382

AbstractNutrient enrichment of water resources has degraded coastal waters throughout the world, including in the United States (e.g., Chesapeake Bay, Gulf of Mexico, and Neuse Estuary). Agricultural nonpoint sources have significant impacts on water resources. As a result, nutrient management planning is the primary tool recommended to reduce nutrient losses from agricultural fields. Its effectiveness requires nutrient management plans be used by farmers. There is little literature describing nutrient management decision-making. Here, two case studies are described that address this gap: (i) a synthesis of the National Institute of Food and Agriculture, the Conservation Effects Assessment Project, and (ii) field surveys from three nutrient-impaired river basins/watersheds in North Carolina (Neuse, Tar-Pamlico, and Jordan Lake drainage areas). Results indicate farmers generally did not fully apply nutrient management plans or follow basic soil test recommendations even when they had them. Farmers were found to be hesitant to apply N at university-recommended rates because they did not trust the recommendations, viewed abundant N as insurance, or used recommendations made by fertilizer dealers. Exceptions were noted when watershed education, technical support, and funding resources focused on nutrient management that included easing management demands, actively and consistently working directly with a small group of farmers, and providing significant resource allocations to fund agency personnel and cost-share funds to farmers. Without better dialogue with farmers and meaningful investment in strategies that reward farmers for taking what they perceive as risks relative to nutrient reduction, little progress in true adoption of nutrient management will be made.

Farmers’ Use of Nutrient Management: Lessons from Watershed Case Studies

Deanna L. Osmond,* Dana L. K. Hoag, Al E. Luloff, Donald W. Meals, and Kathy Neas

Agricultural nonpoint source losses of nitro-gen (N) and phosphorus (P) as well as the global water quality impacts of nutrient losses have been well docu-

mented. Manifestations of nutrient enrichment on coastal waters, particularly N, include green tides off the coast of Britany, France (Charlier et al., 2007); the dead zone in the Gulf of Mexico (Diaz and Rosenberg, 2008; Rabalais and Turner, 2001; Goolsby, 2000); massive fish kills in the Neuse Estuary (Paerl et al., 1998); and degraded water quality in the Chesapeake Bay (Cooper, 1995). Numerous studies have demonstrated production agri-culture’s impact on surface and groundwater nutrient concen-trations (Rupert, 2008; Schilling and Libra, 2000). Although agriculture was not the only source of these nutrients, farming is an inherently leaky system and often constitutes a significant portion of the pollution (Preston and Brakebill, 1999; Randall and Vetsch, 2005; Alexander et al., 2008; White et al., 2014).

To reduce agricultural nutrient losses, conservation systems must be used to control them at their source, as they are transported, and at or in the receiving water. The USDA Natural Resources Conservation Service (NRCS) refers to this as “avoid, control, and trap” (ACT). No single conservation practice is sufficient to control all nonpoint source losses, but nutrient management is critical for reducing nutrients at their source (i.e., the field). Sogbedji et al. (2000) demonstrated 30 to 50% increases in groundwater nitrate-N (NO3–N) leaching with only a 38 kg N ha-1 overapplication to corn. When farmers in the treatment watershed used the late sidedress nitrate test and fertilized accordingly, stream concentrations of NO3–N were lower than the control watershed where farmers did not adjust N fertilizer rates ( Jaynes et al., 2004). These studies, along with many others, demonstrate the importance of nutrient management as a conservation tool to protect water quality.

What is less clear is the relationship between nutrient management planning by agency personnel, nutrient management plans written for farmers, and fertilizer behavior of producers. Anecdotal comments from land grant nutrient

Abbreviations: CEAP, Conservation Effects Assessment Project; NIFA, National Institute of Food and Agriculture; NLEW, Nitrogen Loss Estimation Worksheet; NRCS, Natural Resources Conservation Service; RYE, realistic yield expectations; STP, soil test phosphorus.

D.L. Osmond, North Carolina State Univ., Dep. of Soil Science, Box 7619, Raleigh, NC 27695; D.L.K. Hoag, Colorado State Univ., Agricultural and Resource Economics, Box 1172, Ft. Collins, CO 80523; A.E. Luloff, Pennsylvania State Univ., Agricultural Economics, Sociology, and Education, 114 Armsby Building, University Park, PA; D.W. Meals, Ice.Nine Environmental Consulting, 84 Caroline St., Burlington, VT 05401; K. Neas, USDA–NASS, 2 West Edenton Street, Raleigh, NC 27601. Assigned to Associate Editor Eric Davidson.

Copyright © American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America. 5585 Guilford Rd., Madison, WI 53711 USA. All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. J. Environ. Qual. 44:382–390 (2015) doi:10.2134/jeq2014.02.0091Freely available online through the author-supported open-access option.Received 27 Feb. 2014. *Corresponding author (deanna_osmond@ncsu.edu).

Journal of Environmental QualityIMPROVING NITROGEN USE EFFICIENCY IN CROP AND LIVESTOCK PRODUCTION SYSTEMS

SPECIAL SECTION

Published March 11, 2015

www.agronomy.org • www.crops.org • www.soils.org 383

management specialists often focus on the disconnect between plans and actual nutrient application rates. A recent summary of conservation practice adoption literature found that no attribute (e.g., acres, age, capital, education, experience, income, networks, land tenure awareness) consistently accounted for nutrient management adoption (Prokopy et al., 2008). However, as stated by Prokopy et al. (2008), most studies focused on farm structure (e.g., size of farm, rental acres, crops produced, farmer demographic information) rather than interviews that explored farmer decision-making. Bosch et al. (1995) reviewed the use of N soil testing and test results in fertilizer decision-making in a semi-regulated area in Nebraska; these researchers found that the regulatory approach did not promote the use of soil N test results and that farmers needed more education along with regulation. More recently, Stuart et al. (2014) suggested that in Michigan, fertilizer dealers and seed companies were critical sources in determining N fertilizer rates, and university recommendations and other tools, such as pre-sidedress NO3 test and GreenSeeker technology, were rarely used. Farmers were also increasing their N rates seeking increased yields due to high corn prices (Stuart et al., 2014).

Understanding farmer fertilizer decision-making is critical to understanding nutrient management adoption, which includes basic nutrient management, currently referred to by NRCS and fertilizer industry as the 4Rs: right placement, right rate, right source, and right timing. Understanding how conservation practices are adopted in general is, in some cases, also applicable to nutrient management decision-making. Insights into nutrient management and conservation practice decision-making by farmers are presented in this case study, which includes results from the recently completed synthesis of the National Institute of Food and Agriculture (NIFA) Conservation Effects Assessment Project (CEAP) watersheds study and information from

two regulated river basins (Neuse and Tar-Pamlico) and two regulated watersheds ( Jordan and Falls Lake) in North Carolina.

Materials and MethodsNIFA CEAP

The USDA implemented CEAP in 2002 with the goal of using science to determine the effectiveness of conservation practices at the watershed scale (Duriancik et al., 2008). Multiple USDA organizations were partners, including, but not limited to, the NRCS, the Agricultural Research Service, and NIFA. Other federal agencies, such as the USEPA, the United States Fish and Wildlife Service, and the United States Geological Survey, were close partners. Multiple initiatives were implemented under CEAP, including the NIFA CEAP watersheds study project.

To better understand conservation practice adoption and efficacy at the watershed scale, USDA NIFA funded 13 watershed-scale projects (Fig. 1). Agricultural projects (crop and pasture only) selected as part of the NIFA CEAP protocol were expected to have a minimum of 5 yr of water quality and land use data for the watershed. Each was expected to answer four questions: (i) What was the relationship between conservation practices and water quality change? (ii) What was the relationship between conservation practices to each other relative to water quality change? (iii) What socioeconomic factors increased or decreased adoption of conservation practices? and (iv) What was the optimal placement of practices in the watershed to achieve water quality goals?

As these watershed studies were ending, USDA NIFA funded a synthesis project to develop lessons learned from the 13 projects. A multidisciplinary, interinstitutional team worked alongside the13 project investigators to conduct a synthesis of these projects (Osmond et al., 2012). Lessons were developed

Fig. 1. National Institute of Food and Agriculture Conservation Effects Assessment Project locations.

384 Journal of Environmental Quality

from each of the 13 projects using a systematic project outline filled out by watershed project personnel, project literature (peer-reviewed publications, fact sheets, and reports), presentations, site visits, and key informant interviews conducted in each watershed. For the purposes of this paper, we focus on the socioeconomic lessons learned relative to the adoption of conservation practices, particularly nutrient management (Hoag et al., 2012b; Luloff et al., 2012; Woods et al., 2014).

An interactive, semistructured key-informant survey instrument was developed by synthesis team members to address questions about conservation practice acceptance, implementation, maintenance, and the role of extension and education as viewed by a wider audience than the project team. Key informants (farmers, community leaders, agency personnel, citizens, project personnel, and others knowledgeable about the CEAP project specifically and about water quality and quantity efforts in general) were identified for interviews using the interactive survey that contained broad questions about the project. For example, we asked the following questions: What makes this place and the people where you live unique? What are the most critical water quality concerns in this watershed? What conservation practices were being used here before this project began? Have any new conservation practices been added? This key informant interview guide was pretested, and adjustments were made as necessary. Extensive notes were taken during the key informant process (according to Institutional Review Board defined and accepted rules), transcribed immediately after the interviews so that nuance would not be lost, and saved to electronic files. These files were summarized by the interviewers for each site location and then integrated across sites. Internal validity assessments were established by having multiple interpretations of findings conducted. Where disagreements

occurred, reassessments were undertaken to identify the correct interpretation.

North Carolina Basin and Watershed Agricultural Surveys

Over the past 18 yr, the State of North Carolina adopted regulations in river basins (Neuse and Tar-Pamlico) or watersheds ( Jordan Lake and Falls Lake) considered nutrient impaired (Fig. 2) (NCDENR 2014a,b,c,d). To better determine agricultural activities in these basins or watersheds (Falls Lake is the upper portion of the Neuse River Basin), surveys, using a computerized area frame sampling technique (Monroe and Finkner 1959), were administered within water boundaries. Census blocks were used to exclude data (urban areas) or to include areas (agricultural) and then randomly selected to aggregate into sampling segments (Osmond et al., 2013). The segments, which included 1 to >30 fields, were geolocated and printed on a map. Farmers were interviewed, and fields were surveyed by USDA enumerators who had been trained using a pretested survey. Collected information varied slightly by survey but usually included county, field size, previous crop, current crop, fertilizer applications for previous or current crops (rate and source of N, P, and potassium [K]), tillage type, cover crop use, buffer vegetation type, buffer widths, area affected by the buffers (in ha), presence of water control structures, area affected by the water control structure (in ha), field slope, slope length, receiving slope length, presence of other conservation practices (sediment basin or pond), soil test P levels, frequency of soil testing, livestock types, livestock density, and some farmer socio-demographic information (Osmond et al., 2006; Osmond and Neas, 2007, 2011). Soil mapping units were derived by matching the georeferenced location with USDA Web Soil Survey,

Fig. 2. Neuse and Tar-Pamlico River Basins and Jordan and Falls Lakes Watershed, North Carolina delineations and land uses.

www.agronomy.org • www.crops.org • www.soils.org 385

which allowed the calculation of soil erosion using the Revised Universal Soil Loss Equation (USDA-ARS, 2014), as well as the appropriate N rates for the crop produced in each field. Soil samples were taken for each field and processed through the state soil testing laboratory when farmers did not already have sample results to obtain soil-test P values. Each field survey was reviewed by three different teams before data were entered and analyzed using SAS software. Information from over 5000 useable agricultural records was compiled by water boundaries: 1156 agricultural fields in the Tar-Pamlico River Basin, 3355 agricultural fields in the Neuse River Basin, and 650 agricultural fields from the Jordan Lake Watershed. Each data record was processed through the Nitrogen Loss Estimation Worksheet (NLEW) (Osmond et al., 2001) to compare farmer-applied N rates with state agency–recommended N rates to determine farmer use of N fertilizers. Although the agricultural areas and cropping systems were not homogenous between basins or watersheds, the results provided an excellent understanding of current farmer practices in these areas (Osmond et al., 2006; Osmond and Neas, 2007, 2011).

Results and DiscussionNIFA CEAP Synthesis

A key informant survey conducted in each of the 13 NIFA CEAP watersheds was especially useful in providing information about farmers’ thinking relative to conservation practice adoption and use. Of the 196 watershed key informants, 33 were farmers. Farmers understood the water quality problem in their watershed almost as well as the federal agency personnel and as well as did university, soil and water conservation district, and watershed association personnel. The decision to adopt or not to adopt conservation practices was not associated with

understanding the water quality problem in their watershed; there were other more important factors (Luloff et al., 2012).

Conservation practice implementation was a multifaceted decision (Hoag et al., 2012b; Woods et al., 2014), just as Prokopy et al. (2008) found. Economics were important but could be superseded by other factors, such as yield requirements and ease of management (Table 1). There were also intangibles (Hoag et al., 2012b; Woods et al., 2014), such as the threat of regulation, an ability to see the pollutant of concern (sediment vs. nutrients), land ownership, family dynamics, and the belief systems of the farmer (Reimer et al., 2011), which affected conservation practice adoption. Further, some of these factors increased or decreased adoption depending on individual farming conditions. Some other factors often provided disincentives, such as expensive and inflexible conservation practice design standards, excessive legal requirements, and practices that required a considerable amount of management (Table 1). Management was mentioned by many of the farmers interviewed during the key informant conversations. As one farmer put it, “Conservation competes with the time the farmer could be using to make money.” Farmers often cited time savings as the reason they adopted conservation tillage (Woods et al., 2014), along with costs and technological developments.

The two most disliked conservation practices were riparian buffer installations and nutrient management (Luloff et al., 2012). Riparian buffers were disliked because they removed land from production. Farmers disliked nutrient management for a variety of reasons. In one watershed with significant areas under nutrient management plans on paper, we were told farmers had abandoned these plans. As a key informant in that watershed said, “Nutrient management was a failure. Some folks cheated the system and some just wouldn’t sign up. Farmers want to brag about yields and not return on investments.” Farmers often expressed the need to ensure sufficient nutrients to their

Table 1. Summary of factors found to increase or decrease adoption of conservation practices in National Institute of Food and Agriculture-Conservation Effects Assessment Project project surveys.†

Category Increase adoption Decrease adoptionFarm business Profit practice also improves profits reduces profits Yield practice increases crop yields practice decreases yields Production cost lowers cropping costs increases cropping costs Conservation cost practice is expensive to install and maintain Other government financial incentives, cost share lost farm land, time, attitude that farming is a business, high

commodity prices, lack equipmentConservation efficacy On farm can see the result on-farm (e.g. reduced erosion) Off farm can see off-site results don’t see off-farm consequences Other difficult to install and maintain Stewardship belief that conservation is doing the right thing water quality problems are someone else’s fault/problem,

lack knowledge/awareness of cons. practices or assistance, concervation not acceptable to farming community

Government/NGO‡ Relationship personal attention (time) from experts (e.g., NRCS, Extension);

networking and feedback, trust (knowledge, local)lack of trust

Approach bottom up, not top down, not being told top down, prescribed, inflexible practice recommendations/requirements, lost control of property/decisions, paperwork

Other regulations

† From Hoag et al. (2012b).

‡ NGO, nongovernmental organization.

386 Journal of Environmental Quality

crops for optimum yielding years as a guide for determining N rates (Nebraska and Kansas); conversely, farmers often did not believe the university fertilizer recommendations—particularly regarding N application—were correct (Kansas and Indiana). These farmers are not unlike those surveyed in Michigan, who rarely use university recommendations and who were concerned that reductions in fertilizer would reduce yields (Stuart et al., 2014).

One NIFA CEAP study (Nebraska) focused on reducing N in the groundwater (Meals et al., 2012b). Due to high NO3–N in the aquifer, farmers were semiregulated in the area of study. They were required to test and report N content in irrigation water, animal waste, and soil to a depth of 0.9 m. Fertilizer timing and record keeping was mandated, as was education. A survey of 600 farms in the larger Central Platte Natural Resource District indicated a large number of farmers using soil testing (regular soil samples, 91%; deep soil samples, 72%), crediting legumes (75%), and testing irrigation water for NO3 (50%). However, in our key informant survey, one farmer stated that he stopped using his moisture meters because too much management was required. Despite a high level of nutrient management practices, many farmer key informants in this project were clear that ensuring sufficient N fertilizer rates was essential (Woods et al., 2014). Perceived effectiveness of the nutrient management efforts varied relative to the key informant’s profession. For instance, one agency official said, “There are a lot of outreach and education programs for producers through University of Nebraska Lincoln extension, the Natural Resource Districts, and INE staff. They advertise the programs though newsletters, news media, farm magazines. I think the programs have been very effective given the high demand for participation, especially in the nutrient management and transfer irrigation systems.” However, a farmer stated the following: “… probably the least effective [conservation practices] would be the seminars explaining to farmers how they can reduce the N that they use. Even if they have facts and field data, farmers always say, well that was your field and your conditions but they hate to put too little on in case it’s a really good year. They want the N [to be] there to use it. Farmers are always optimists—this is going to be a good year.” Although Bosch et al. (1995) believed that education would lead to behavior change, this did not appear to be the case.

The reluctance of Nebraska farmers to change N rates was confirmed through an analysis of reductions in groundwater NO3–N from 1986 to 2002 (Exner et al., 2010). The NO3–N reductions of 0.26 mg L-1 yr-1 were attributed not to reduced

N fertilizer rates (which were in fact level) but to increased corn yields and lower irrigation rates as a consequence of changes from furrow to sprinkler irrigation.

Nutrient management was widely adopted in two of the NIFA-CEAP watersheds: Arkansas and New York. The Arkansas Lincoln Lake watershed NIFA CEAP project focused on nutrient management (Hoag et al., 2012a) on cattle pastures that receive litter from broiler operations. Lincoln Lake is part of the Illinois River system, and P reductions were required due to litigation between the states of Arkansas and Oklahoma. Conservation practices focused on nutrient management, exclusion fencing, and pasture management adoption (Table 2). Farmer adoption was low until a dedicated extension agent was hired who gained the trust of watershed farmers. By working with a small number of farmers (<65) on a one-to-one basis, this agent helped producers trust soil test results, lower litter application rates, and export litter outside the watershed. Nutrient management change in this watershed required resources to hire an extension agent (Hoag et al., 2012a,b) and the threat from the lawsuit. Generally, nutrient management plans are written without additional outreach to farmers, so it is not surprising that in many NIFA CEAP watersheds the plans were not implemented.

New York City’s drinking water comes from multiple reservoirs in upstate New York; the City pays for agricultural conservation practices in these watersheds to protect water quality. The Cannonsville Reservoir, one of these water resources, was the focus of the NIFA CEAP study (Osmond et al., 2012). Many conservation practices were funded by New York City, but the cornerstone of water quality protection was nutrient management. In the Cannonsville Reservoir watershed, nutrient management plans were rewritten every 3 yr, and color-coded maps were generated that specified the timing and rate of dairy manure application by field. Farmers used calendars for their nutrient management record-keeping and indicated the date and loads of manure by field in the calendar. On a yearly basis, the calendars were submitted to a peer-review panel consisting of farmers who certified compliance (Osmond et al., 2012). Certification allowed farmers to receive supplementary reimbursement that helped offset additional costs incurred in following the nutrient management plan. Nutrient management plans were used in this watershed because plans were simplified, significant financial resources were provided, and long-term and sufficient technical assistance was available to farmers. In a poignant comment, one farmer whom the research team visited

Table 2. Conservation practices and USDA–NRCS Conservation Practice Standard Number for Lincoln Lake NIFA-CEAP Watershed Project Study.

Best management practice USDA–NRCS conservation practice standard numberAdoption of an approved nutrient management plan 590: Nutrient ManagementSoil testing 590: Nutrient ManagementLitter application based on soil test P results 590: Nutrient ManagementReduced litter applications 590: Nutrient ManagementUse of legumes to decrease or offset decreased applications of N in poultry litter 512: Forage and Biomass PlantingRotational or controlled grazing 528: Prescribed GrazingAmendment of poultry litter with alum to decrease ammonia volatilization and sequester P 591: Amendments for Treatment of Agricultural WasteUse of poultry litter storage or staking sheds 313: Waste Storage FacilityFilter strips and riparian areas 390: Riparian Herbaceous CoverStream fencing to exclude grazing cattle 382: Fence; 578: Stream CrossingConstruction of detention ponds 378: Pond

www.agronomy.org • www.crops.org • www.soils.org 387

told them he would probably abandon all of his practices if subsidies ceased, even though he thought that they were valuable.

Unintentional outcomes to water quality occurred when nutrient management behavior changed relative to placement, timing, and rate due to the adoption of conservation tillage. In the Rock Creek (Ohio) watershed, part of the Lake Erie basin, researchers measured significant reductions in total P found in water resources in the area after considerable adoption of conservation tillage (>60%) (Richards et al., 2009). However, several years later, the trend in dissolved P increased dramatically. These increases were initially attributed to soil P stratification due to conservation tillage, which was believed to have substantially increased surface soil test P, which in turn would escalate dissolved P runoff (Sharpley, 1980). Further investigation suggested that some tillage occurred and that stratification was not as great as expected. Farmers appeared to have changed their nutrient management behavior when they adopted conservation tillage: P applications moved away from the spring to the fall (timing), P fertilizer was applied on the surface rather than mixed into the soil (placement), and farmers applied P less often but in greater quantities (rate) (Meals et al., 2012a; Sharpley et al., 2012; Sharpley et al., 2013). Shifting nutrient management behavior, particularly for timing and rate, caused a shift from one pollutant form to another.

For the NIFA CEAP watersheds, nutrient management was successful when there was purposeful implementation of the practice, supported by education, outreach, and significant and constant financial incentives. This required human resources and financial capital. In watersheds where nutrient management plans were written, as they generally were, without significant local input, trusted advisors, and adequate resources, they frequently were unused.

North Carolina Basin and Watershed Agricultural Surveys

Surveying over 5000 agricultural fields required the workload to be spread over different years. The Tar-Pamlico survey occurred in 2004, Jordan survey in 2006, and Neuse survey in 2009. The results presented here focus on reported N and P application rates, soil test data, and farmer nutrient management behavior, although the latter information was collected only for the Neuse River Basin and Jordan Lake watershed fields.

Agricultural areas in the Tar-Pamlico and Neuse River Basins spanned physiographic regions in North Carolina that included the coastal plain and the piedmont. Land use maps (Fig. 2) clearly showed the difference in agricultural systems between these two regions, with the majority of croplands in the coastal plain and pasture or hayland in the piedmont. Neuse survey data indicated that approximately 36% of agricultural fields were in soybeans, 20% in pasture or hay, and the remaining land in other annual crops such as corn and cotton (Osmond and Neas, 2011); this basin is mostly in the coastal plain, but the upper portion is within the piedmont physiographic region. Conversely, 45 to 100%, depending on the county, of agricultural lands were in pasture or hay in the Jordan Lake survey, an area entirely contained in the piedmont physiographic region (Osmond and Neas, 2007). Cropping system differences were important relative to nutrient management decisions and farmer behavior.

Nitrogen requirements within a field and between fields can vary substantially from one year to another (Lory and Scharf, 2003; Mamo et al., 2003; Schmidt et al., 2007). Thus, predicting N requirements in any given year and field is difficult but necessary to reduce N losses (Sogbedji et al., 2000). In North Carolina, N management has been based on realistic yield expectations (RYE) (i.e., the best three yields over a 5-yr time period) for each crop by soil series. Crop N rates are posted by the state Interagency Nutrient Management committee (NCINMC, 2014).

When interviewed, farmers were asked to list the current crop, fertilizer rate, and fertilizer type for each field enumerated; then, the N and P fertilizer rates were calculated. Initially, farmers did not understand that fertilizer meant applied N (usually liquid N as urea-ammonium nitrate); they interpreted the question to mean a complete fertilizer, such as 15–15–15, or a starter fertilizer, such as diammonium phosphate. Most often, the reported corn N rates were suspect. Even after we retrained enumerators to gather all fertilizer data, including additional applications of N, the N fertilizer rates for corn often had to be sent back for further verification. The N rate data for other crops were more reliable because many of these crops have only one or no fertilizer application during the growing season. For example, soybean producers either applied no fertilizer or diammonium phosphate as a source of starter P (Osmond et al., 2006; Osmond and Neas, 2007, 2011). Obtaining accurate N fertilizer application rates, even with trained enumerators and data checks, is not easy or trivial.

To determine overapplication of N, reported N rates were compared with the RYE N rates in the NLEW tool. In the Tar-Pamlico River Basin, N was as often underapplied as overapplied (Table 3). This was almost identical to the pattern in N fertilization observed in the Jordan Lake watershed. In almost half of these counties, the amount of applied N fertilizer was less than what the crops needed based on RYE rates, which are a function of soil mapping unit (Table 3). When applied N rates exceeded RYE-based crop needs, the amount of excess N was generally quite small. The NLEW analysis also did not include pasture, many of which were not fertilized or were underfertilized by approximately 60% of N needs (Osmond and Neas, 2007). If nutrient management plans were implemented in the Jordan Lake watershed, applied N would increase. Overall, the data demonstrated that farmers were not matching crop N needs to N fertilizer rates regardless of whether they had a nutrient management plan; most, however, did not have plans.

Soil sampling is an important tool for nutrient management decision-making relative to P and other nutrients, such as K, calcium, and micronutrients. In the Neuse River Basin, 97% of the fields were soil tested during the past 3 yr, whereas in the Jordan Lake watershed, only 36% of the fields had been tested over the same time period. The mean range of Mehlich 3 soil test P (STP) on agricultural fields in the Neuse River Basin was 135 mg kg-1, with a minimum of 3 mg kg-1 and a maximum of 865 mg kg-1; generally, over 60 mg kg-1 no additional P is recommended (Hardy et al., 2014). Twelve counties had mean STP levels above 100 mg kg-1, reflecting very high soil test ratings, and the other five counties were high. Because dissolved- or sediment-attached P losses increase as STP

388 Journal of Environmental Quality

increases, lowering STP is important for reducing nutrient losses from cropland (Sharpley, 1980). Applied P was similar for fields in the Neuse Basin regardless of P needs (37 kg P ha-1 for soils that did not need P and 39 kg P ha-1 for fields that did). In the Tar-Pamlico River Basin, average STP levels were very high in seven counties, high in four counties, and medium in one county. Over 67% of the fields did not need P applications, yet P application rates were identical (29 kg P ha-1) regardless of whether the soil needed P or not. Finally, in the Jordan Lake watershed, 65% of the agricultural fields had STP levels in the medium or low category; the majority of agricultural land use was pasture based, and farmers reduced costs by minimizing fertilizer application. The average added P to fields in the Jordan Lake watershed was 79 kg P ha-1 for soils testing high or very high but only 27 kg P ha-1 for soils testing low or medium. This discrepancy in P fertilizer application reflected extant cropping systems: low P fertilization of pastures and high P rates for tobacco. Large quantities of P were still being applied to tobacco.

In one county in the Neuse River Basin, an extension agent worked with farmers on nutrient management plans, which enabled the farmers to write their own nutrient management plans. This effort was based on the belief that if farmers had more ownership in the content of the plans they would use them. Nutrient management plans were written for over 16,000  ha. In addition, farmers were required to attend a 1-d nutrient management training session conducted by North Carolina

Cooperative Extension county agents (NCDENR, 2014c), and over 2000 farmers attended. Despite these efforts, the agent reported that none of the farmers was using the plans 2 yr after plan writing and nutrient management training (R. Pleasants, personal communication, 2006).

Using fertilization rates and crop information, we determined if an individual farmer was changing fertilizer rates on different fields for the same crop. In the Neuse River Basin, farmers with multiple fields of the same crop did not vary their fertilization between fields on cotton (92%), corn (73%), flue-cured tobacco (86%), pasture/hay (73%), soybeans (90%), wheat (82%), or miscellaneous crops (81%). In some cases, such as soybeans, fields fertilized identically received no fertilizer; the effective fertilizer application rate was 0 on approximately 50% of the soybean fields (Osmond and Neas, 2011). Similar results were obtained in the Tar-Pamlico basin (Osmond et al., 2006). In the Jordan Lake watershed, 120 producers with multiple fields of the same crop always used the same fertilizer regime, whereas 9 (7%) farmers changed fertilization rates for the same crop depending on the field (Osmond et al., 2013).

Fertilizer decision-making was an essential component of nutrient management regardless of the existence of plans. Fertilizer decisions when no nutrient management plans were available were made by the farmers (60% Neuse and 70% Jordan; Table 4) (Osmond and Neas, 2007, 2011). Twenty percent of the farmers in the Neuse relied on fertilizer dealers to make these decisions, whereas only 10% did so in the Jordan Lake watershed. When no plans were available, the role of extension, NRCS, and paid consultants relative to fertilizer recommendation was low (<5%).

Application of nutrients, although not identical to fertilizer decision-making, followed the same pattern. In Neuse, 50% of the fertilizer was applied by farmers and 36% by fertilizer dealers, whereas in Jordan, farmers applied 69% of the fertilizer and dealers applied 17%. Although N rates did not appear excessive in the regulated basins, many farmers had over a 30-yr supply of P in their soils (Kamprath, 1999). Despite this, P fertilizer additions continued without regard for soils test recommendations or university research that demonstrated even starter P was unnecessary on these soils (Cahill et al., 2013).

Together, these data demonstrated that producers were making many of their fertilization decisions without technical advice and that most were not following nutrient management plans that would prescribe different fertilizer

Table 3. Amount of nitrogen fertilizer applied to all crop acres enumerated by county vs. the amount of nitrogen recommended by North Carolina Interagency Nutrient Management Committee for those crops by county in the Tar-Pamlico River Basin and the Jordan Lake watershed.

County N applied by farmersN needed per North

Carolina nutrient recommendations

———————— kg ————————Tar-Pamlico River Basin

Beaufort 3,925 3,990Edgecombe 5,875 7,808Franklin 5,334 5,985Granville 1,687 2,752Halifax 3,723 4,598Hyde 20,448 7,789Martin 4,406 3,599Nash 4,217 4,928Pamlico 493 240Pitt 6,183 6,690Washington 3,004 2,902Wilson 2,091 3,705

Jordan Lake watershedAlamance 14,614 39,303Caswell 1,871 1,659Chatham 44,286 30,260Forsyth 1,519 4,786Guilford 50,730 125,504Orange 49,191 48,524Randolph 10,325 7,656Rockingham 16,192 36,849Wake 161 103

Table 4. Organization or individual who makes fertilizer decisions on agricultural fields for farmers in the Neuse River Basin or Jordan Lake watershed when no nutrient management plan is used.

Organization or individual

Neuse River Jordan LakeNo.of people % No. of people %

Fertilizer dealer 391 19.7 15 8.5Paid consultant 51 2.6 0 0NRCS 5 0.3 5 2.9Extension 10 0.5 2 1.1Friend/other farmer 15 0.8 5 2.9Self 1192 60.0 123 70.3Other 239 12.3 7 4.0No commercial

fertilizer applied81 4.1 18 10.3

www.agronomy.org • www.crops.org • www.soils.org 389

rates based on soil test results and RYE. The Neuse River Basin survey analysis indicated that nutrient management was not occurring. Due to the high or very high STP, farmers could have discontinued their use of P fertilizer in many fields, particularly in the coastal plain where STP levels were usually high or very high (Cahill et al., 2013). Nitrogen did not appear to be excessively applied in either physiographic region, but for different reasons. In the coastal plain, mixed cropping systems, which included corn, soybeans, tobacco, cotton, sorghum, and a variety of vegetable crops, predominated. Soybeans had little N applied, and cotton and tobacco used lower levels of N to minimize excess vegetative growth at the end of the season (McCants and Woltz, 1967). The piedmont area was predominantly pasture and hay used for were cow/calf operations. Most of the pastures were underfertilized with N by as much as 60% because the farmers could not afford more N fertilizer. Regardless, whether farmers were applying N or P, fertilizer decision-making did not appear to be based on soil test recommendations, state-approved RYEs, or nutrient management plans.

ConclusionNutrient management is a cornerstone conservation

practice for reducing agricultural nonpoint-source pollution, yet nutrient management plans generally have not translated into on-the-ground, active nutrient management in the NIFA CEAP watersheds or nutrient-impaired watersheds in North Carolina. Comprehensive success will not be realized until public policies and local government assistance are in sync with what drives farmers to adopt nutrient management practices. Information from our key informant survey and individual project information suggested several reasons for lack of participation: (i) farmers were found to be hesitant to apply N at university recommended rates because they did not trust the recommendations, (ii) farmers viewed abundant N as insurance, (iii) or used recommendations made by fertilizer dealers. Exceptions were found when watershed education, technical support, and funding resources focused around nutrient management that included easing management time, actively and consistently working directly with a small group of the same farmers, and substantial resource allocations.

A significant disconnect exists between farmer behavior relative to nutrient management and its importance as the first line of defense in reducing agricultural nonpoint-source pollution. Market forces on profits limit adoption, but failures of current policies to promote greater nutrient management are also a function of programs failing to understand and address social and cultural forces as identified in this paper. Understanding motivations for better nutrient use will require detailed key informant interviews that focus solely on farmer nutrient management decision-making.

AcknowledgmentsThe North Carolina watershed studies were funded by three separate USEPA grants as pass-through funds to the NC Department of Environment and Natural Resources. The NIFA-CEAP watershed study synthesis was funded by USDA NIFA and NRCS (Agreement No. 2007-51130-18575). This work resulted from a conference supported by NSF Research Coordination Network award DEB-1049744 and by

the Soil Science Society of America, the American Geophysical Union, The International Plant Nutrition Institute, The Fertilizer Institute, and the International Nitrogen Initiative.

ReferencesAlexander, R.B., R.A. Smith, G.E. Schwarz, E.W. Boyer, J.V. Nolan, and J.W.

Brakebill. 2008. Differences in phosphorus and nitrogen delivery to the Gulf of Mexico from the Mississippi River Basin. Environ. Sci. Technol. 42:822–830. doi:10.1021/es0716103

Bosch, D.J., Z.L. Cook, and K.O. Fuglie. 1995. Voluntary versus mandatory agricultural policies to protect water quality: Adoption of nitrogen testing in Nebraska. Rev. Agric. Econ. 17:13–24. doi:10.2307/1349651

Cahill, S., R. Gehl D. Osmond, and D. Hardy. 2013. Evaluation of an organic copolymer fertilizer additive on phosphorus starter fertilizer response by corn. Crop Manage. 12. doi:10.1094/CM-2013-0322-01-RS

Charlier, R.H., P. Mor, C.W. Finkl, and A. Thys. 2007. Green tides on the Brittany coasts. Environ. Res. Eng. Manage. 41:52–59.

Cooper, S.R. 1995. Chesapeake Bay watershed historical land use: Impact on water quality and diatom communities. Ecol. Appl. 5:703–723. doi:10.2307/1941979

Diaz, R.J., and R. Rosenberg. 2008. Spreading dead zones and consequences for marine ecosystems. Science 321:926–929. doi:10.1126/science.1156401

Duriancik, L.F., D. Bucks, J.P. Dobrowolski, T. Drewes, S.D. Eckles, L. Jolley, R.L. Kellogg, D. Lund, J.R. Makuch, M.P. O’Neill, C.A. Rewa, M.R. Wallbridge, R. Parry, and M.A. Weltz. 2008. The first five years of the Conservation Effects Assessment Project. J. Soil Water Conserv. 63:185A–197A. doi:10.2489/jswc.63.6.185A

Exner, M.D., H. Perea-Estrada, and R.F. Spalding. 2010. Long-term response of groundwater nitrate concentrations to management regulations in Nebraska’s Central Platte Valley. Sci. World J. 10:286–297. doi:10.1100/tsw.2010.25

Goolsby, D.A. 2000. Mississippi Basin nitrogen flux believed to cause Gulf hypoxia. Eos Trans. AGU 81:321–327. doi:10.1029/00EO00244

Hardy, D.H., M.R. Tucker, and C.E. Stokes. 2014. Crop fertilization based on North Carolina soil tests: Circular 1. North Carolina Department of Agriculture, Raleigh, NC. http://www.ncagr.gov/agronomi/pdffiles/obook.pdf (accessed 19 Jan. 2014).

Hoag, D.L.K., I. Chaubey J. Popp M. Gitau L. Chiang J. Pennington G. Rodriguez E. Gbur M. Nelson, and A.N. Sharpley. 2012a. Lincoln Lake Watershed, Arkansas: National Institute of Food and Agriculture–Conservation Effects Assessment Project. In: D. Osmond et al., editors, How to build better agricultural conservation programs to protect water quality: The National Institute of Food and Agriculture-Conservation Effects Assessment Project experience. Soil and Water Conservation Society, Ankeny, IA. p. 171–186.

Hoag, D., A. Luloff, and D. Osmond. 2012b. Socioeconomic analysis: National Institute of Food and Agriculture–Conservation Effects Assessment Project. In: D. Osmond et al., editors, How to build better agricultural conservation programs to protect water quality: The National Institute of Food and Agriculture-Conservation Effects Assessment Project experience. Soil and Water Conservation Society, Ankeny, IA. p. 120–137.

Jaynes, D.B., D.L. Dinnes, D.W. Meek, D.L. Karlen, C.A. Cambardella and T.S. Colvin. 2004. Using the late spring nitrate test to reduce nitrate loss within a watershed. J. Environ. Qual. 23:669–677. doi:10.2134/jeq2004.6690

Kamprath, E.J. 1999. Changes in phosphate availability of ultisols with long‐term cropping. Commun. Soil Sci. Plant Anal. 30:909–919. doi:10.1080/00103629909370256

Lory, J.A., and P.C. Scharf. 2003. Yield goal versus delta yield for predicting fertilizer nitrogen need in corn. Agron. J. 95:994–999. doi:10.2134/agronj2003.0994

Luloff, A.E., D.L.K. Hoag D.L. Osmond B.R. Woods J.S. Gordon J. Gruver K. Roka C.M. Raboanarielina C. Longmire M. Ward, and J.L. Weigle. 2012. Key informant survey to understand what farmers, agency personnel, and stakeholders think: National Institute of Food and Agriculture–Conservation Effects Assessment Project. In: D. Osmond et al., editors, How to build better agricultural conservation programs to protect water quality: The National Institute of Food and Agriculture-Conservation Effects Assessment Project experience. Soil and Water Conservation Society, Ankeny, IA. p. 12–35.

Mamo, M., G.L. Malzer, D.J. Mulla, D.R. Huggins, and J. Strock. 2003. Spatial and temporal variation in economically optimum nitrogen rate for corn. Agron. J. 95:958–964. doi:10.2134/agronj2003.0958

McCants, C. B., and W.G. Woltz. 1967. Growth and mineral nutrition of tobacco. Adv. Agron. 19:222–232.

390 Journal of Environmental Quality

Meals, D.W., P. Richards R. Confesor K. Czajkowski J. Bonnell D.L. Osmond D.L.K. Hoag J. Spooner, and M.L. McFarland. 2012a. Rock Creek Watershed, Ohio: National Institute of Food and Agriculture–Conservation Effects Assessment Project. In: D. Osmond et al., editors, How to build better agricultural conservation programs to protect water quality: The National Institute of Food and Agriculture-Conservation Effects Assessment Project experience. Soil and Water Conservation Society, Ankeny, IA. p. 316–326.

Meals, D.W., R. Spalding M. Exner R.B. Ferguson M.L. McFarland D.L. Osmond, and J. Spooner. 2012b. Groundwater beneath a Phase III Management Area in the Central Platte Natural Resources District, Nebraska: National Institute of Food and Agriculture–Conservation Effects Assessment Project. In: D. Osmond et al., editors, How to build better agricultural conservation programs to protect water quality: The National Institute of Food and Agriculture-Conservation Effects Assessment Project experience. Soil and Water Conservation Society, Ankeny, IA. p. 286–295.

Monroe, J., and A.L. Finkner. 1959. Handbook of area sampling. Chilton, Philadelphia, New York.

NCDENR. 2014a. Falls Lake nutrient strategy. North Carolina Department of Environment and Natural Resources, Division of Water Resources. http://portal.ncdenr.org/web/fallslake/home (accessed 19 Jan. 2014).

NCDENR. 2014b. Jordan nutrient strategy. North Carolina Department of Environment and Natural Resources, Division of Water Resources. http://portal.ncdenr.org/web/jordanlake/home (accessed 19 Jan. 2014).

NCDENR. 2014c. Neuse nutrient strategy. North Carolina Department of Environment and Natural Resources, Division of Water Resources. http://portal.ncdenr.org/web/wq/ps/nps/neuse (accessed 19 Jan. 2014).

NCDENR. 2014d. Tar-Pamlico nutrient strategy. North Carolina Department of Environment and Natural Resources, Division of Water Resources. http://portal.ncdenr.org/web/wq/ps/nps/tarpamlico (accessed 19 Jan. 2014).

NCINMC. 2014. Realistic yield expectations for North Carolina. North Carolina Interagency Nutrient Management Committee, Raleigh, NC. http://nutrients.soil.ncsu.edu/yields/ (accessed 15 Feb. 2014).

Osmond, D.L., K. Cassel, and K. Neas. 2006. Delineating agriculture in the Tar-Pamlico River Basin. Final report to NCDENR, Division of Water Quality for USEPA 319 program, Raleigh, NC. http://www.soil.ncsu.edu/publications/river/TarPam_NCANAT_FinalRpt_Jan30_06.pdf (accessed 19 Jan. 2014).

Osmond, D., D. Meals M. Arabi, and D. Hoag. 2012. Cannonsville Reservoir, NY: National Institute of Food and Agriculture–Conservation Effects Assessment Project. In: D. Osmond et al., editors, How to build better agricultural conservation programs to protect water quality: The National Institute of Food and Agriculture-Conservation Effects Assessment Project experience. Soil and Water Conservation Society, Ankeny, IA. p. 120–137.

Osmond, D.L., and K. Neas. 2007. Delineating agriculture in the Lake Jordan River Basin. Final report to NCDENR, Division of Water Quality for USEPA 319 program, Raleigh, NC. http://www.soil.ncsu.edu/publications/river/Cape_Fear_NCANAT_FinalRpt_11_9_07.pdf (accessed 19 Jan. 2014).

Osmond, D.L., and K. Neas. 2011. Delineating agriculture in the Neuse River Basin. Final report to NCDENR, Division of Water Quality for USEPA 319 program, Raleigh, NC. http://www.soil.ncsu.edu/publications/river/Neuse_Report.Final.11.29.2011.pdf (accessed 19 Jan. 2014).

Osmond, D.L., K. Neas, A.M. Johnson, and S.L. Cahill. 2013. Fertilizer use in regulated river basins: Is it what we think? J. Contemp. Water Res. Educ. 151:20–26. doi:10.1111/j.1936-704X.2013.03148.x

Osmond, D.L., L. Xu, N.N. Ranells, S.C. Hodges, R. Hansard, and S.H. Pratt. 2001. Nitrogen Loss Estimation Worksheet (NLEW): An agricultural nitrogen loading reduction tracking tool. In: J. Galloway, E. Cowling, J. Erisman, J. Wisniewski, C. Jordan, editors, Optimizing nitrogen management in food and energy production and environmental protection. Contributed papers from the 2nd International Nitrogen Conference. 14–18 Oct. 2001, Potomac, MD. p. 777–783.

Paerl, H.W., J.L. Pinckney, J.M. Fear, and B.L. Peierls. 1998. Ecosystem responses to internal and watershed organic matter loading; consequences for hypoxia in the eutrophying Neuse River Estuary, North Carolina, USA. Mar. Ecol. Prog. Ser. 166:17–25. doi:10.3354/meps166017

Preston, S.D., and J.W. Brakebill. 1999. Application of spatially referenced regression modeling for the evaluation of total nitrogen loading in the Chesapeake Bay watershed. USGS Water-Resources Investigations Report 99-4054. USGS, Reston, VA.

Prokopy, L.S., K. Floress, D. Kllatthar-Weinkauf, and B. Baumgart-Getz. 2008. Determinants of agricultural best management practice adoption: Evidence from the literature. J. Soil Water Conserv. 63:300–311. doi:10.2489/jswc.63.5.300

Rabalais, N.N., and R.E. Turner. 2001. Hypoxia in the northern Gulf of Mexico: Description, causes and change. Coastal Estuarine Stud. 58:1–36. doi:10.1029/CE058p0001

Randall, G.W., and J.A. Vetsch. 2005. Nitrate losses in subsurface drainage from a corn–soybean rotation as affected by fall and spring application of nitrogen and nitrapyrin. J. Environ. Qual. 34:590–597. doi:10.2134/jeq2005.0590

Reimer, A.P., A.W. Thompson, and L.S. Prokopy. 2011. The multi-dimensional nature of environmental attitudes among farmers in Indiana: Implications for conservation adoption. Agric. Human Values 29:29–40. doi:10.1007/s10460-011-9308-z

Richards, R.P., D.B. Baker, and J.P. Crumrine. 2009. Improved water quality in Ohio tributaries to Lake Erie: A consequence of conservation practices. J. Soil Water Conserv. 64(3):200–211. doi:10.2489/jswc.64.3.200

Rupert, M.G. 2008. Decadal-scale changes of nitrate in ground water of the United States, 1988–2004. J. Environ. Qual. 37:S240–S248. doi:10.2134/jeq2007.0055

Schilling, K.E., and R.D. Libra. 2000. The relationship of nitrate concentrations in streams to row crop land use in Iowa. J. Environ. Qual. 29:1846–1851. doi:10.2134/jeq2000.00472425002900060016x

Schmidt, J.P., N. Hong, A. Dellinger, D.B. Beegle, and H. Lin. 2007. Hillslope variability in corn response to nitrogen linked to in-season soil moisture redistribution. Agron. J. 99:229–237. doi:10.2134/agronj2006.0187

Sharpley, A.N. 1980. The enrichment of soil-phosphorus in runoff sediments. J. Environ. Qual. 9:521–526. doi:10.2134/jeq1980.00472425000900030039x

Sharpley, A., P. Richards, S. Herron, and D. Baker. 2012. Case study comparison between litigated and voluntary nutrient management strategies. J. Soil Water Conserv. 67:442–450. doi:10.2489/jswc.67.5.442

Sharpley, A., H.P. Jarvie, A. Buda, L. May, B. Spears, and P. Kleinman. 2013. Phosphorus legacy: Overcoming the effects of past management practices to mitigate future water quality impairment. J. Environ. Qual. 42:1308–1326. doi:10.2134/jeq2013.03.0098

Sogbedji, J.M., H.M. van Es, C.L. Yang, L.D. Geohring, and F.R. Magdoff. 2000. Nitrate leaching and nitrogen budget as affected by maize nitrogen rate and soil type. J. Environ. Qual. 29:1813–1820. doi:10.2134/jeq2000.00472425002900060011x

Stuart, D., R.L. Schewe, and M. McDermott. 2014. Reducing nitrogen fertilizer application as a climate change mitigation strategy: Understanding farmer decision-making and potential barriers to change in the US. Land Use Policy 36:210–218. doi:10.1016/j.landusepol.2013.08.011

USDA–ARS. 2014. Revised Universal Soil Loss Equation (RUSLE): Welcome to Version 1 and 2. USDA–ARS. http://www.ars.usda.gov/Research/docs.htm?docid=5971 (accessed 6 Apr. 2014).

White, M.J., C. Santhi, N. Kannan, J.G. Arnold, D. Harmel, L. Norfleet, P. Allen, M. DiLusio, X. Want, J. Atwood, E. Haney, and M. Vaughn Johnson. 2014. Nutrient delivery from the Mississippi River to the Gulf of Mexico and effects of cropland conservation. J. Soil Water Conserv. 69:26–40. doi:10.2489/jswc.69.1.26

Woods, B.R., A.E. Luloff, D.L. Osmond, and D.L. Hoag. 2014. Towards a synthesis: Lessons from thirteen cropland watershed-scale studies. Soc. Nat. Resour. 27:341–357. doi:10.1080/08941920.2013.861551