TY - JOUR

T1 - Arthropod but not bird predation in Ethiopian homegardens is higher in tree-poor than in tree-rich landscapes

AU - Lemessa, Debissa

AU - Hambäck, Peter A.

AU - Hylander, Kristoffer

N1 - The Swedish International DevelopmentCooperation Agency (SIDA) (contract number: SWE-2009-134) and Formas (contract number: 229-2009-991) provided fund support [to KH]. The funders hadno role in study design, data collection and analysis,decision to publish, or preparation of the manuscript.

PY - 2015/5/11

Y1 - 2015/5/11

N2 - Bird and arthropod predation is often associated with natural pest control in agricultural landscapes, but the rates of predation may vary with the amount of tree cover or other environmental factors. We examined bird and arthropod predation in three tree-rich and three tree-poor landscapes across southwestern Ethiopia. Within each landscape we selected three tree-rich and three tree-poor homegardens in which we recorded the number of tree species and tree stems within 100 x 100 m surrounding the central house. To estimate predation rates, we attached plasticine caterpillars on leaves of two coffee and two avocado shrubs in each homegarden, and recorded the number of attacked caterpillars for 7-9 consecutive weeks. The overall mean daily predation rate was 1.45% for birds and 1.60% for arthropods. The rates of arthropod predation varied among landscapes and were higher in tree-poor landscapes. There was no such difference for birds. Within landscapes, predation rates from birds and arthropods did not vary between tree-rich and tree-poor homegardens in either tree-rich or tree-poor landscapes. The most surprising result was the lack of response by birds to tree cover at either spatial scale. Our results suggest that in tree-poor landscapes there are still enough non-crop habitats to support predatory arthropods and birds to deliver strong top-down effect on crop pests.

AB - Bird and arthropod predation is often associated with natural pest control in agricultural landscapes, but the rates of predation may vary with the amount of tree cover or other environmental factors. We examined bird and arthropod predation in three tree-rich and three tree-poor landscapes across southwestern Ethiopia. Within each landscape we selected three tree-rich and three tree-poor homegardens in which we recorded the number of tree species and tree stems within 100 x 100 m surrounding the central house. To estimate predation rates, we attached plasticine caterpillars on leaves of two coffee and two avocado shrubs in each homegarden, and recorded the number of attacked caterpillars for 7-9 consecutive weeks. The overall mean daily predation rate was 1.45% for birds and 1.60% for arthropods. The rates of arthropod predation varied among landscapes and were higher in tree-poor landscapes. There was no such difference for birds. Within landscapes, predation rates from birds and arthropods did not vary between tree-rich and tree-poor homegardens in either tree-rich or tree-poor landscapes. The most surprising result was the lack of response by birds to tree cover at either spatial scale. Our results suggest that in tree-poor landscapes there are still enough non-crop habitats to support predatory arthropods and birds to deliver strong top-down effect on crop pests.

KW - Sustainability Science

KW - Gender and Diversity

KW - Biology

UR - http://www.scopus.com/inward/record.url?scp=84930640142&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0126639

DO - 10.1371/journal.pone.0126639

M3 - Journal articles

C2 - 25961306

AN - SCOPUS:84930640142

VL - 10

JO - PLoS ONE

JF - PLoS ONE

SN - 1932-6203

IS - 5

M1 - e0126639

ER -