NPWRC :: Evidence for Edge Effects on the Multiple Levels in Tallgrass Prairie

**TABLE 1.** Prairie fragments in southwestern Missouri and the types of study done in each indicated by an "X".
Prairie fragment	Size (ha)	Artificial nests (1995)	Natural nests (1995-1997)	Mammals (1996-1997)
Pawhuska	31.0	X	X	X
Little Osage	32.0	X
Mon-Shon	32.4	X	X	X
Catlin	43.7		X	X
Gay Feather	47.0	X		X
Treaty Line	64.0		X
Clear Creek	77.7	X		X
Cook Memorial Meadow	113.0	X
Mo-No-I	122.0	X	X	X
Niawathe	130.0	X	X	X
Stony Point	259.0		X
Osage	506.0	X	X	X
Prairie State Park	1,084.0		X

**TABLE 2.** Tests of effects of distance to edge, fragment size, and vegetation structure on nesting success and its correlates. Presented are results from stepwise logistic regression or repeated-measures ANCOVA (animal track data only). In the logistic regressions, the dependent variables were log (ln) fragment size and three principal components describing vegetation structure at the nest. In repeated-measures ANCOVA, distance to edge (categorical) and ln fragment size were the explanatory variables.
Logistic regression	n	Variables entered	Slope ± SE^a		P
Artificial nest survival	193	PC1	-0.50 ± 0.12	18.5	<0.001
Nest success
Dickcissel^b	111	ln fragment size	-0.44 ± 0.004	8.5	<0.01
Dickcissel^b	111	PC1	-0.34 ± 0.13	6.5	0.01
Henslow's Sparrow^b	49	Distance to edge	-0.25 ± 0.11	5.6	<0.03
Brood parasitism^b	111	Distance to edge	1.14 ± 0.14	4.8	0.01
Depredation of artificial nests by all mammals	169	PC1	0.38 ± 0.12	9.6	<0.01
Depredation of artificial nests by all mammals	169	PC3	-0.31 ± 0.15	4.5	<0.05
mid-sized carnivores	169	Distance to edge	0.74 ± 0.37	4.1	<0.05
small mammals	169	PC1	0.40 ± 0.13	9.2	<0.01
small mammals	169	PC3	-0.41 ± 0.16	6.5	0.01

Repeated-measures ANCOVA	n	Variables in model	Slope ± SE	F	P
all mammals	48	Distance to edge^c		11.0	<0.001
all mammals	48	ln fragment size	-0.03 ± 0.01	5.8	0.05
mid-sized carnivores	48	Distance to edge^c		7.7	0.001
mid-sized carnivores	48	ln fragment size	0.001 ± 0.007	0.0	0.83
small mammals	48	ln fragment size	-0.01 ± 0.001	6.5	<0.01
small mammals	48	Distance to edge^c		5.3	0.06
^a Logistic regression calculated the probability that a nest fails or that a nest is parasitized. Thus a negative parameter indicates that the probability that a nest fails or gets parasitized decreases with increasing fragment size or vegetation variable. ^b The analysis includes nests in proximity to shrubby edge only. ^c Slopes are presented graphically in Figure 2.

**TABLE 3.** Percent of total Dickcissel and Henslow's Sparrow nests found at different edge types and within different distance intervals, estimates of daily Mayfield nest survival (±SE), and percent cowbird parasitism within three distance intervals from edge habitats in southwestern Missouri, 1995-1997.
Edge Type	Variable	Dickcissel (n^a = 225)			Henslow's Sparrow (n = 50)
Edge Type	Variable	≤ 50 m	50-100 m	>100 m	≤ 50 m	50-100 m	>100 m
Agricultural	Nests found (%)	0.04	0.09	1.7	0	0	2
	Nesting success	1.00 ± 0.00	0.95 ± 0.05	1.00 ± 0.00	--	--	1.00 ± 0.00
	Brood parasitism (%)	0	0	0	--	--	0
Forest	Nests found (%)	1.3	4	5.8	4	2	8
	Nesting success	0.96 ± 0.03	0.95 ± 0.03	0.93 ± 0.02	0.96 ± 0.04	--	0.96 ± 0.04
	Brood parasitism (%)	0.08	0	1.6	0	--	0
Road	Nests found (%)	12	8.4	12	0	4	8
	Nesting success	0.95 ± 0.01	0.90 ± 0.03	0.91 ± 0.02	1.00 ± 0.00	1.00 ± 0.00	1.00 ± 0.00
	Brood parasitism (%)	1.6	0.04	1	0	0	0
Shrubland	Nests found (%)	13.4	16.7	24.5	16	26	30
	Nesting success	0.93 ± 0.01	0.94 ± 0.02	0.95 ± 0.01	0.88 ± 0.04	0.95 ± 0.03	0.98 ± 0.01
	Brood parasitism (%)	7.5	2.7	2.4	1	1	1
^a Number of nests for which distance to edge was recorded.

**TABLE 4.** Effects of fragment size on artificial nests, natural nests, and nest predator distribution in southwestern Missouri prairie fragments.
Variable	Fragments used	Slope ± SE^a	R²	F	P
Artificial nest survival	9	0.20 ± 0.24	0.09	0.7	0.43
Nesting success:
Dickcissel	16^b	0.02 ± 0.01	0.22	4.3	0.05
Henslow's Sparrow	9	0.01 ± 0.02	0.07	0.6	0.45
Cowbird parasitism of Dickcissel nests	9	-0.06 ± 0.11	0.04	0.3	0.59
*Predators at artificial nests*
All mammals	9	-0.45 ± 0.20	0.38	4.9	0.06
Mid-sized carnivores	9	-0.12 ± 0.05	0.46	6.0	<0.05
Small mammals	9	-0.15 ± 0.15	0.13	1.0	0.35
Birds	9	-0.04 ± 0.06	0.06	0.5	0.51
Snakes	9	0.05 ± 0.09	0.03	0.3	0.60
*Visitors at track plates*
All mammals	8	-0.03 ± 0.01	0.56	7.6	0.03
Mid-sized carnivores	8	-0.001 ± 0.007	0.01	0.03	0.87
Small mammals	8	-0.03 ± 0.02	0.44	3.9	0.10
^a Positive slopes indicate an increase of nesting success, cowbird parasitism, or predator occurrence with fragment size. ^b n is larger than the number of prairie fragments used in the study, because data from several years were combined.


Home	About NPWRC	Our Science	Staff	Employment	Contacts	Common Questions	About the Site

Evidence for Edge Effects on Multiple Levels in Tallgrass Prairie