Northern Prairie Wildlife Research Center
During the study, birds were counted on 153 fields totaling 4,570 ha (range 6 - 96, = 30 ha). Horned larks (Eremophila alpestris) were the most frequently observed species in all seasons, but the most common species was red-winged blackbirds seen mainly in fall (Table 1). The frequency and number of other species varied considerably among seasons.
In spring, we counted 18 species and 4,184 birds on 86 fields. Migrating chestnut-collared longspurs and lapland longspurs (Calcarius lapponicus) were the second and third most frequently encountered species, but they were the most numerous because they occurred in large flocks. No other species exceeded 10% of the sightings during the spring.
In summer, we counted 70 bird species and 9,440 birds on 153 fields. The vesper sparrow (Pooecetes gramineus) was the second most frequently observed species after horned lark. However, vesper sparrows were less numerous than yellow-headed blackbirds (Xanthocephalus xanthocephalus), red-winged blackbirds, brown-headed cowbirds, and lapland longspurs, which typically occur in large flocks.
There were insufficient data to test differences in the number of species among crops, field types, and years for spring, fall, and winter, but there were sufficient data for summer counts. The rate of increase between the number of species and the natural logarithm of the size of the field was consistent among all combinations of years, crops, and field types (F = 0.71; 17,93 df; P = 0.781). Thus, the regression lines were parallel and differences among the treatments would be the same at any value of covariate. The 2-way and 3-way interactions were not significant (P > 0.05), which makes differences in the main effects interpretable. No differences in the mean number of species were found among field types (F = 0.19; 2,110 df; P = 0.827) or between years (F = 2.89; 1,110 df; P = 0.092). However, there were differences among the crops (F = 11.09; 2,110 df; P < 0.001). The LSD tests indicated that the number of species ( = 7.3, SE = 0.472) on fallow fields was larger (P < 0.001) than the number of species on sunflower fields ( = 4.6, SE = 0.400) and wheat fields (= 4.8, SE = 0.340). No difference was found between sunflower and wheat fields (P = 0.726).
During the 2 fall surveys, we observed 40 species totaling 12,169 birds on 88 fields. Vesper sparrows were again second to horned larks in frequency, but both species were less numerous than red-winged and yellow-headed blackbirds, which composed 63% of all birds observed in fall. Eighty-three percent of the blackbirds seen in fall were observed in conventional sunflower fields.
In winter, we observed 11 species and 2,651 birds on 85 fields. The most common species were horned larks, snow buntings (Plectrophenax nivalis), and lapland longspurs. These 3 species commonly remained on our fields until snow depths exceeded 15-25 cm. Also present in the fields in the winter were resident game birds, including sharp-tailed grouse (Tympanuchus phasianellus), gray partridge (Perdix perdix), and ring-necked pheasant (Phasianus colchicus). These 6 species included 96% of all birds seen in winter.
Bird Density by Season
Because there was a crop-field type interaction (F = 2.66; 4,71 df; P = 0.039) in the spring, we tested for crop differences within field types and field type differences within crops. Bird densities in minimum-tillage fallow were higher than those in organic fallow. Bird densities in minimum-tillage fallow fields were higher than densities in minimum-tillage sunflower or wheat fields (Table 2).
In summer, there was a significant (F = 2.71; 4,134 df; P = 0.033) year-crop interaction so differences among crops were interpreted within years. In 1992, fallow fields had higher bird densities than sunflower and wheat fields (Table 3). In 1993, fallow and wheat fields had higher bird densities than sunflower fields. Bird densities in fallow fields were enhanced mainly by breeding birds, not transient migrants.
In fall, there were no differences among field types (F = 0.32; 2,73 df; P = 0.725) or among crops (F = 2.66; 2,73 df; P = 0.077). Sunflower fields were attractive to birds, mainly blackbirds, but counts were variable and no differences were detected (Table 2).
Similarly, in winter there was large variation in bird counts and no effects due to crops (F = 0.77; 2,70 df; P = 0.469) or field types (F = 1.06; 2,70 df; P = 0.353) were detected.
Nesting Species and Nesting Period
Nineteen species nested in the crop fields, and brown-headed cowbirds (Molothrus ater) laid parasitic eggs in 39% of the passerine nests. Horned lark, vesper sparrow, and killdeer (Charadrius vociferus) initiated 66% of all nests located (Table 4). Passerines initiated 75% of all nests, shorebirds 17%, waterfowl 5%, doves 3%, and upland game 1%.
Nesting started in mid-April and extended into early August (Fig. 1). The peak period for active nests was late May and early June. Most wheat was seeded by late May, most sunflowers were planted by mid-June, but the cultivating and spraying of fallow fields occurred during the entire period.
Figure 1: Chronology of active nests and the length of the nesting period of the primary species compared to the normal planting period for wheat and sunflowers and the normal fallow tillage dates in southeastern North Dakota, 1991-93.
We compared the number of nesting species per field for 1992 and 1993. The rate of increase between the number of species and the natural logarithm of the size of the field was consistent for all combinations of years, crops, and field types (F = 0.90; 17,93 df; P = 0.578). Therefore, the regression lines were parallel and the differences in the years, crops, and field types were tested at the average value of the covariate. The 2-way and 3-way interactions were not significant (P > 0.05), making differences in the main effects interpretable. There was a difference between years (F = 4.07; 1,110 df; P = 0.046). The number of species in 1993 ( = 1.69, SE = 0.168) was larger than the number of species in 1992 ( = 1.24, SE = 0.150). There were differences (F = 7.01; 1,110 df; P = 0.001) in the number of species among field types (Table 5). Minimum-tillage and organic fields had a wider variety of nesting species, including upland sandpipers (Bartramia longicauda), red-winged blackbirds, grasshopper sparrows (Ammodramus savannarum), and lark buntings (Calamospiza melanocorys) than conventional fields. Conventional fields were used mainly by the killdeer, horned lark, and vesper sparrow. There was no difference in the number of nesting species between minimum-tillage and organic fields.
There was a marginal difference (F = 2.83; 2,110 df; P = 0.063) in the number of species per field among crops. Fallow fields had significantly more species than sunflower fields, but no differences were detected between fallow fields and wheat fields or between sunflower fields and wheat fields. Overall, 16 species nested in fallow fields compared to 12 for wheat fields, and 10 for sunflower fields.
There were no significant 2-way and 3-way interactions between year, field type, and crop when comparing nest densities. Thus, differences in the years, crops, and field types could be interpreted. We did find significant differences in nest densities among field types (F = 4.01; 2,134 df; P = 0.020) and among crops (F = 3.07; 2,134 df; P = 0.049). Among field types, minimum-tillage and organic fields had significantly more nests than conventional fields (Table 6). There was no difference between minimum-tillage and organic fields. Fallow fields had more nests than sunflower fields and wheat fields, but there was no difference between sunflower and wheat fields.
Hatching Success and Fledging Success
We found no significant interactions for passerine nest loss between field type and crops ( = 3.91, 4 df, P = 0.419). Also, there were no differences in DSR among field types ( = 2.26, 2 df, P = 0.324) or crops ( = 2.50, 2 df, P = 0.287; Table 7). For passerine nestling survival, we found no interaction between field type and crop ( = 4.44, 4 df, P = 0.349). Also, there were no differences among the crops ( = 1.99, 2 df, P = 0.369) or among the field types ( = 1.10, 2 df, P = 0.576). Although hatching success of passerine nests was higher than that of shorebirds, nest success for passerine nests was low: 13% for conventional fields, 18% for minimum-tillage fields, 11% for organic fields, and 14% overall. For shorebird nests there was no field type-crop interaction ( = 4.55, 4 df, P = 0.337), no differences among the crops ( = 2.58, 2 df, P = 0.275), or among field types ( = 2.81, 2 df, P = 0.245).
Passerine clutches that were unsuccessful (n = 173) were lost to predators (62%), farming activities (22%), abandonment (9%), and other (7%). Losses of shorebird nests (n = 47) were attributed to predators (47%), farming activities (41%), abandonment (9%), and other (3%). Unsuccessful waterfowl nests (n = 16) were destroyed by a combination of predators (50%), farming activities (44%), and abandonment (6%).
We also examined the DSR of passerine eggs when losses were due to factors other than predators, mainly farming activities and weather. There was no interaction between crop and field type ( = 3.31, 4 df, P = 0.508) and no differences among field types ( = 2.13, 2 df, P = 0.344). However, there were differences among the crops ( = 11.81, 2 df, P = 0.003). The DSR (0.985, SE = 0.004) for wheat fields was significantly larger than the DSR (0.962, SE = 0.008) for fallow fields (P = 0.014) and the DSR (0.939, SE = 0.017) for sunflower fields (P = 0.008). No differences were found between fallow fields and sunflower fields (P = 0.221).
For shorebird egg losses due to factors other than predators, there was no field type-crop interaction ( = 1.93, 4 df, P = 0.758) and no difference among crops ( = 3.50, 2 df, P = 0.174). However, there were differences among field types ( = 10.24, 2 df, P = 0.006). The DSR (0.9947, SE = 0.005) for minimum-tillage fields was significantly larger than the DSR (0.9516, SE = 0.014) for organic fields (P = 0.002). There was no difference between the DSR (0.9744, SE = 0.015) for conventional fields and minimum-tillage fields (P = 0.188) or between conventional fields and organic fields (P = 0.242).
Bird and Nest Density Associations with Vegetative Cover
Plant cover and litter cover were used to compare the relation between vegetation in the fields and bird populations and nest density. Plant height was not used because it was highly correlated with plant cover (r = 0.893), and visual obstruction values were not used because they were correlated with litter cover (r = 0.436).
In summer, no relation was found between percent plant cover and species of birds that prefer dense cover (r = -0.072, P = 0.378, n = 151) or species that prefer sparse cover (r = -0.061, P = 0.456, n = 151). Also, we found no significant correlations between litter cover and species that prefer sparse cover (r = 0.035, P = 0.668, n = 151) but there was a significant association between species that prefer dense cover and percent litter cover (r = 0.304, P = 0.001, n = 151).
There also were significant correlations between percent litter cover and the nesting densities of species that prefer dense cover (r = 0.231, P = 0.004, n = 151) and the nesting densities of species that prefer sparse cover (r = 0.162, P = 0.047, n = 151). However, the correlation coefficients (r) for the associations were small, suggesting that other factors affect this relation. There were no correlations between nest densities and plant cover (P > 0.05).
Plant cover did not differ among field types (F = 2.39; 2,134 df; P = 0.095), but litter cover did vary (F = 7.55; 2,134 df; P = 0.008). Minimum-tillage fields had a higher (P < 0.001) litter cover ( = 39.216, SE = 4.678) than conventional fields ( = 22.125, SE = 3.373) or organic fields ( = 20.732, SE = 2.918).
There were differences between crops for both plant cover and litter cover, but tests were made within years due to the year-crop interaction. In 1991, percent plant cover for wheat (= 41.842, SE = 3.359) was significantly higher (P < 0.001) than sunflower (= 16.777, SE = 5.539). In 1992, percent plant cover for wheat (= 51.847, SE = 2.487) was higher (P < 0.001) than fallow (= 26.181, SE = 2.867) and (P = 0.022) sunflower (= 16.495, SE = 3.061).
In 1992, percent litter cover in fallow (= 43.930, SE = 3.861) was higher (P < 0.001) than litter cover in sunflower (= 24.024, SE = 4.121) and wheat (= 12.614, SE = 3.348). In 1993, percent litter cover for fallow (= 38.400, SE = 4.565) was marginally higher (P = 0.063) than wheat (= 27.892, SE = 3.292), and percent litter cover for sunflower (= 38.193, SE = 4.040) was significantly larger (P = 0.050) than wheat.
The number of tillages averaged 1.1/year for minimum-tillage fields compared to 2.8/year for conventional fields and 4.0/year for organic fields. Overall, the fewest tillage treatments were applied to fallow fields (0.5/yr for minimum-tillage, 2.3/yr for conventional, and 3.6/yr for organic) and the most to sunflower fields (2.0/yr for minimum-tillage, 3.8/yr for conventional, and 5.4/yr for organic). The mean number of tillage treatments was not related to DSR (r = -0.423, P = 0.257, n = 9), but it was related to nest density (r = -0.176, P = 0.045, n = 129).
We found a negative correlation (r = -0.292, P = 0.041, n = 49) between tillage treatments and nest density for organic fields, but a non-significant correlation for conventional (r = -0.022, P = 0.889, n = 24) or minimum-tillage fields (r = -0.179, P = 0.296, n = 36). For crops there was a similar negative relation (r = -0.319, P = 0.050, n = 38) between tillage treatments and nest density for sunflowers. However, there was a non-significant relation between tillage treatments and nest density for fallow fields (r = -0.127, P = 0.446, n = 38) and wheat fields (r = -0.101, P = 0.472, n = 53).