Elsevier

Environmental Modelling & Software

Volume 75, January 2016, Pages 388-401
Environmental Modelling & Software

Modeling the effect of social networks on adoption of multifunctional agriculture

https://doi.org/10.1016/j.envsoft.2014.09.015Get rights and content

Highlights

  • Social networks composed of weak ties and strong ties affect adoption of multifunctional agriculture.

  • Network effectiveness is contingent on social and spatial factors.

  • Programs targeting network structure and function could facilitate expansion of alternative systems.

Abstract

Rotational grazing (RG) has attracted much attention as a cornerstone of multifunctional agriculture (MFA) in animal systems, potentially capable of producing a range of goods and services of value to diverse stakeholders in agricultural landscapes and rural communities, as well as broader societal benefits. Despite these benefits, global adoption of MFA has been uneven, with some places seeing active participation, while others have seen limited growth. Recent conceptual models of MFA emphasize the potential for bottom-up processes and linkages among social and environmental systems to promote multifunctionality. Social networks are critical to these explanations but how and why these networks matter is unclear. We investigated fifty-three farms in three states in the United States (New York, Wisconsin, Pennsylvania) and developed a stylized model of social networks and systemic change in the dairy farming system. We found that social networks are important to RG adoption but their impact is contingent on social and spatial factors. Effects of networks on farmer decision making differ according to whether they comprise weak-tie relationships, which bridge across disparate people and organizations, or strong-tie relationships, which are shared by groups in which members are well known to one another. RG adoption is also dependent on features of the social landscape including the number of dairy households, the probability of neighboring farmers sharing strong ties, and the role of space in how networks are formed. The model replicates features of real-world adoption of RG practices in the Eastern US and illustrates pathways toward greater multifunctionality in the dairy landscape. Such models are likely to be of heuristic value in network-focused strategies for agricultural development.

Introduction

Multifunctional agriculture (MFA) promises benefits over many other forms of agriculture but its potential is unrealized in many places (Jordan and Warner, 2010). Agriculture is considered multifunctional when it produces standard commodities such as food or biomass as well as social and ecological outputs. These range from economic contributions such as ecotourism in agricultural landscapes to provision of ecological services including denitrification and protection of human systems from climate change impacts (Flora, 2001, Renting et al., 2009). Despite these benefits, MFA adoption has been very uneven, with some locales like New Zealand and several European countries seeing active participation and growth, while others including the United States have not (Jordan and Warner, 2010).

Why is MFA distribution uneven? It does well under top-down policies that explicitly support it as a form of production satisfying broader social and environmental goals (Potter and Burney, 2002, Uthes et al., 2011). Less clear is how MFA grows absent such support. As detailed below, some recent explanations emphasize ‘bottom-up’ processes and self-reinforcing feedbacks of systems at a variety of scales. These feedbacks rely on linkages among social and environmental systems that connect farmers, farmers to markets, organizations, and environments that benefit from MFA and subsequently support it (Jordan and Warner, 2010, Selman and Knight, 2006).

Social networks that help create linkages among social and environmental systems are critical to these explanations of MFA growth, but there are unanswered questions about how and why these networks matter. In this paper, we draw on an empirical case study to develop a stylized model of social networks and systemic change for a form of MFA called rotational grazing dairy farming (RG). We are particularly interested in differentiating between the effects on farmer decision making of weak-tie social relationships, which bridge across disparate people and organizations, and strong-tie relationships, which are shared by groups with closely integrated members. We also examine the effect of spatial distance on network formation and activation on RG practices within a broader context of confinement practices in dairy farming (CP). We draw on existing research and our own empirical work to develop an agent based model (ABM) that explores these networks and their spatiotemporality. This model of dairy farmers in the United States examines how social networks affect adoption of new forms of agriculture, and in turn how social and spatial factors influence these networks. The model is kept fairly abstract to examine general dynamics of these networks and the agricultural system.

This work makes several contributions to modeling and analysis of the role of social networks in land use change. The study of social networks in landscape dynamics and in agricultural extension is both important and relatively neglected (Beilin et al., 2013, Entwisle et al., 2008, Hellin, 2012), which has led to calls for research on the role of social networks in complex resource systems such as MFA (Agrawal et al., 2013, Bodin and Crona, 2009). This paper answers these calls by specifying possible impacts of different kinds and configurations of networks. This work also addresses the need to understand MFA dynamics at the farm scale (Renting et al., 2009), especially because the effectiveness of policy interventions must be measured and understood at the farm scale (Uthes et al., 2010). Finally, this work meets the specific need to create ABM to understand MFA (Van Berkel and Verburg, 2012) and addresses the more general paucity of tools to understand and design multifunctional landscapes (O'Farrell and Anderson, 2010, Rossing et al., 2007, Zander et al., 2007).

The remainder of this section describes multifunctional agriculture in the United States and the potential roles that social networks could play in the greater adoption of one kind of MFA, dairy production based on rotational grazing (RG). Section 2 describes methods centered on a stylized yet empirically based model of network dynamics and their role in the growth of RG in the northern United States. Section 3 offers results and discussion, focusing in particular on the impact on RG of network processes that were reported by farmers about their transition to RG production. Section 4 concludes with a recap of results and examines future research and policy directions.

MFA in the United States is less established and studied than elsewhere (Jordan and Warner, 2010). The top-down policies that support MFA in other contexts are largely absent and are likely to remain so given a policy focus on standard commodity production (Uthes et al., 2011). Other countries design policies to encourage MFA for broader social and environmental reasons, ranging from preservation of rural livelihoods to encouraging biodiversity (Baylis et al., 2008, Cocklin et al., 2006, Potter and Burney, 2002). In contrast, US agricultural programs are largely incidental to MFA. They focus instead on specific environmental issues such as minimizing soil erosion by taking land out of production, and economic concerns such as supporting commodity agriculture. There is concomitantly less known about the biophysical and social nature of MFA in the US (Boody et al., 2005, Polasky et al., 2011).

How might MFA arise in the absence of top-down policies? One promising route is for farmers and other stakeholders to endogenously develop networks that create self-reinforcing flows of information and resources about multifunctionality (Jordan and Warner, 2010, Selman and Knight, 2006). This perspective posits an important role for social networks that are composed of nodes (people or other actors) that are joined by ties (relationships that convey information or resources). Networks can serve as conduits for information and resources, exert social influence, and shape actions (Knoke et al., 2008, Wasserman and Faust, 1994). Social networks link farmers to a broader array of actors and human-environment systems (Ortiz-Miranda et al., 2010, Sonnino and Marsden, 2006, Steyaert et al., 2007). These systems can include farmers and their neighbors trading production strategies (Conley and Udry, 2001), farmers and consumers forging links with cultural, social, and ecological meaning such as increased landscape amenity or healthful food (Van Huylenbroeck et al., 2007), or environmental organizations and scientists sharing information about benefits of multifunctionality (Ploeg, 2003).

Research on the social context of natural resources management and agriculture points to critical distinctions among kinds of relationships in networks (Moore and Westley, 2011). Such network ties are of three types: bonding or strong ties, bridging or weak ties, and linking or vertical ties (Woolcock and Narayan, 2000). Strong ties are relationships within largely homogeneous groups such as families that provide benefits to members. Weak ties bridge nodes among heterogeneous groups across sociocultural or geographical gaps, creating flexible networks that can transmit new ideas. Vertical ties are a variant of weak ties that serve as conduits among nodes at one level with nodes at another level of organization, such as between farmers and state agencies.

Network effectiveness is contingent on its structure and dynamics. A combination of weak and strong ties may be necessary for propagating new practices like MFA, judging by the work on related forms of production like agroforestry and sustainable agriculture. Weak ties in particular may help information generation and alliance building (Nelson et al., 2014, Olsson et al., 2006), creating knowledge across sectors (Steyaert et al., 2007), and developing ties among disparate organizations (Ortiz-Miranda et al., 2010, Sonnino and Marsden, 2006). Networks can form and dissolve as a function of physical space, social distance, and happenstance in ways that affect their ability to convey information and resources (Carolan, 2005, Nelson et al., 2014, Olsson et al., 2006). The structure of networks can affect how they function in natural resource situations (Bodin and Crona, 2009), including how well the effects of networks at one spatial scale or level of organization are manifested at other scales, or across different sectors of social systems (Moore and Westley, 2011).

Rotational grazing is an excellent system in which to study the network dynamics of MFA. The majority of dairy farms exemplify standard commodity agriculture in their use of confinement practices (CP) while a minority use rotational grazing (RG). While both CP and RG dairy farms use a range of practices and methods (Lyon et al., 2011), in CP farms, cows spend much of their time in barns and eat processed feed. CP has what is commonly termed a standard productivist focus on maximizing production and efficiency (see for in depth treatment Wilson, 2008). In contrast, RG dairy farms move the grazing herd through a series of paddocks in which the livestock are held for one to seven days (Taylor and Foltz, 2006). RG centers on ‘good grass farming,’ which translates into minimizing costs while promoting a sustainable environment for healthy animals that create higher profits than conventional production methods (Gloy et al., 2002, Lyon et al., 2011). Grazing reduces input costs through having cows eat grass in pastures instead of prepared feed and via lower costs for manure disposal and veterinary care (Undersander et al., 2002). RG has biophysical benefits including lower soil-erosion rates than row cropping (Gantzer et al., 1990) and improved wildlife habitat (Vondracek et al., 2005). Additional motivations include perceived lifestyle and community benefits, such as RG offering greater labor flexibility and more readily maintaining family farms (Murphy, 1998, Parsons et al., 1998).

Rotational grazing is also an interesting system to study because despite offering advantages for farmers and society, it has not yet expanded beyond roughly 25% of dairy farms in a few areas in the northern US and is far less in most places. In Wisconsin, the number of RG farms grew rapidly during the 1990's, from approximately 7% of dairy farms in 1993 to its current plateau of 26% in 2005 (Taylor and Foltz, 2006) and after (Kloppenburg et al., 2012). In Northeast US, the rate of RG plateaus in the range of 8–22% depending on sub region (Winsten et al., 2010). In addition to being limited to a small proportion of dairy farms, most of the grazing farms we studied were also spatially limited, forming small ‘islands’ of grassland embedded in a larger matrix of various land uses.

We focused on household farms as the primary entry point into examining MFA. Household farmers are defined as those who rely primarily on the immediate family for labor and are not heavily capitalized, reflecting the fact that dairy has seen steady consolidation into small numbers of large industrial operations and decreasing numbers of household farms (Hart, 2003; Winsten et al., 2010). These farms are important to study because they blend the market orientation of commercial farms with social elements such as values that are linked to MFA practices and the life course, or the sequence of socially and culturally defined events and structures that define an individual (Taylor and Foltz, 2006). At the same time, they are most likely to experiment with labor-intensive practices, such as rotational grazing, outside of the mainstream focus on capital-intensive commercial production.

Section snippets

Methods

We develop an agent based model of social network dynamics in the growth of RG in the northern United States. There are other excellent approaches for modeling MFA, including econometrics, farm-scale simulations, and integrated socioeconomic simulations (Jongeneel et al., 2008, Renting et al., 2009). These methods offer advantages including long histories and integration with specific bodies of research, such as derivation of econometrics from economics or basing farm-scale simulations on in

Results

We used the model to explore the impact of a sequence of network processes on RG adoption. As described above, farmers were introduced to RG via strong-tie networks like family, relied on vertical-tie networks for subsidizing switching costs, then on weak ties to help consolidate. The goal was to explore how network attributes and related interactions might affect the prevalent dynamics in our study region, namely an initial phase of rapid adoption followed by a static plateau. Our primary

Discussion

The motivating question for this work is how can MFA, as exemplified by RG, grow in the absence of large scale, top-down support? One promising answer to this question lies in the potential for networks of farmers and other actors in the social system to host and generate ‘bottom-up’ processes that create self-sustaining growth of RG. Our work illustrates ways in which these social networks may be important to forming pathways toward MFA but their effectiveness is contingent on social and

Acknowledgments

This work is supported in part by the National Science Foundation grant BE/CNH: Understanding the Importance of Weak-Tie Networks in Complex Human–Environment Systems: Ecosocial Feedback in Multifunctional Agriculture (0709613), National Aeronautics and Space Administration New Investigator Program in Earth-Sun System Science (NNX06AE85G), the National Institutes of Health supported Minnesota Population Center (R24 HD041023), and the Resident Fellowship program of the Institute on the

References (81)

  • S.M. Manson

    Land use in the Southern Yucatan Peninsular Region of Mexico: scenarios of population and institutional change

    Comput. Environ. Urban Syst.

    (2006)
  • P.J. O'Farrell et al.

    Sustainable multifunctional landscapes: a review to implementation

    Curr. Opin. Environ. Sustain.

    (2010)
  • C. Potter et al.

    Agricultural multifunctionality in the WTO—legitimate non-trade concern or disguised protectionism?

    J. Rural Stud.

    (2002)
  • H. Renting et al.

    Exploring multifunctional agriculture. A review of conceptual approaches and prospects for an integrative transitional framework

    J. Environ. Manage.

    (2009)
  • W.A.H. Rossing et al.

    Integrative modelling approaches for analysis of impact of multifunctional agriculture: a review for France, Germany and The Netherlands

    Multifunct. Agric. Tools Methods Impact Assess. Valuat.

    (2007)
  • P. Steyaert et al.

    The role of knowledge and research in facilitating social learning among stakeholders in natural resources management in the French Atlantic coastal wetlands

    Environ. Sci. Policy

    (2007)
  • S. Uthes et al.

    Regional impacts of abolishing direct payments: an integrated analysis in four European regions

    Agric. Syst.

    (2011)
  • S. Uthes et al.

    Modeling a farm population to estimate on-farm compliance costs and environmental effects of a grassland extensification scheme at the regional scale

    Agric. Syst.

    (2010)
  • D. Valbuena et al.

    Effects of farmers' decisions on the landscape structure of a Dutch rural region: an agent-based approach

    Landsc. Urban Plan.

    (2010)
  • S.J. Walsh et al.

    Design of an agent-based model to examine population–environment interactions in Nang Rong District, Thailand

    Appl. Geogr.

    (2013)
  • G.A. Wilson

    From “weak” to “strong” multifunctionality: conceptualising farm-level multifunctional transitional pathways

    J. Rural. Stud.

    (2008)
  • J.R. Winsten et al.

    Trends in the Northeast dairy industry: large-scale modern confinement feeding and management-intensive grazing

    J. Dairy Sci.

    (2010)
  • P. Zander et al.

    Multifunctionality of agriculture: tools and methods for impact assessment and valuation

    Agric. Ecosyst. Environ.

    (2007)
  • R. Beilin et al.

    Transition landscapes and social networks: examining on-gound community resilience and its implications for policy settings in multiscalar systems

    Ecol. Soc.

    (2013)
  • G. Boody et al.

    Multifunctional agriculture in the United States

    Bioscience

    (2005)
  • M.S. Carolan

    Barriers to the adoption of sustainable agriculture on rented land: an examination of contesting social Fields

    Rural Sociol.

    (2005)
  • C. Cocklin et al.

    From market to multifunctionality? Land stewardship in Australia

    Geogr. J.

    (2006)
  • T. Conley et al.

    Social learning through networks: the adoption of new agricultural technologies in Ghana

    Am. J. Agric. Econ.

    (2001)
  • B. Entwisle et al.

    An agent-based model of household dynamics and land use change

    J. Land Use Sci.

    (2008)
  • T.P. Evans et al.

    An agent-based model of agricultural innovation, land-cover change and household inequality: the transition from swidden cultivation to rubber plantations in Laos PDR

    J. Land Use Sci.

    (2011)
  • C.B. Flora

    Shifting agroecosystems and communities

  • C.J. Gantzer et al.

    Estimating soil erosion after 100 years of cropping on Sanborn Field

    J. Soil Water Conserv.

    (1990)
  • M.S. Granovetter

    The strength of weak ties

    Am. J. Sociol

    (1973)
  • M.T. Hansen

    The search-transfer problem: the role of weak ties in sharing knowledge across organization subunits

    Admin. Sci. Q.

    (1999)
  • J.F. Hart

    The Changing Scale of American Agriculture

    (2003)
  • N. Hassanein et al.

    Where the grass grows again: knowledge exchange in the sustainable agriculture movement

    Rural Sociol.

    (1995)
  • J. Hellin

    Agricultural extension, collective action and innovation systems: lessons on network brokering from Peru and Mexico

    J. Agric. Educ. Ext.

    (2012)
  • W. Hunt et al.

    Agricultural extension: building capacity and resilience in rural industries and communities

    Rural. Soc.

    (2011)
  • M.E. Isaac et al.

    Transfer of knowledge on agroforestry management practices: the structure of farmer advice networks

    Ecol. Soc.

    (2007)
  • R. Ison et al.

    Insights into operationalizing communities of practice from SSM-based inquiry processes

    Syst. Pract. Action Res.

    (2012)
  • Cited by (67)

    • Agent-based modelling to simulate farmers’ sustainable decisions: Farmers’ interaction and resulting green consciousness evolution

      2022, Journal of Cleaner Production
      Citation Excerpt :

      Social network analysis (SNA) and ABM are both valuable tools to analyse human interactions in a given environmental and societal context. Manson et al. (2016) combined SNA and ABM to model farmer transition to rotational grazing production in the United States (US). In their approach each tie between agents represents a certain type of relationship according to a predefined definition.

    View all citing articles on Scopus

    Thematic Issue on Modelling Systemic Change.

    View full text