Subhumid pasture plant communities entrained by management Julie E. Woodis *, Randall D. Jackson University of Wisconsin-Madison, Department of Agronomy, 1575 Linden Drive, Madison, WI 53706, USA 1. Introduction In many temperate areas of the USA, species-poor annual cropping systems are being converted to more diverse perennial grasslands for managed grazing (Hanson et al., 1998; Ostrom and Jackson-Smith, 2000; Foltz and Lang, 2005). Historically, farmers were encouraged to plant these pastures to monocultures or binary plant mixes (Sanderson et al., 2004), but recent work suggests that a higher number of species in grasslands may benefit both production and the environment (Hector et al., 1999; Minns et al., 2001; Tilman et al., 2001; Moore et al., 2004; Tracy and Sanderson, 2004). Generally speaking, grassland plant diversity decreases as management inputs increase (Tallowin et al., 2005), but land managers still lack specific information on how to manage species- rich plant communities. Cool-season (C3) pastures with two or more species have been characterized as difficult to sustain (Tracy and Faulkner, 2006), manage, and as inherently unpredictable (Frame and Newbould, 1986; Kessler and Nosberger, 1994). Other work indicates that our ability to predict vegetation change under grazing is limited because plant species frequently vary in their response to grazing (Vesk and Westoby, 2001; Pakeman, 2004). The complexity of grassland communities may not allow high levels of control given our limited ability to scale population-level responses to the plant community—the scale where pasture management is most relevant. In a review by Chapman et al. (2007), the authors stated: ‘‘Our ability to manipulate these pasture types to achieve increasingly challenging productivity and environmental targets depends on gaining further understanding of the interactions between grazing behavior and intake, feed supply and demand balance, pasture composition, and grazing method, against a backdrop of spatial and temporal variability introduced from a range of sources’’ (emphasis added). To what extent does the backdrop of spatial and temporal variability matter? Do the effects of management outweigh this variability, and if so, by how much? Little evidence exists about the proportional influence of management and the environment on community composition in these relatively resource-rich pasture ecosystems. Understanding the extent to which management Agriculture, Ecosystems and Environment 129 (2009) 83–90 ARTICLE INFO Article history: Received 7 January 2008 Received in revised form 7 July 2008 Accepted 15 July 2008 Available online 22 August 2008 Keywords: Burning Carbon addition Grazing Nitrogen addition Nonmetric multidimensional scaling (NMS) Ordination Variance partitioning ABSTRACT Many annual cropping lands are being converted to perennial pasture in humid and subhumid areas of the USA as the economic, social, and environmental benefits of managed grazing become evident. Recent evidence about the positive effects of plant diversity on productivity is driving farmers towards management for more complex species mixes than were promoted historically for grazed pastures, but this emphasis has highlighted a lack of understanding about plant community management in pastures. To better understand this system we examined sources of variability in the plant community of a subhumid grassland that was subjected to combinations of disturbance (graze or burn), and amendment (nitrogen, carbon, or ambient), and native grass seeding treatments over 3 years (2004 to 2006). We estimated species cover five times during the study period and ordinated these data with nonmetric multidimensional scaling (NMS). The ordination explained 80% of the variation in the plant community data set. Variance in ordination site scores was ascribed to management (disturbance, soil amendment, and native grass seeding time), the environment (year, season within year, and block), or unexplained (residual) realms with a linear mixed-effects model. The overwhelming source of variability in the plant community (62–99%) was attributed to management. Ordination results illustrated that plant groups sorted out along management gradients: burned plots were shifted towards grasses while forbs and clovers were associated with grazed plots. We demonstrate how a community-level approach can be used to evaluate management regimes in agroecosystems. Our results indicate that disturbance and soil amendments can be used to shape plant communities in subhumid grasslands for specific agronomic and conservation goals. ß 2008 Elsevier B.V. All rights reserved. * Corresponding author. Tel.: +1 608 890 0268. E-mail address: jedoll@wisc.edu (J.E. Woodis). Contents lists available at ScienceDirect Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee 0167-8809/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.agee.2008.07.006