Land Use Policy 51 (2016) 236–243
Contents lists available at ScienceDirect
Land Use Policy journal homepage: www.elsevier.com/locate/landusepol
Exploring agricultural advice networks, beneﬁcial management practices and water quality on the landscape: A geospatial social-ecological systems analysis Julia Baird a,∗ , Marilyne Jollineau a,b , Ryan Plummer a,c , Josh Valenti b a
Environmental Sustainability Research Centre, Brock University, St. Catharines Ontario L2S 3A1, Canada Department of Geography, Brock University, St. Catharines Ontario L2S 3A1, Canada c Stockholm Resilience Centre, Stockholm University, SE-106 91 Stockholm, Sweden b
a r t i c l e
i n f o
Article history: Received 2 September 2015 Received in revised form 2 November 2015 Accepted 14 November 2015 Keywords: Agricultural practices Social network analysis Surface water quality 3D geovisualization
a b s t r a c t Agricultural practices have been linked to detrimental effects on ecosystems, with water quality of particular concern. Research has been devoted to understanding uptake of beneﬁcial, or best, management practices (BMPs) in agriculture; however, sources of advice and subsequent effects on the landscape have not been elucidated. This study set out to understand (1) what sources of information agricultural producers rely on when making land-management decisions; (2) the characteristics of their advice networks; and (3) how the advice network linked spatially to water quality on the landscape. A watershed in Alberta was used as a case study and respondents identiﬁed that regional advisors were relied upon most often for advice and these advisors had the most inﬂuence on the adoption of BMPs. Results indicate that respondents with connections to regional actors implemented more BMPs that those without. Regional government actors had a greater effect than regional non-governmental actors. Local actors played a lesser role in advice networks related to BMP adoption. A 3D geovisualization was used to explore linkages among advisors, BMPs, and water quality. This technique may be useful for other scenarios and can contribute to policy development and enhanced practices. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction The intensiﬁcation and industrialization of agriculture has led to apprehension about the myriad of potential impacts on the environment, the need for continuous improvement of practices, and the overarching imperative of sustainability. Potentially serious detrimental effects on ecosystems have accompanied marked gains in productivity (Tilman, 1999; Tilman et al., 2002; Plummer et al., 2008). These effects are not restricted to large or intense operations. Small and moderate sized farms may also cause environmental damage, and in some instances, their practices may have a greater environmental impact due to lack of capital, absence of technologies, and less stringent reporting requirements (FitzGibbon et al., 2002). While soil erosion and sedimentation are longstanding effects associated with agriculture, the adverse effects from modern
∗ Corresponding author. E-mail address: [email protected]
(J. Baird). http://dx.doi.org/10.1016/j.landusepol.2015.11.017 0264-8377/© 2015 Elsevier Ltd. All rights reserved.
practices have grown in terms of diversity and extent (see Stoate et al., 2009; Gomiero et al., 2011 for examples of reviews). The effects on water quality from agricultural practices are pervasive, a growing issue of public concern, and a challenge to policy makers. Drawing upon data from water bodies surveyed in the National Water Quality Inventory, the United States Environmental Protection Agency (USEPA, 2002) reported that nonpoint source pollution from agriculture was the leading cause of degradation of water quality in rivers and lakes, the second greatest contributor to wetland impairments, and a major contributor to groundwater and estuary contamination. Within the Organisation for Economic Co-operation and Development countries, “. . . agriculture is often the main source of water pollution” and attempts to address it costs billions of dollars (OECD, 2012, p. 9). Pollution from agricultural activities is almost exclusively nonpoint source in nature (Weersink et al., 1998). Nonpoint sources of pollution are diffuse as opposed to originating from a single identiﬁable source. Nonpoint sources are also complicated. In speciﬁc reference to agriculture as a nonpoint source of water pollution, (1) the point of origin is typically ‘invisible’ or low in concentration
J. Baird et al. / Land Use Policy 51 (2016) 236–243
Fig. 1. Respondent advice network. White circle nodes represent respondents and grey and black nodes represent non-respondents (grey triangle = governmental regional actor; black triangle = non-governmental regional actor; square = local actor; diamond = other business).
and the paths by which pollution enters water systems are complex; (2) the impact from diffuse sources over a large area tend to have a cumulative impact; and (3) the impacts are inﬂuenced not only by farm activities on the land but also stochastic events and physical properties (Baird, 2012; OECD, 2012). Effectively addressing nonpoint sources of pollution is a substantive challenge. Agriculture is emblematic of the limitations associated with the regulatory approach due to the volume and diffuseness of sources, technical complexities and costs of monitoring, uncertain causal relationships and difﬁculties in assigning attribution, and cumulative effects (Weersink et al., 1998; Plummer et al., 2008). The effectiveness and efﬁciency of traditional regulations as a way to garner compliance regarding agriculture pollution has been critically questioned (Young and Karkosky, 2000; Pretty et al., 2001; Dowd et al., 2008; Plummer et al., 2008). Dowd et al. (2008) reviewed the agricultural nonpoint source pollution policy literature and categorized it into voluntary programs, command and control programs, and economic instruments. Beneﬁcial management practices (or ‘best management practices’ in the United States) (BMPs) can apply to all three categories and are integral to mediating environmental degradation effects from agricultural practices. The application of BMPs in the United States grew out of the Soil and Conservation District movement of the mid 1930s (see Ice, 2004 for a history). A BMP is generally considered “a farming method that minimizes risk to the environment without sacriﬁcing economic productivity” (Hillard et al., 2002 p. i). Agricultural BMPs to protect water quality encompass on-ﬁeld improvements, edge of ﬁeld sinks, and multiple BMPs at watershed scales (see USDA, 2013 for examples). While the characteristics of nonpoint source pollution make it challenging to document their effectiveness (Mulla et al., 2006), insights are being gained into their effects on water quality at different spatial and temporal scales (Chaubey et al., 2010; Meals et al., 2010; Lam et al., 2011). Adoption of BMPs has been used as an indicator of the sustainability and resilience of the agricultural sector (MacKay and Hewitt, 2010). Despite ongoing extension efforts and ﬁnancial support for adoption of BMPs, uptake has been relatively low –40%
throughout Canada (Eilers et al., 2010) and 56% in Alberta in 2014 (IPSOS REID, 2014). To the authors’ knowledge, similar broad assessments of agricultural BMP uptake have not been conducted in other countries; however, a commodity-speciﬁc assessment in Australia revealed approximately 60% of land area devoted to cotton production had been registered in a BMP auditing program (WWF Australia, 2005). Research has been devoted to understanding how and why BMPs and conservation practices are adopted. From their analysis of more than 2500 research reports on this subject in 2006, the United States Department of Agriculture Natural Resources Conservation Service highlights that producers most commonly go through six stages associated with the adoption-diffusion model (awareness, interest, evaluation, trial, adoption, and adaptation) when implementing conservation (American Farmland Trust, 2013). Investigating variables that inﬂuence the adoption of conservation practices in agriculture have been the subject of several reviews (e.g., Pannell et al., 2006; Knowler and Bradshaw 2007; Prokopy et al., 2008). Given methodological limitations and relative insigniﬁcance of ﬁndings from previous reviews, Baumgart-Getz et al. (2012), (p. 17) conducted a meta-analysis of 46 studies on BMP adoption in the United States from 1982 to 2007 and found that the largest impact on adoption was from “. . . access to and quality of information, ﬁnancial capacity, and being connected to agency or local networks of farmers or watershed groups.” Accessibility of information is one of the main reasons BMP adoption is limited (Brethour et al., 2007; Bjornlund et al., 2009) and the pathways by which agricultural producers access information and advice regarding their land management decisions remains unclear. It is important to understand whom farmers are relying on for information and/or advice regarding land management practices to potentially inﬂuence decision-making at the individual scale (Prell et al., 2009; Knoot and Rickenbach 2011) and to promote continual improvement and development of new practices by researchers (Council of Canadian Academies, 2013). The effectiveness of formal government information pathways may be limited as agricultural producers often mistrust these agencies and/or ﬁnd the costs to outweigh the beneﬁts in gaining access
J. Baird et al. / Land Use Policy 51 (2016) 236–243
Fig. 2. 3D advice network structure, BMPs, and the nutrient water quality sub-index.
to incentives for BMP implementation (Baird, 2012). Conversely, peers were identiﬁed as key actors for information exchange and advice in an analysis of informal farmer information exchange and
advice networks in a developing country context (Isaac et al., 2007). In the same region, linkages to organizations including government and non-governmental organizations proved essential for
J. Baird et al. / Land Use Policy 51 (2016) 236–243
farmers who adopted innovative and sustainable land management practices (Isaac, 2012). Isaac (2012) and Isaac et al. (2007) highlight the importance of maintaining both farmer–organization ties and farmer–farmer ties for successful and sustainable agricultural practices. The importance of the peer-to-peer advice network has also been found in the context of private forest management (Knoot and Rickenbach, 2011) as well as countryside stakeholders in the United Kingdom (Prell et al., 2009). Thus, informal (peerbased) entry points into farmer advice networks may be required by government agencies to effectively disseminate information, inﬂuence BMP adoption, and ultimately improve ecological outcomes. The spatial inﬂuence of peer-to-peer communication in inﬂuencing BMP adoption by others, as well as the role of location of land in relation to water bodies, have been identiﬁed as contributors to the adoption of BMPs in a region of the United States (Kaufman, 2011). The inﬂuence of social networks on decision making and sustainable practice adoption, and the spatial arrangement of these networks in relation to ecological conditions has been identiﬁed by Isaac (2012) as well as Bodin and Tengö (2012) as important and under studied areas within this scholarship with valuable applied implications. In this research study, we explored the advice network of agricultural producers from a spatially explicit social-ecological systems perspective and how this particularly important aspect of BMP adoption relates to water quality on the landscape. Specifically, the research objectives were to (1) query the individuals, organizations, and other sources of information agricultural producers turn to for advice when making decisions about land management practices, and in particular BMPs; (2) probe the characteristics of this advice network; and (3) link the advice network of individuals spatially and incorporate water quality data.
2. Research context Situated within the foothills of southern Alberta, the study watershed is located within a relatively high precipitation region with undulating topography and well-drained soils. A main water channel ﬂows from south to north within the watershed. Smaller tributaries drain the north-central and northwest portions of the watershed, and these tributaries drain into the northern end of the main channel. This channel is ephemeral; however, water ﬂow is manipulated by a dam, but ﬂow is primarily driven by snowmelt and rainfall events. The watershed is in an area that is mostly rural and sparsely populated (approximately 38,000 people in the Census Division of approximately 14,000 km2 in 2011), and agriculture is the main industry (Statistics Canada, 2013). Ranching, or raising cattle on pasture land, conﬁned feedlot operations and grain farming are all common activities in the region (Statistics Canada, 2006). Approximately two-thirds of farm operators are male and the average age of a farm operator in the division is 53.5 years (Statistics Canada, 2006). This watershed, encompassing 140 km2 of land, was selected for this study given that it is located within a region where agricultural effects on water quality are of special concern. Given these concerns, agricultural activities and the results of water-quality assessments have been well-documented over the past eight years. Agricultural landowners within the watershed have also shown interest in improving water quality through BMP adoption.1
1 In order to maintain conﬁdentiality agreements with survey respondents, the geographic name of the watershed and some of the information sources used to describe the watershed cannot be disclosed in this paper.
3. Research design and methods Each agricultural landowner in the watershed (N = 68) was provided with a brief introduction to the study via print material and then contacted by phone two weeks later in the spring of 2013. Three attempts were made by phone to invite participation, and after the second attempt, a voice message was left where possible providing information and a contact number. Appointments were scheduled to conduct in-person surveys during a one-week period with those who agreed to participate. Prior to beginning the inperson survey, respondents were informed of the content of the ethics protocol. The survey queried agricultural producer attributes, land parcel locations, agricultural activities and BMPs on each land parcel, reasons for adopting BMPs, and the sources of information and advice respondents accessed when making land management decisions (network questions). A map was used so that each respondent could indicate the land parcels they owned and managed. The in-person approach used ensured that surveys were completed fully and correctly, and that respondents fully understood the questions. For the network questions, a name generator and interpreter approach was used (Marsden, 1990), where respondents were asked to list up to 10 names of those with whom they communicate for information and/or advice, and the frequency with which they communicated to illuminate the strength of that relationship. Respondents were also asked to report the direction of the connection to each source of advice they identiﬁed (i.e., does the respondent contact the source for advice, does the source contact them, or do both occur?). SPSS 20 (IBM) was used to generate descriptive statistics (e.g., means, medians) for survey responses related to respondent attributes, agricultural activities, BMP adoption, and network measures. Social network data were entered as a matrix in Ucinet 6 (Analytic Technologies) and visualized using Netdraw (Analytic Technologies). An egocentric network analysis was performed (i.e., the focus of the analysis was on the network of individual respondents rather than connections among respondents) to understand who inﬂuenced decision making (‘alters’) for each respondent (‘ego’) (Marsden, 1990) and the mean size of respondents’ ego networks were calculated. Ego network analyses can vary in focus from collecting network information from only the ego, or extending data collection to also ask alters for their network information; however, in this study only the ego was surveyed (Knoot and Rickenbach, 2011). Land parcels identiﬁed on the maps were subsequently digitized within a geographic information system (GIS) using ArcGIS version 10.2.2 (Esri Canada). These parcels were spatially georeferenced (i.e., they were assigned to speciﬁc real-world geographic coordinates) and linked to an attribute table containing the survey data. Surface water quality data acquired from 10 watershed monitoring stations were also incorporated into the geospatial database in order to perform geospatial data analyses, such as spatial query and buffering. A spatial query was used to produce a GIS data layer containing land management practices (BMPs) based on the land parcel data. A spatial buffer was subsequently used to determine the proximity of each land parcel to the main water channel and tributaries in the watershed. Using the ArcScene 3D Visualizer in ArcGIS, the results of the social network analysis (SNA) were superimposed onto the newly created data layers to permit 3D geovisualization of the spatial relationships between the structure of the advice network, land management practices (BMP adoption), and a 2012 nutrient water quality sub-index provided to us by a third party. The nutrient water quality sub-index provided an overall measure of surface water quality in 2012 using nutrient parameters (several forms of nitrogen and phosphorus) and the number, frequency, and amount by which water quality objectives were not met for each watershed monitoring station
J. Baird et al. / Land Use Policy 51 (2016) 236–243
(following Wright et al., 1999). Index scores ranged from 0 to 100 (or poor to excellent) with ﬁve categories within this range. The nutrient water quality sub-index values represented the relative water quality integrated for each portion of the watershed upstream from each of the 10 water quality monitoring stations included in this study. The index value at the outlet represented a cumulative measure for the entire watershed. 4. Results Sixteen agricultural producers agreed to participate in the faceto-face survey. This represents a 24% response rate, which is consistent with other agricultural studies in the region (e.g., Nicol et al., 2010; Johnston et al., 2001). Most respondents (n = 14) were male, and the most common age of landowners was 56–65 years and ranged from 26–36 to 76+. More than half of the landowners (n = 9) produced grain and livestock (cattle), and only one landowner did not produce any livestock (i.e., grain farmer only). The results are presented in terms of the three objectives set out for the study. First, a description of the individuals, organizations, and other sources of information agricultural producers turn to for advice when making decisions about land management practices is provided. Then, the characteristics of the advice network are probed. Finally, the advice network of individuals is examined in relation to the spatial dimension and connections to water quality data. 4.1. Sources of information for advice Respondents identiﬁed four main sources of advice. The most common source of advice was regional actors with 63% of respondents identifying this source in their ego network. This included municipal ﬁeldmen (agricultural advisors hired by municipalities), regional employees for provincial agencies, and regional to national non-governmental organizations. Local actors, such as family and neighbors, were identiﬁed by 25% of respondents. Other businesses that were not considered local were included in 13% of respondents’ ego networks. 13% of respondents also listed text-based sources as part of their network, however, these sources were not included in subsequent analysis. It is important to note that 25% of respondents (n = 4) did not identify any source of advice or information regarding land management decisions. 4.2. Agricultural advice network structure The advice network was not fully connected (i.e., the network was fragmented), and where connections existed, the density of ties (number of connections relative to the total possible) was low (Fig. 1). Respondents 2, 12, 13, and 16 reported no advice network at all, and thus they are isolated from the network. Surprisingly, there were no direct connections among respondents. Regional and provincial extension agents (‘regional actors’) tended to be the nodes that held the network together, and in some cases this was through passive efforts where agricultural producers contacted them (e.g., Regional extension 1 node) and in others it was through passive and active sharing of information and advice (e.g., Provincial extension 1). Local actors also played a substantial role as sources of advice (Fig. 1) with six local sources of advice identiﬁed by respondents. Most of these were neighbors, however, these were not the same neighbors that responded to the survey. The role of the local watershed group was also of interest, but only two respondents reported a connection to this group, and in both cases the respondents indicated that they received advice but did not seek it from this source. Because there were no direct connections reported among respondents, the overall structure of the advice network of
respondents could not be analyzed. However, the role of the regional actors as advisors in agricultural land management decision making was clearly important and the respondents were grouped by the presence (n = 10) or absence (n = 6) of a regional actor in their ego network (Table 1). Those respondents connected to a regional actor tended to use their concern for water quality and economic beneﬁt more often as motivations for using BMPs. They also implemented more BMPs per land parcel and a greater diversity of BMPs overall (Table 1). Respondents identifying a regional actor in their network also used BMPs important for water quality management such as erosion control and riparian management more often. The reason for the lack of adoption of BMPs by the two respondents was that they did not apply to their particular operation. The size of the advice ego network for each respondent ranged from zero to four, and those connected to a regional actor tended to have a larger and more diverse ego network than those who were not. Also of interest was the distinction between governmental and non-governmental regional actors. These are shown in Fig. 1, where grey triangles represent government agencies and black triangles represent non-governmental organizations. From the network diagram it is clear that government agencies are more often identiﬁed in the advice networks of respondents than nongovernmental organizations.
4.3. Inﬂuence of the advice network structure on surface water quality The 3D geovisualization permitted visual examination of the spatial relationships between the structure of the advice network, BMP adoption, and a nutrient water quality sub-index (Fig. 2). The upper level of the advice network shows the regional, local, and other actors nodes operating in the watershed. These actors appear to be “ﬂoating” above the landscape because they are not situated within it, unlike the respondents. As an example of how to read the upper level, the nodes connected by the black solid lines connect respondents to regional actors (such as a municipal ﬁeldmen or a provincial extension) (red parcels), regional and local actors (purple parcels), or regional, local and other actors (light green parcels). The middle level of Fig. 2 shows the corresponding BMPs implemented on the land parcels. In this study we focused on BMPs that are particularly relevant to water quality in the area, such as improved cropping systems, grazing management, erosion control, manure management, and riparian management. Most respondents used at least one BMP, but in some areas, none were used (grey shaded land parcels). Respondents connected to regional advisors were more likely to use erosion control than those who were not (70% vs. 17%). The lower level of Fig. 2 shows the location of land parcels in relation to the main water channel and select tributaries in the watershed. The channel ﬂows from south to north and several water monitoring sites (7 out of 10) were distributed along the main channel. The 2012 nutrient water quality sub-index at each monitoring station ranged from good to poor. The sub-index values decreased from upstream to the downstream outlet of the main channel, as expected. The outlet represented a cumulative measure of water quality (i.e., poor) for the entire watershed. While it is inappropriate to make explicit links to water quality from one year of data collection, the 3D geovisualization technique is heuristic in visual examination of the spatial relationships between the advice network, BMPs, and the nutrient water quality sub-index results. This information is useful for establishing a baseline dataset from which changes (e.g., land management practices) that occur in the entire watershed could be monitored with time, especially if new
J. Baird et al. / Land Use Policy 51 (2016) 236–243
Table 1 Respondent attributes grouped by presence or absence of a regional advisor in the ego network. Respondent attribute
Connected to regional actor (n = 10)
Not connected to regional actor (n = 6)
Reasons for BMP use: Economic beneﬁt Concern for water quality Non-economic beneﬁts Others suggested it
Proportion 0.5 0.6 0.4 0.1
Proportion 0.3 0.3 0.3 0
BMP usage: Practices per ﬁeld Diversity of practices overall (9 possible)
Mean 2.8 3.3
Mean 1.6 2.0
Use of speciﬁc BMPs: Nutrient management planning Pest management Erosion/runoff control Grazing management Riparian management Improved cropping Manure/wastewater management Farmyard waste management Other No BMPs used
Proportion 0 0.10 0.70 0.70 0.70 0.20 0.20 0.30 0.30 0
Proportion 0 0 0.17 0.67 0.17 0.17 0.33 0 0.50 0.30
SNA measures: Ego network size Ego network diversity
Mean 2.1 0.4
Mean 1.5 0.25
information (e.g., the date of BMP adoption and edge of ﬁeld water quality measurements) was added as implemented. 5. Discussion Implementation of BMPs is essential to addressing nonpoint source pollution from agriculture, especially in terms of water quality. The BMPs respondents used on their land relating to water quality were mostly focused on erosion and runoff control, grazing management, and riparian management. Data for the region and province for adoption rates for individual BMPs are not available. Broad estimates of BMP adoption in Alberta based on a 2006 survey indicate that approximately half of respondents had not adopted BMPs (Alberta Research Council, 2006). Further, a Canadian study found that, of those Alberta agricultural producers who use BMPs, approximately 30–40% of all appropriate BMPs are implemented (Mackay et al., 2010). In this study, most respondents (n = 14) indicated that they had implemented at least one BMP and the average was 3.3 for those with regional actor connections and 2 for those without such a connection. Motivation for adopting BMPs was distributed among a range of reasons, from concern for water quality to economics to non-economic beneﬁts, and this is consistent with the ﬁndings of other studies in Alberta (Alberta Research Council, 2006; Banack and Hvenegaard, 2010). The reason for not implementing BMPs by respondents in the current study was that BMPs did not apply to their situation. This reason was different than ﬁndings from other Alberta and Canadian studies where lack of economic or personal beneﬁt, social implications, and time and labor requirements dominated responses (Alberta Research Council, 2006; Banack and Hvenegaard, 2010). Landlord-tenant relationship was also identiﬁed as a constraint in a meta-analysis of adoption of conservation practices in the United States (American Farmland Trust, 2013) and this was alluded to by some respondents. This research study concentrates on communication regarding land management practices because access to quality information and advice has the greatest impact on adoption (Prokopy, 2011; Baumgart-Getz et al., 2012) and is a major reason limiting the uptake of BMPs (Brethour et al., 2007). Social networks have been identiﬁed as an important factor in adoption (Prokopy, 2011; American Farmland Trust, 2013). Findings from this study show that the role of the regional actors was key to inﬂuencing land
management decision making, speciﬁcally concerning the adoption of erosion control and riparian management BMPs, and that local actors (neighbors and family) held a lesser role in advice networks. These ﬁndings are different from other agricultural and forestry advice networks, where local (‘non-expert’) actors held a more prominent position in the networks (Isaac et al., 2007; Knoot and Rickenbach, 2011). Isaac (2012) suggested that connections only to similar others (‘homophily’) can be a hindrance to the exchange of information and this is consistent with the results of this study, as those without a regional actor connection tended to adopt fewer BMPs. In addition, Knoot and Rickenbach (2011) acknowledge that, while respondents in their study communicated with local nonexperts, they tended to turn to experts for decision support. Further, the role of regional actors is key in the ﬁrst stages of the ‘AdoptionDiffusion’ model of BMPs (awareness and interest); however, local actors’ inﬂuence is considered important in several subsequent stages of the model (evaluation, trials, and adoption) (American Farmland Trust, 2013). Since access to information and advice is considered to be the most important factor in adoption of BMPs, it is clear that the role of regional actors is imperative to improve BMP adoption and thereby water quality on agricultural landscapes. When regional actors were separated into governmental and non-governmental actors, employees of government agencies appeared to have a greater effect than representatives of nongovernmental organizations. This ﬁnding is supported by the American Farmland Trust (2013), reporting that producers with greater awareness of water pollution, and the effects their own practices have on it, as a result of information from government sources were more likely to implement pollution control measures. Thus, the role of regional actors in government agencies is important for communicating and providing advice about land management practices (and speciﬁcally BMPs) to agricultural producers in this region, despite reductions in support for government-funded extension services (Millburn et al., 2010), which is part of a global trend (Swanson, 2008). The 3D geovisualization technique developed in this study provided an opportunity to explore the spatial dimension of the relationship between agricultural advice networks, BMP adoption, and surface water quality within an agricultural landscape. These relationships are not readily apparent through visualization of the social network alone (Fig. 1). While the results of the SNA were
J. Baird et al. / Land Use Policy 51 (2016) 236–243
superimposed onto agricultural land parcels, BMPs, and water quality data in the current study; GIS data layers could be conveniently exchanged to permit geovisualization of other BMPs and/or water quality parameters. Overall, the approach used in this study could be adopted to produce baseline datasets to which new information could be added with time. Equipped with this information, progress toward the objectives of water quality monitoring programs could be measured. In addition, this technique could be adapted for use in watersheds elsewhere. The results of this study are particularly relevant as they contribute to a nascent but growing body of literature that uses SNA to explore social-ecological systems by incorporating a geospatial dimension to these analyses (Ernstson et al., 2010; Bodin and Tengö, 2012; Bergsten et al., 2014; Bodin et al., 2014). For example, Bodin and Tengö (2012) linked social networks of clans in Madagascar via social networks and then to the landscape through pollen transmission among forest patches where clans accessed ecosystem services. Their approach used motifs, or variations of connections among two social nodes and two ecological nodes, to analyze governance in social-ecological systems. Ernstson et al. (2010) used geospatial methods to address mismatches of scale between ecological and social processes in Stockholm, Sweden, and identiﬁed scale-crossing brokers and mid-scale managers as important for the continued provision of ecological services at the appropriate scale. The 3D visualization presented therein (Ernstson et al., 2010 Fig. 2) served as the inspiration for the visualization presented in this paper. While this study contributes to the above literature, its focus on sources of advice and understanding the connections between those sources and ecological conditions is unique in comparison.
Starting with the social network layer, which actors are relied upon most often for advice by agricultural producers? How do those actors in turn, inﬂuence BMP adoption? And, ultimately, how does that translate into ecological outcomes? Through this process of interrogating the data, focus and resources can be allocated toward those actors that are most inﬂuential in the landscape (in terms of inﬂuencing agricultural producers to adopt the BMPs that have a positive impact on ecological conditions). In the present study, this might mean focusing on improving access to governmental actors at the regional level (e.g., extension agents) and re-establishing funding for these actors that has been diminished with time (Millburn et al., 2010). However, this study represents a ﬁrst, localized step in a much broader effort that would be required for policy decisions. For example, implementing the approach in diverse agricultural watersheds throughout a province or nationally may result in identiﬁable trends or illuminate case-speciﬁc differences. This is a potentially fruitful area for future research. The ability to connect the social and ecological aspects of agricultural land management also provides an opportunity to consider its practical applications. This is illustrated by considering Fig. 2 from the bottom-up. For example, where are ecological outcomes within a desirable range on the landscape? What BMPs are being used in those locations? And, which actors are connected to those locations? Asking these questions identiﬁes what BMPs and connections to advisors would be most beneﬁcial in inﬂuencing positive change on the landscape for locations where undesirable ecological outcomes are prevalent.
Acknowledgements 6. Conclusions There is a pressing need to address negative effects on the environment from agriculture operations of all sizes. These effects on water quality are of particular concern and BMPs are central to mitigating nonpoint source pollution from agricultural activities. While research has been directed at gaining insights into the adoption of BMPs, the rate of uptake remains generally low (MacKay et al., 2010). In exploring the advice network of agricultural producers in an Alberta watershed from a spatially explicit social-ecological systems perspective, this research study draws attention to the inﬂuences of regional actors, and in particular regional government actors. As such, it stands apart from previous empirical investigations that point to local actors as most inﬂuential in agricultural and forestry networks (Isaac et al., 2007; Knoot and Rickenbach, 2011) and questions about the effectiveness of information provision by governments in BMP implementation due to mistrust (Baird, 2012). At the same time, it reinforces that experts are looked to when making decisions as opposed to local non-experts (Knoot and Rickenbach, 2011). More fully understanding who, how, and under what circumstances, inﬂuence is exerted in agricultural advice networks is needed. Determining the roles and inﬂuences of different actors is further complicated as they are expected to change throughout the dynamic process of BMP adoption, as captured in the adoption diffusion model (American Farmland Trust, 2013). Clear connections in this research were identiﬁed among advice networks as a particularly important aspect of BMP adoption and implementation. The 3D geovisualization in this work illustrates the establishment of a baseline dataset. Explicit linkages to water quality on the landscape may be established over time. Overall, the 3D geovisualization captures a wealth of insights and offers a touchstone to inform the development of programs and policies. Questions of it may be asked by individuals concerned with BMP adoption, water quality, and conservation efforts more broadly.
The authors wish to extend their thanks to the participants in this project and to Alberta Agriculture and Rural Development for their important role in supporting this research. Thanks also to research assistant Katrina Krievins for data collection support and Loris Gasparotto for his assistance with illustrations presented in this article. Funding from the Post-doctoral Fellowship in Sustainability Science by the Environmental Sustainability Research Centre at Brock University and the Social Sciences and Humanities Council of Canada is gratefully acknowledged.
References Alberta Research Council, 2006. Study on identifying rural sociological barriers to adoption. In: Alberta Research Council, Agriculture and Agri-Food (eds.). Canada, Alberta Agriculture and Rural Development and The Alberta Farm Plan Company. Edmonton, AB. American Farmland Trust, 2013. The Adoption of Conservation Practices in Agriculture. Report, DeKalb, Illinois. Baird, J., 2012. Performance-based approaches to agri-environmental water quality policy in Canada. In: PhD Dissertation. Department of Environment and Sustainability University of Saskatchewan, Saskatoon, Saskatchewan. Banack, S.A., Hvenegaard, G.T., 2010. Motivations of landowners to engage in biodiversity-friendly farming practices in Alberta’s Central Parkland Region. Hum. Dimens. Wildl. 15, 67–69. Baumgart-Getz, A., Prokopy, L.S., Floress, K., 2012. Why farmers adopt best management practice in the United States. J. Environ. Manage. 96 (1), 17–25. Bergsten, A., Galafassi, D., Ö. Bodin, 2014. The problem of spatial ﬁt in social–ecological systems: detecting mismatches between ecological connectivity and land management in an urban region. Ecol. Soc. 19 (4), 6 [online] URL: http://www.ecologyandsociety.org/vol19/iss4/art6/#results11. Bjornlund, H., Nicol, L., Klein, K.K., 2009. The adoption of improved irrigation technology and management practices—a study of two irrigation districts in Alberta, Canada. Agric. Water Manage. 96 (1), 121–131. Bodin, Ö., Tengö, M., 2012. Disentangling intangible social–ecological systems. Global Environ. Chang. Part A: Hum. Policy Dimens. 22 (2), 430–439. Bodin, Ö., Crona, B., Thyresson, M., Golz, A.-L., Tengö, M., 2014. Conservation success as a function of good alignment of social and ecological structures and processes. Conserv. Biol. 28 (5), 1371–1379. Brethour C., Sparling B., Cortues B., Klimas M., Moore T., Sadler Richards J., 2007. An economic evaluation of beneﬁcial management practices for crop nutrients in Canadian agriculture. Final Report. Guelph, Ontario, 165.
J. Baird et al. / Land Use Policy 51 (2016) 236–243 Chaubey, I., Chiang, L., Gitau, M.W., Mohamed, S., 2010. Effectiveness of best management practices in improving water quality in a pasture-dominated watershed. J. Soil Water Conserv. 65 (6), 424–437. Council of Canadian Academies, 2013. Water and Agriculture in Canada: Towards Sustainable Management of Water Resources, The Expert Panel on Sustainable Management of Water in the Agricultural Landscapes of Canada, Council of Canadian Academies. Dowd, B., Press, D., Los Huertos, M., 2008. Agriculture nonpoint source water pollution policy: the case of California’s Central Coast. Agric. Ecosyst. Environ. 128, 151–161. Eilers, W., MacKay, R., Graham, L., Lefebvre, A., 2010. Executive summary. In: Eilers, R., Graham, L., Lefebvre, A. (Eds.), Environmental Sustainability of Canadian Agriculture Agri-Environmental Indicator Report Series-Report #3. Agriculture and Agri-Food Canada, Ottawa, Ontario, p. 244. Ernstson, H., Barthel, S., Andersson, E., Borgström, S.T., 2010. Scale-crossing brokers and network governance of urban ecosystem services: the case of Stockholm. Ecol. Soc. 15 (4), 28 [online] URL: http://www.ecologyandsociety. org/vol15/iss4/art28/. FitzGibbon, J., Hammel, S., Matrunec, S., 2002. Report on Nutrient Management By-Laws In the Province of Ontario, Report SR. 9086, Guelph, ON: Ontario Ministry of Agriculture, Food and Rural Affairs. Gomiero, T., Pimentel, D., Paoletti, M.G., 2011. Is there a need for a more sustainable agriculture? Crit. Rev. Plant Sci. 30 (1–2), 6–23. Hillard C., Scott, N., Lessa, A., Reedyk, S., 2002. Agriculturee best management practices for the Canadian Prairies: a review of literature. File No.: 6672-1-12-1-18. Agriculture and Agri-Food Canada. Ice, G., 2004. History of innovative best management practice development and its role in addressing water quality limited waterbodies. J. Environ. Eng. 130, 684–689. IPSOS REID, 2014. Environmentally sustainable agriculture tracking survey. In: Summary Report. Alberta Agriculture and Rural Development, Calgary, Alberta. Isaac, M., 2012. Agriculture information exchange and organization ties: the effect of network topology on managing agrodiversity. Agric. Syst. 109, 9–15. Isaac, M., Erickson, B., Quashie-Sam, S., Timmer, V., 2007. Transfer of knowledge on agroforestry management practices: the structure of farmer advice networks. Ecol. Soc. 12 (2), 32. Johnston, T.R.R., Kromm, D.E., Byrne, J.M., 2001. On-farm water conservation practices in Southern Alberta. J. Am. Water Resour. Assoc. 37 (3), 737–750. Kaufman, Z., 2011. Agricultural best management practice adoption decisions and spatial dependence in southern Pennsyvania farms and watersheds. In: Master’s Thesis. Pennsylvania State University. Knoot, T.G., Rickenbach, M., 2011. Best management practices and timber harvesting: the role of social networks in shaping landowner decisions. Scand. J. For. Res. 26 (2), 171–182. Knowler, D., Bradshaw, B., 2007. Farmers’ adoption of conservation agriculture: a review and synthesis of recent research. Food Policy 32 (1), 25–48. Lam, Q., Schmalz, B., 2011. The impact of agricultural best management practices on water quality in a North German lowland catchment. Environ. Monit. Assess. 183 (1–4), 351–379. MacKay, R., Bennett, E.M., Lefebvre, A., 2010. Using a beneﬁcial management practice (BMP) adoption index in agri-environmental policy in Canada. In: OCED Workshop on Agri-Environmental Indicators, Leysin, Switzerland, p. 14. MacKay, R., Hewitt, J., 2010. Farm environmental management. In: Eilers, R., Braham, L., Lefebre, A. (Eds.), Environmental Sustainability of Canadian Agriculture: Agri-Environmental Indication Report Series-Report #3. Agriculture and Agri-Food Canada, Ottawa, Ontario. Marsden, P.V., 1990. Network data and measurement. Annu. Rev. Soc. 16, 435–463. Meals, D.W., Dressing, S.A., Davenport, T.E., 2010. Lag time in water quality response to best management practices: a review. J. Environ. Qual. 39 (1), 85–96. Millburn, L., Mulley, S., Kline, C., 2010. The end of the beginning and the beginning of the end: the decline of public agricultural extension in Ontario. J. Ext. 46 (6), 6FEA7.
Mulla, D.J., Birr, A.S., Kitchen, N.R., David, M.B., 2006. Limitations of evaluating the effectiveness of agricultural management practices at reducing nutrient losses to surface waters. In: Final Report: Gulf Hypoxia and Local Water Quality Concerns Workshop. American Society of Agricultural and Biological Engineers, St. Joshep, Minnesota, pp. 189–212. Nicol, L., Henning, B., Klein, K.K., 2010. Private irrigators in Southern Alberta: a survey of their adoption of improved irrigation technologies and management practices. Can. Water Resour. J. 35 (3), 339–350. Organisation for Economic Co-operation and Development (OECD), 2012. Water Quality and Agriculture: Meeting the Policy Challenge. OECD Studies on Water, 2012. Pannell, D.J., Marshall, G.R., Barr, N., Curtis, A., Vanclay, F., Wilkinson, R., 2006. Understanding and promoting adoption of conservation practices by rural landholders. Aust. J. Exp. Agric. 46 (11), 1407–1424. Plummer, R., Spiers, A., Summer, R., Fitzgibbon, J., 2008. The contributions of stewardship to managing agro-ecosystem environments. J. Sustain. Agric. 31 (3), 55–84. Prell, C., Hubacek, K., Reed, M., 2009. Stakeholder analysis and social network analysis in natural resource management. Soc. Nat. Resour. 22 (6), 501–518. Pretty, J., Brett, C., Gee, D., Hine, R., Mason, C., Morison, J., Rayment, M., Van Der Bijl, G., Dobbs, T., 2001. Policy challenges and priorities for internalizing the externalities of modern agriculture. J. Environ. Plann. Manage. 44 (2), 263–283. Prokopy, L.S., 2011. Agricultural human dimensions research: the role of qualitative research methods. J. Soil Water Conserv. 66 (1), 9A–12A. Prokopy, L.S., Floress, K., Klotthor-weinkauf, D., Baumgart-getz, A., 2008. Determinants of agricultural best management practice aoption: evidence from the literature. J. Soil Water Conserv. 63 (5), 300–311. Statistics Canada, 2006. 2006 Census of Agriculture, Farm Data and Farm Operator Data. Catalogue No. 95-629-XWE, Ottawa. Statistics Canada, 2013. Division No. 3, CDR, Alberta (Code 4803) (table). National Household Survey (NHS) Proﬁle. 2011 National Household Survey. Catalogue No. 99-004-XWE, Ottawa, 2013. Stoate, C., Báldi, A., Beja, P., Boatman, N.D., Herzon, I., van Doorn, A., de Snoo, G.R., Rakosy, L., Ramwell, C., 2009. Ecological impacts of early 21st century agricultural change in Europe—a review. J. Environ. Manage. 91 (1), 33–46. Swanson, B.E., 2008. Global Review of Good Agricultural Extension and Advisory Service Practices. Food and Agriculture Organization of the United Nations, Rome. Tilman, D., 1999. Global environmental impacts of agricultural expansion: the need for sustainable and efﬁcient practices. Proc. Natl. Acad. Sci. U. S. A. 96 (11), 5995–6000. Tilman, D., Cassman, K., Matson, P., Naylor, R., Polasky, S., 2002. Agricultural sustainability and intensive production practices. Nature 418 (6898), 671–677. Usda, 2013. Research – Phosphorous – BMPs. http://www.usawaterquality.org/ themes/npm/research/p bmp.html (accessed 2014.). USEPA, 2002. National Water Quality Inventory: Report to Congress, 2002 Reporting Cycle, United States Environmental Protection Agency. [Online] URL: http://www.epa.gov/305b/. Weersink, A., Livernois, J., Shogren, J.F., Shortle, J.S., 1998. Economic instruments and environmental policy in agriculture. Can. Public Pol. 24 (3), 309–327. WWF Australia, 2005. Case study: development of the Australia cotton industry Best Management Practice (BMP) Program. Australia. Wright, C.R., Saffran, K.A., Anderson, A.-M., Neilson, R.D., MacAlpine, N.D., Cooke, S.E., 1999. A Water Quality Index for Agricultural Streams in Alberta: The Alberta Agricultural Water Quality Index (AAWQI). Prepared for the Alberta Environmentally Sustainable Agriculture Program (AESA). Published by Alberta Agriculture, Food and Rural Development. Edmonton, AB. 35 pp. Young, T.F., Karkosky, J., 2000. Green evolution: are economic incentives the next step in nonopint source pollution control? Water Policy 2, 151–173.