA = Cache River, IR = Illinois River, WI = Wisconsin) in the Upper MS River and Great Lakes Region during spring 2006 and 2007 (least-squares means ± standard error). Different letters indicate significantly different means. Table 1.25. Results from a MANCOVA model evaluating the effects of transect (Tran), date (Jul), habitat (Hab), and year (Yr) on proportions of invertebrates and seeds consumed by ring-necked ducks in the Upper MS River and Great Lakes Region during spring 2006 and 2007.
Source DF Type III SS Mean Square F value Pr > F ________________________________________________________________________
Figure 1.15. Percent of seeds and invertebrates consumed by ring-necked ducks at western/eastern transect in the Upper MS River and Great Lakes Region during spring 2006 and 2007 (least-squares means ± standard error). Different letters indicate significantly different means. the annual life cycle of waterfowl (e.g., reproduction), it is likely that feeding niches reduce competition among species and have a large influence on diet (Nudds 1983), as larger dabbling ducks (e.g., mallards) are able to utilize foods vertically deeper in the water column than smaller dabbling ducks (e.g., blue-winged teal). This niche separation has also been demonstrated in foraging depths of dabbling and diving ducks (Green 1998). Diets of ducks during clutch formation tend to be less variable than other times during the annual cycle, as ducks must meet protein demands during egg production by taking advantage of abundant invertebrates on wetlands typically utilized during reproduction. Swanson et al. (1985) reported breeding female mallards in North Dakota consumed > 70% animal matter during RFD. Similarly, the diet of female blue-winged teal, scaup, and ring-necked ducks during RFD resemble that of breeding mallards (Dirschl 1969, Swanson et al. 1974, Hohman 1985). The gadwall, however, is one of the most herbivorous waterfowl species and literature indicates a weaker dependence on invertebrates during RFD, as they may consume as little as 50% animal matter (Ankney and Alisauskas 1991). The diet of mallards and blue-winged teal I collected that had begun various stages of RFD, varied considerably from the previously mentioned diet studies conducted on traditional breeding areas (e.g., prairie pothole region) during the breeding season. My sample size of follicle developing blue-winged teal was small (e.g., 9), thus, my inferences are limited; my estimate of the proportion of invertebrates consumed by blue-winged teal in RFD (58% invertebrates) was similar to estimates reported by Dirschl (1969), but considerably lower than the 95% reported by Swanson et al. (1974). Similarly, I documented mallards in RFD only consumed 27.7% invertebrates in comparison to the 72% previously reported by Swanson et al. (1985). The difference between my study and previous studies is likely a difference in nutrient demand associated with stage of RFD of ducks breeding on my study sites or a difference in food availability. Because I collected females only at the beginning of the nesting season, the females I collected may have been in an earlier stage of egg development, thus having a lower demand for protein than females from other studies. Alternatively, Elmberg et al. (2000) discovered a relationship between invertebrate abundance and duck-use during the breeding season in Sweden, suggesting the importance of wetlands with high invertebrate densities during RFD. Most wetlands in the breeding range of ducks (e.g., prairie pothole and parkland habitat) are seasonal in nature (e.g., have standing water only through midsummer). Neckles et al. (2006) found seasonal wetlands in Manitoba supported the highest densities of invertebrates. It is likely, therefore, that ducks in my study areas may have had fewer invertebrates available (e.g., compared to breeding ducks in prairie and aspen wetlands) because most wetlands encountered in the UMR/GLR exhibit a longer hydroperiod than seasonal wetlands, thus ducks were supplementing their diet with seeds that are high in carbohydrates, but also provide important amino acids. In contrast to diets during RFD, diet breadth increases during fall and winter, as most species consume large amounts of carbohydrates (Delnicki and Reinecke 1986, Thompson et al. 1992) and some species consume large amounts of invertebrates (Rogers and Korschgen 1966). Typically, animal matter comprises only a small portion (< 5%) of a ducks diet during winter (Paulus 1982, Delnicki and Reinecke 1986, Thompson et al. 1992, Peters and Afton 1993). This is likely a consequence of the need for high-energy foods (e.g., seeds and agricultural grains) to fuel migration and endure sub-freezing temperatures. Greater variation exists however during this time as compared to breeding diets, as algae and green vegetation are the main components of gadwall diets (Paulus 1982) during winter. Likewise, Rogers and Korschgen (1966) reported scaup diets during winter to consist of > 50% animal matter. Despite the wealth of knowledge regarding wintering ducks, little is known about duck diets during spring. Existing literature indicates spring diet is similar to winter diet for some species (e.g., mallards), yet very different for others (e.g., blue-winged teal). I found blue-winged teal consumed ~ 35% invertebrates at southern study sites (e.g., the Cache and Scioto River) and ~ 40% – 60% at northern study sites (e.g., Wisconsin and Saginaw Bay). My estimates of invertebrate consumption by blue-winged teal were similar to previous spring diet studies at southern latitudes (e.g., Louisiana) but lower than those reported in North Dakota (Manley et al. 1992, Swanson et al. 1974). The only study, however, that examined blue-winged teal diet at similar latitudes as my study was conducted in Missouri and reported teal consumed 70% animal matter (Taylor 1978). This study evaluated only 10 teal, however, which were collected in seasonally flooded wetlands (i.e., the same habitat type). My estimates of mallard diet during spring are similar to previous studies (e.g., < 1% animal matter in Nebraska and 12% − 24% in Illinois) and indicate a heavy reliance on seeds and agricultural grains (Jorde 1981, Smith 2007). Mallard diets in my study ranged from 7% − 27% invertebrates, with mean percentages being the least at southern sites (e.g., 9% at Cache River and 7% at Scioto River) and highest at northern sites (e.g., 23.7% at Wisconsin and 27% at Saginaw Bay). Because of the lack of spring diet studies of gadwall, I was unable to contrast my results. My data indicated that gadwall diet during spring in the UMR/GLR was composed of approximately 53% vegetation, 38% seeds, and 9% invertebrates. I did not, however, anticipate gadwall to consume large amounts of invertebrates because of their herbivorous food-habits (Ringelman 1990). I observed a noticeable difference in the spring diet of the diving ducks (e.g., scaup and ring-necked ducks). At southern and mid-latitude study sites, I detected scaup consuming similar proportions of invertebrates and seeds. At northern sites, however, I detected differing trends, with scaup at Saginaw Bay consuming ~ 80% invertebrates and scaup at Wisconsin consuming ~ 30% invertebrates. Previous spring diet studies of scaup indicated a heavier reliance on invertebrates than I observed (Gammonley and Heitmeyer 1990, Anteau and Afton 2006, Badzinski and Petrie 2006a). I suggest these differences may be related to habitats in which the scaup were collected (i.e., differences in availability) and differences in latitude of collected ducks (i.e., I collected scaup at more southerly latitudes than previous spring studies). Anteau and Afton (2008) only collected scaup in semi-permanent and permanent wetlands, whereas scaup in this study were collected from all wetland types in which they were regularly observed foraging in (i.e., often seasonal emergent wetlands). More recently, spring diet studies of scaup at southerly latitudes produced similar results, indicating there may be more reliance on plant foods than historically perceived (Smith 2007, Strand et al. 2008). No diet data existed on spring-migrating ring-necked ducks but Hohman’s (1985) results from pre-laying females in Minnesota may provide an indication of what to expect. He found pre-laying females consumed 36% animal matter (Hohman 1985). This estimate was higher than my estimate of ring-necked duck invertebrate consumption at all staging sites in the UMR/GLR except for Saginaw Bay, where they consumed 60% invertebrates. Given that none of the ring-necked ducks collected at Saginaw Bay had begun RFD and that their diet was similar to female ring-necked ducks that had initiated egg-laying in Minnesota (Hohman 1985), I speculate invertebrate consumption at Saginaw Bay was a function of availability. Temporal and Latitudinal Variation in Spring Diet Within a Species Some waterfowl rely on endogenous lipid to produce a clutch of eggs, thus as spring progressed, I expected to see an increase in consumption of invertebrates to supplement stored lipids. I detected this pattern in blue-winged teal, and to a lesser extent in mallards and ring-necked ducks, and found some evidence that this increase in invertebrate consumption was influenced by (1) latitude (i.e., diet shift as they moved further north during the spring) and (2) date (i.e., diet shift as spring progressed). Because blue-winged teal have a late and protracted spring migration when compared with other species in this study, I was able to describe this temporal pattern while controlling for latitude of collection site; whereas these 2 factors (i.e. latitude and time) were confounded with the other species in this study. For example, my data included teal that were collected in late April at both southern and northern study sites, whereas mallards had departed southern study sites by late April, thus were collected only at northern study sites in late April. No significant differences in invertebrate consumption existed among study-sites for blue-winged teal after controlling for date (Figure 1.4). However, there was a significant increase in invertebrate consumption with time, indicating that diet of blue-winged teal during spring migration is likely more influenced by date rather than latitude. It is possible that this is a consequence of a nutritional demand for more protein as breeding neared or that there were simply more invertebrates available as spring progressed; the latter of which is likely not the case (e.g., see Chapter 2). There was a temporal trend among transects, with mallards and ring-necked ducks increasing invertebrate consumption throughout the spring; however, this trend was not apparent when controlling for study site in which they were collected (e.g., a non-significant effect of date when evaluating diets at the level of study site). This temporal trend was observed in ring-necked ducks collected in the eastern transect and in mallards collected in both the western and eastern transects. These results indicate that mallards and ring-necked ducks appear to be transitioning to diets consisting of more invertebrates as spring progressed. Diet varied little by latitude among dabbling ducks in my study. Specifically, my data indicated no significant differences in diet composition between study sites for mallards (Figure 1.6) and gadwall (Figure 1.8) when controlling for collection date. Blue-winged teal collected at Saginaw Bay consumed less seeds than blue-winged teal collected in Wisconsin, but invertebrate consumption did not differ among study sites (Figure 1.4). These results suggested that latitude did not strongly influence the diet of the dabbling ducks collected in my study. In contrast to dabbling ducks, some study-site variation existed among diving ducks in my study. These differences, however, occurred with no latitudinal pattern and therefore may exist solely because of differences in availability at these sites. Scaup consumed significantly more invertebrates and fewer seeds at Saginaw Bay than scaup at Lake Erie, Illinois River, and Wisconsin (Figure 1.10). Ring-necked ducks at Saginaw Bay consumed significantly more invertebrates and fewer seeds than ring-necked ducks at all other sites. I found availability of both invertebrates and seeds to be scarce to diving ducks at Saginaw Bay, whereas seeds were more available to diving ducks at Lake Erie and the Illinois River (see Chapter 2, Table 2.1). This may explain why fewer seeds were consumed by scaup and ring-necked ducks at Saginaw Bay. Additionally, ring-necked ducks consumed significantly more invertebrates and less seeds at the Scioto River than the Wisconsin site (Figure 1.12). Previous work shows photoperiod is one of the most important factors influencing migration patterns (Beason 1978). Because latitude of migration appears to have little impact on diet during spring, yet I found date to influence diet of several species, I speculate that photoperiod (e.g., indirectly date) plays a larger role in diet habits.
Analyses indicated longitudinal differences in diet for all species in my study except mallards (Figure 1.7). The similarity in the diet of female mallards among transects (i.e., high use of carbohydrate and low use of high-protein foods) likely reflects their preference for high-carbohydrate foods for migration and their efficiency at acquiring their preferred food. Female blue-winged teal (Figure 1.5), scaup (Figure 1.11), and ring-necked ducks (Figure 1.13) consumed a higher percentage of seeds and fewer invertebrates in the western transect than the eastern transect. Invertebrate consumption did not differ among transects for gadwall, but more seeds and less vegetation was consumed by female gadwalls collected in the western transect than those collected in the eastern transect (Figure 1.9). Possible explanations for the differences in diet among transects in blue-winged teal, scaup, and ring-necked ducks include (1) ducks collected on the eastern transect may nest in a different region than ducks collected on the western transect, resulting in different foraging strategies dependant on the distance they are from their breeding areas and the quality of wetlands at respective breeding areas; (2) differences in wintering populations, resulting in body conditions that were dependant on the quality of wetlands in respective wintering areas, and; (3) resource availability (see Chapter 2). Most blue-winged teal winter in Central and South America and it is possible that teal collected in the western transect wintered along the Gulf Coast of North America and Central America, whereas teal collected on the eastern transect wintered in South America (Bellrose 1980). Badzinski and Petrie (2006b) found scaup that wintered in the Atlantic flyway used parts of the lower Great Lakes during spring migration. Scaup collected in the western transect, therefore, likely wintered in Central America or along the Gulf Coast of Louisiana and Mississippi, whereas scaup collected on the eastern transect likely wintered in Florida or along the Atlantic Coast (Bellrose 1980). Ring-necked ducks staging along the western and eastern transects in the UMR/GLR during spring probably wintered in similar areas as scaup (Bellrose 1980). Lastly, according to the migration corridors provided by Bellrose (1980), it appears that the gadwall encountered in the eastern transect spent winters on the east coast of North America, whereas the gadwall encountered in the western transect likely wintered in Louisiana and Mississippi. Diet of scaup and ring-necked ducks that winter in the Atlantic and Mississippi Flyways may differ, possibly explaining some of the discrepancy I observed in diet during spring among the western and eastern transects and possibly supporting the notion that I collected different wintering populations of birds in different transects. In support of this idea, I found the variation in diet was slightly higher in ring-necked ducks and scaup collected in the eastern transect than those in the western transect (avg. SE in western transect = 0.075 and avg. SE in eastern transect = 0.086). Scaup staging in coastal South Carolina (i.e., Atlantic Flyway) during winter consumed < 1% animal matter (Kerwin and Webb 1972), whereas scaup staging in coastal Louisiana (i.e., Mississippi flyway) during winter consumed 63% animal matter (Rogers and Korschgen 1966). Conversely, ring-necked ducks staging in Louisiana during winter consumed < 1% animal matter (Peters and Afton 1993), whereas ring-necked ducks staging in South Carolina during winter consumed 58% animal matter (Hoppe et al. 1986). Diet data for gadwall during the winter was only available for the Mississippi Flyway (Paulus 1982, McKnight and Hepp 1998) as no diet data exists for gadwall wintering along the Atlantic Flyway. The lack of gadwall diet studies on the wintering grounds in the Atlantic Flyway makes it difficult to hypothesize if I may have encountered different wintering populations of gadwall during the spring in the UMR/GLR based on their diets alone. Likewise, winter diet data for blue-winged teal is sparse and has only been collected in Central America (Thompson et al. 1992). Lastly, if distance to breeding area or wetland productivity at respective breeding areas differed among transects, I may have detected a difference in diet because ducks that were closer to breeding areas may have consumed more invertebrates to increase protein reserves. It has been suggested that ducks nesting in different areas may differ in their reliance on endogenous reserves as a consequence of wetland productivity at breeding areas (Young 1993). For example, migratory mallards breed in both aspen parkland and prairie habitats and the invertebrate food base in prairie wetlands are more likely to exhibit annual variation than aspen parkland wetlands. Existing literature, however, does not allow me to make any predictions regarding where the ducks I am encountering in each transect may breed. In any case, my data indicates that latitude has no affect on diet of spring migrating ducks, so it is likely that distance to breeding areas has little influence on diet. Although neither of the previously hypothesized explanations can be excluded, I believe the most likely explanation for longitudinal differences in diet was simply a result of differences in availability, as availability varied among the western and eastern transects in 2006 (J. Straub, The Ohio State University, thesis in progress, R. Schultheis, Southern Illinois University, dissertation in progress). Seed abundance was considerably higher than invertebrate abundance at western transect study sites, whereas seed and invertebrate abundances were more similar at eastern transect study sites. Yüklə 0,59 Mb. Dostları ilə paylaş:
|
