Landscape effects on reproduction of Euschistus servus (Hemiptera: Pentatomidae), a mobile, polyphagous, multivoltine arthropod herbivore

Authors: Olson, Dawn; PRESCOTT, KRISTINA - University Of Minnesota; ZEILINGER, ADAM - Dominican University Of California; HOU, SUQIN - Harvard School Of Public Health; Coffin, Alisa; Smith, Coby; RUBERSON, JOHN - Kansas State University; ANDOW, DAVID - University Of Minnesota

Submitted to: Environmental Entomology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/15/2018
Publication Date: 3/15/2018
Citation: Olson, D.M., Prescott, K.K., Zeilinger, A.R., Hou, S., Coffin, A.W., Smith, C.M., Ruberson, J.R., Andow, D.A. 2018. Landscape effects on reproduction of Euschistus servus (Hemiptera: Pentatomidae), a mobile, polyphagous, multivoltine arthropod herbivore. Environmental Entomology. 47(3):660-668. https://doi.org/10.1093/ee/nvy045.
DOI: https://doi.org/10.1093/ee/nvy045

Interpretive Summary: Community structure, species abundance, and biotic interactions of invertebrate species in farmlands are influenced by processes at the regional and landscape levels. While previous work makes clear the importance of landscape factors for natural enemy populations, relatively less is known about their influence on arthropod pest populations. The economically important brown stink bug, Euschistus servus, is a mobile pest of many crops in southeastern USA. We estimated its reproduction in 192 fields in two regions on four major crop hosts—maize, peanut, cotton and soybean—over three years in southern Georgia. We used a geographic information system (GIS) to characterize the surrounding landscape structure (proportion of land-use types, distance between fields of different crop types). We identify subsets of local and landscape characteristics that account for variation in E. servus reproduction. Common measures of landscape complexity, such as % non-crop habitat (woodland, pasture and non-crop hosts), were not associated with brown stinkbug reproduction rate. However, a higher proportions of corn and peanut and lower proportions of cotton and soybean in the landscape, as well as shorter distances to corn and soybean fields were associated with higher reproduction. These results indicate a more complex pattern of factors controlling reproduction and distribution of the brown stinkbug than what has been found for other arthropod pests.

Technical Abstract: The brown stink bug is an economic pest of many crops in the southeastern US. In this study we characterized the complexity, land-use intensity and distances between differently-cropped fields on landscapes in two regions of southern Georgia, USA. We estimated brown stink bug reproduction, predator abundance and adult and nymph parasitism in the major crops maize, peanut, cotton and soybean in these landscapes, and associated landscape characteristics and natural enemy density with brown stink bug populations. We selected two regions (Southwest and East-Central) of the Coastal Plain of Georgia to sample stink bugs during the years 2009-2011. The regions were separated by approximately 150 km and differed in the area planted to maize, cotton and peanut with more area of these crops in the Southwest than the East-Central region. Each region contained two or three 4.8 x 4.8 km (2330 ha) landscapes and 3 corn, cotton, peanut, and soybean fields were sampled in each landscape. For each landscape within each region, we determined the proportion of the landscape that contained E. servus host crops (intensively produced maize, peanut, cotton and soybean) and non-host semi-natural habitat (woodland, pasture and non-crop hosts in the woodland edges), host cropland perimeter to area ratios (PA), and average distance to the nearest cropland in a radius of 1000 m from the focal sampling field. To obtain brown stink bug reproduction, we calculated the area under the colonist-incidence curve and the progeny-incidence curve, and estimated relative net population growth as the ratio of these two. All geo-spatial manipulations and analyses were conducted using ArcGIS software. There was a positive relationship between stink bug reproduction and the proportion of corn and peanut in the landscape. There was a negative relationship between stink bug reproduction and the proportion of cotton and soybean in the landscape and the mean distance from the sampled site to maize and soybean fields. The proportion of woodland and pasture and perimeter to area ratio of cropland had no effect on E. servus reproduction. Rather the composition of crop hosts within the landscape had a much greater influence on E. servus reproduction.