HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Cabot, P. E. Articles by Nowak, P. Search for Related Content PubMed PubMed Citation Articles by Cabot, P. E. Articles by Nowak, P. Agricola Articles by Cabot, P. E. Articles by Nowak, P. Related Collections Animal Waste Nutrient Management Phosphorus

ENVIRONMENTAL ISSUES

Planned versus Actual Outcomes As a Result of Animal Feeding Operation Decisions for Managing Phosphorus

^a Department of Biological Systems Engineering and Nelson Institute for Environmental Studies, University of Wisconsin-Madison, 420 Agriculture Hall, 1450 Linden Drive, Madison, WI 53706
^b Department of Rural Sociology and Nelson Institute for Environmental Studies, University of Wisconsin-Madison, 420 Agriculture Hall, 1450 Linden Drive, Madison, WI 53706

^* Corresponding author (pnowak{at}facstaff.wisc.edu)

Received for publication December 26, 2002.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
The paper explores how decisions made on animal feeding operations (AFOs) influence the management of manure and phosphorus. Variability among these decisions from operation to operation and from field to field can influence the validity of nutrient loss risk assessments. These assessments are based on assumptions that the decision outcomes regarding manure distribution will occur as they are planned. The discrepancy between planned versus actual outcomes in phosphorus management was explored on nine AFOs managing a contiguous set of 210 fields in south-central Wisconsin. A total of 2611 soil samples were collected and multiple interviews conducted to assign phosphorus index (PI) ratings to the fields. Spearman's rank correlation coefficients (r_S) indicated that PI ratings were less sensitive to soil test phosphorus (STP) levels (r_S = 0.378), universal soil loss equation (USLE) (r_S = 0.261), ratings for chemical fertilizer application (r_S = 0.185), and runoff class (r_S = –0.089), and more sensitive to ratings for manure application (r_S = 0.854). One-way ANOVA indicated that mean field STP levels were more homogenous than field PI ratings between AFOs. Kolmogorov–Smirnov (K–S) tests displayed several nonsignificant comparisons for cumulative distribution functions, S(x), of mean STP levels on AFO fields. On the other hand, the K–S tests of S(x) for PI ratings indicated that the majority of these S(x) functions were significantly different between AFOs at or greater than the 0.05 significance level. Interviews suggested multiple reasons for divergence between planned and actual outcomes in managing phosphorus, and that this divergence arises at the strategic, tactical, and operational levels of decision-making.

Abbreviations: AFO, animal feeding operation • AU, animal unit • CNMP, comprehensive nutrient management plan • FS, farm system • K–S, Kolmogorov–Smirnov • PI, phosphorus index • r_S, Spearman's rank correlation coefficients • STP, soil test phosphorus • S(x), cumulative distribution function • USLE, universal soil loss equation

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
THE USDA AND USEPA (1999) support the adoption of comprehensive nutrient management plans (CNMPs) to reduce the potential for nonpoint-source pollution from AFOs. These operations constitute a sector of agricultural land uses identified as sources of excessive nutrients in eutrophic water bodies (USEPA, 1998, 2000). Focusing on those operations with the largest herd sizes, the Unified National Strategy for Animal Feeding Operations mandates that approximately 5% of AFOs must adopt CNMPs, while the remaining 95% of AFOs are encouraged to voluntarily adopt CNMPs by 2009 (USDA and USEPA, 1999). The regulatory provision of the Strategy primarily affects AFOs managing more than 1000 animal units (AUs), but special cases may also designate smaller AFOs as environmental threats.

This policy of using traditional thresholds on AU numbers or densities to define the regulatory identity of AFOs is preserved from the Clean Water Act (CWA) of 1977. The use of these thresholds, however, has come under scrutiny in recent years. For instance, some critics speculate that operations housing fewer animals, but using antiquated technology and inefficient management techniques, may be comparable with larger operations in terms of their overall environmental impact (Norris and Seidl, 2001). Others support this argument with findings that indicate larger operations are more likely to comply with recommended manure management practices (Jackson-Smith and Nevius, 1998). This compliance is probably driven by the threat of fines or punitive actions or possibly more efficient economies of scale in manure storage and distribution (Jackson-Smith et al., 1995). Although the policy implications of understanding the relative contributions of AFOs to nonpoint-source pollution are extensive, this issue has not been thoroughly and systematically researched to date.

Quantifying the environmental impacts of livestock operations is complex because of the fundamental difference between point- and nonpoint-source pollution (Fig. 1) . Distribution of manure on the landscape alters the process of pollutant delivery from a relatively simple point-source problem to a complicated situation involving hydrologic, biophysical, and management variables. The possible point sources from AFOs are typically addressed through engineered remedies that are promoted as best management practices (BMPs) or best available control technologies (BACTs). These remedial practices are designed to collect and store manure somewhere near the animal confinement facilities. Solutions that focus on storage and collection are advantageous for two reasons. First, they provide more flexibility to producers, regardless of operation size, to plan the timing and location of manure distribution. Second, the relationship between environmental risk and herd size is effectively neutralized by designing the storage and collection system to accommodate the volume of manure generated. Nevertheless, manure must still be distributed on the landscape, and this distribution phase of the management process is a more complicated situation to control.

View larger version (30K): [in this window] [in a new window]   Fig. 1. Point and nonpoint-source pathways of P delivery. (1) Direct discharge from the barn to waterbody. (2) Direct and concentrated runoff from barnyard to waterbody. (3) Indirect and diffuse discharge from cropland to waterbody mediated by farmer or operator.

The potential divergence between planned and actual outcomes from manure distribution decisions is important with respect to the accuracy of risk assessments for nonpoint-source pollution from agricultural fields. This divergence emphasizes the variability among animal feeding operations and their decisions regarding the location, rate, and timing of manure placement (Shepard, 2000). The differences between planned and actual outcomes in manure distribution are often overlooked in CNMPs, which instead tend to focus on manure generation and storage. Furthermore, whether at the scale of an individual farm or an entire watershed, unforeseen nutrient losses during the actual manure distribution phase may be significant enough to nullify many of the planned reductions in pollution based on engineered BMPs or BACTs. We acknowledge that this distribution phase involves a "human dimension" that may complicate assessments of nonpoint-source pollution and CNMPs for mitigating this problem. This paper attempts to illustrate, however, that these assessments and plans will be more viable if they encompass the generation, storage, and distribution phases of manure management.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Study Site and Farm Characteristics
Data were gathered in a south-central Wisconsin watershed where AFOs comparable with those found in other parts of Wisconsin and the Midwest Lake States exist. This watershed is located in a county that is currently at 25 to 50% of its assimilative capacity for P (Gollehon et al., 2001). Information on the total cropland area, herd size (AUs), livestock species, and AU density (AU ha^–1) was collected (Table 1). A code is used to maintain operator confidentiality. All information was obtained directly from the nine producers who operate contiguous farms within a common watershed boundary.

The nine AFOs manage a total of 210 fields, which were geographically mapped using Trimble (Sunnyvale, CA) global positioning system technology. Field boundaries were georeferenced in the Lambert Conformal Conic projection and stored in a geographic information system using ArcView 3.3 GIS (ESRI, 2002).

The following method published by Beegle et al. (1997) was used to calculate the number of AUs managed:

[1]

The following relationship was then used to calculate the animal to land ratio or AU density for each operation:

[2]

For six of the operations, all fields and consequently all cropland managed by the operator were delineated in the GIS and used to calculate AU density. Three of the larger AFOs (FS-1/3, FS-2/4, and FS-8/9, where FS is farm system) operated fields that were outside of the watershed boundary and away from the main animal feeding location. For these three AFOs, the effective cropland base was determined from conservation plans and Code 590 Nutrient Management Plans, which were provided by the producer and verified with the county land conservation department. Pastures and fields enrolled in the Conservation Reserve Program were not included in the cropland base, but these fields accounted for an insignificant fraction of the total field area.

Soil Sampling
Soils were sampled and analyzed to assess nonpoint-source pollution risk from the 210 fields managed by AFOs in this study area. Risk was assessed on a P basis for three reasons. First, agronomic soil tests for this nutrient are relatively inexpensive, so a large number of fields such as is used for this paper could be accommodated. Second, public concerns regarding waterbody eutrophication (Parry, 1998; Sharpley et al., 2000) have initiated a nationwide program to develop indexing tools for assigning P loss risk to agricultural fields. Although agricultural runoff contains relatively low P concentrations compared with soil or manure, these concentrations regularly exceed critical P levels (20–100 µg L^–1) associated with water quality impairment (Correll, 1998). Third, the manure application rating within these indexing tools is an example of a planned decision and outcome relating to the overall distribution phase of manure management by AFOs.

Sample points were located using FarmGPS software Version 1.55 (Red Hen Systems, 1999) and an unaligned systematic sampling protocol (Wollenhaupt et al., 1994). Soil samples were collected in September and October of 2000. A total of 1013 base samples were collected every 1 ha (2.5 acres) across all 210 fields. One out of every five sample points served as the base for a stratified transect sample. The primary transect direction was randomly selected from the four principal points of the compass (north, east, west, south), and samples were taken at 5, 10, 20, and 40 m from the base. A second transect was then taken 60° clockwise from the base transect to form a V shape. The purpose of extending the transects in two directions was to capture spatial variability at small scales, thereby preventing samples with extremely high localized STP levels from biasing the mean STP value for the field. This method also improved the sampling of fields that were strip-cropped. Finally, because the 1 ha (100 x 100 m) grid is typical in nutrient management plans, the use of the transect data allowed for better stratification. The number of transect samples was 1598, bringing the total number of samples to 2611. The density of soil sampling varied from one sample per field on three small fields (<2.75 ha) that had no transects to 96 samples on a 14.92-ha field with eight transects.

Each sample consisted of eight cores that were collected within a circle (1.5-m radius) centered on each sample point. The samples were taken at a depth of 5 cm to represent the effective depth of runoff–soil interaction from which most P is delivered (Sharpley et al., 1996). The soil was then air-dried and analyzed for extractable P using the Bray-P1 method (Bray and Kurtz, 1945), commonly used in Wisconsin for agronomic and soil testing guidelines (Kelling et al., 1998).

Soil and Management Data
The soil P status of each field was evaluated with two of the measures specified in NRCS Code 590 for P-based management where manure, organic by-products, or fertilizer are used: (i) the level of P in the soil based on an agronomic soil test, and (ii) a phosphorus index (PI) rating using integrated site characteristics (Natural Resources Conservation Service, 2002).

Soil test P levels are recognized as an option specified in NRCS Code 590 for evaluating the need for P-based management, but these levels alone cannot predict P losses from agricultural fields (Sharpley et al., 1996). For this reason, a PI attempts to account for other variables that may also control off-site P migration. Because PI ratings vary from state to state (Sharpley et al., 2003), and had not been finalized specifically for Wisconsin at the time of this study, the generalized PI (Table 2; Lemunyon and Gilbert, 1993) was used to calculate the nutrient loss risk for each field. To calculate a PI for these fields, soil erosion potential (Mg ha^–1) was estimated from the universal soil loss equation (USLE) developed by Wischmeier and Smith (1978) using a digital elevation model with 10-m resolution (LS factor) and cropping and conservation information (C and P factors) obtained during field visits. Also used was the runoff class obtained from the local land conservation department and mean Bray-P1 (mg kg^–1) from the soil analysis. Lastly, P fertilizer and manure management information was obtained from the producers.

View this table: [in this window] [in a new window]   Table 2. The phosphorus indexing method (adapted from Lemunyon and Gilbert, 1993).

Producers were able to easily provide total field area and herd sizes, but many lacked basic information regarding manure application rates, manure spreader calibration, and distribution locations. In such cases, application rates were estimated based on the size of the spreader and how much area the producer believed was required to place the manure on a daily basis. Most of them viewed the manure distribution process as a cumbersome but necessary chore in the business of managing animals. During subsequent farm visits and discussions with these producers, more information was acquired on management decisions and aspects of manure distribution.

Soil P and management data for all fields are given in Table 3, which summarizes the cropping system, management characteristics, mean STP levels (0–5 cm), USLE, and PI ratings for all 210 fields in the study area. Not shown in Table 3 are fields that were in pasture or enrolled in the Conservation Reserve Program, or where only one soil sample had been taken making STP calculations unreliable.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Soil Phosphorus Levels and Phosphorus Index Ratings
The field-level PI ratings tended to be most sensitive to the manure management category rating. On the basis of Spearman's rank correlation (r_S), PI ratings overall were moderately sensitive to the category ratings assigned to STP levels (r_S = 0.378) and USLE (r_S = 0.261). For these operations, chemical fertilizer applications (r_S = 0.185) did not influence the PI rating strongly. Runoff class (r_S = –0.089) was also not a factor. The PI ratings were the most sensitive to the category rating for manure application (r_S = 0.854). These correlations suggest that site vulnerability as quantified by this particular indexing approach is more often a function of the ratings assigned to manure management decisions, rather than STP levels at the scale of the field. In most cases, fields with higher PI ratings were associated with operations that engage in at least two of the following practices: broadcast manure distribution, high-rate fertilizer application, or high-rate manure application.

Mean STP levels were also more homogenous on a farm-wide basis than PI ratings (Table 4), as indicated by the one-way ANOVA, which did not reject the null hypothesis of equal mean between groups (p = 0.083). The Levene homogeneity-of-variances test also did not reject the null hypothesis of equal variances for STP levels between the AFOs (p = 0.266). These differences are emphasized by stem-and-leaf plots for STP level and PI rating (Fig. 2) . For the PI ratings, on the other hand, the null hypothesis of equal means was also rejected (p < 0.001), but it should also be noted that the Levene homogeneity-of-variances test rejected the null hypothesis of equal variance (p < 0.001).

View larger version (26K): [in this window] [in a new window]   Fig. 2. Error plots of mean (a) soil test phosphorus (STP) levels and (b) phosphorus index (PI) ratings on 210 fields managed by nine animal feeding operations (AFOs) in south-central Wisconsin.

It was also evident that between-group variation was minimal for mean field STP levels, but significant for PI ratings. A Kolmogorov–Smirnov (K–S) test displayed broad nonsignificant comparisons for mean STP levels on the fields managed by each operation (Table 5). The S(x) functions for STP levels are shown in Fig. 3 , illustrated for each AFO. Generalizing from the K–S tests, it can be stated that all nine AFOs manage fields with a similar range of mean STP levels. All operations, for instance, consisted of fields ranging from P-deficient to P-enriched. This may be partly due to a similarity of landscape features, rolling topography, and soil characteristics (e.g., pH and P sorption capacity) across the study area. These similarities can affect STP levels through hydrologic processes regardless of the size or density of the animal feeding operation. The homogeneity of STP levels may also be partly due to the fact that operators use similar decision metrics in distributing manure to these fields. For example, it is not unusual to find elevated STP levels closer to the locations where animals are housed.

View larger version (58K): [in this window] [in a new window]   Fig. 3. Cumulative distribution functions, S(x), of soil test phosphorus (STP) levels for 210 fields managed by nine animal feeding operations (AFOs) in south-central Wisconsin.

View larger version (56K): [in this window] [in a new window]   Fig. 4. Cumulative distribution functions, S(x), of phosphorus index (PI) ratings for 210 fields managed by nine animal feeding operations (AFOs) in south-central Wisconsin.

From the S(x) functions shown in Fig. 4, it appears that some operations (FS-1/3, FS-2/4, FS-5/7, FS-6/5) have fields that are grouped by high and low PI ratings, while others manage fields with smoother distributions of PI ratings. These groupings are not unique, however, to the number of animals managed or the animal density for the operation. Furthermore, fields with elevated PI ratings were not more likely to be found on operations with larger herds or higher densities.

Planned versus Actual Manure Applications
The preceding discussion illustrates that PI ratings at the field level are generally more sensitive to management decisions than are STP levels. The significance of the management rating may seem a promising route to address the environmental risk of AFOs by attempting to control management practices, but our interviews suggested that there is a divergence between planned and actual outcomes in managing manure. These decisions are simplified when manure distribution practices are quantified on the basis of what rate or method a producer may plan to use when applying manure to a given field. There is some difficulty in quantifying these practices, however, because of the gap between what producers may be planning to accomplish when distributing manure versus the actual outcomes of their decisions.

The difficulty in quantifying the distribution phase of manure management is evident from the application rate data provided by the producers. In the case of the smaller producers, these application rates had to be calculated from anecdotal heuristics that each producer used in their system of managing manure. One of the larger producers (FS-1/3) could not provide specific manure application rates for their liquid spreader, but instead told us that their tankers (4800 and 5000 gallons, or approximately 18200 and 18900 L, respectively) were used to apply manure at "two loads per acre." The manure application rates for all fields (provided by the producers or estimated from their best ability to quantify these rates) were then used to calculate the total P that would be applied to cropland, based on the total P in manure estimated from MidWest Plan Service (2000) as summarized in Table 7. Although we did not sample all the fields of the two largest AFOs (FS-1/3, FS-2/4) due to their location outside the study watershed, these operators provided us with their nutrient management plans and total field area, so a full estimate of overall P application was still possible. Table 7 also includes an approximation of the total P generated on each farm, based on manure nutrient content estimates (MidWest Plan Service, 2000). What should be obvious from the table are the significant differences between the quantities of P generated versus the estimated quantities applied.

Strategic Decisions
A logical recommendation to lower the STP levels on various fields may be for the operation to acquire more land on which to place manure to reduce the density of the animal feeding operation. This solution was not feasible for the farms in this study area because of urban pressures, zoning ordinances, and travel distances to available land. Moreover, shifting to a lower density of animals as a means of reducing the risk of P loss on individual fields is based on the implicit assumption that manure application rates are uniform across all fields managed by the operation. Producers in this study area indicated, however, that the amount of land actually used for manure distribution frequently differed from the amount of land potentially available or allocated in the nutrient management plan. Others have reported similar observations for different study areas in Wisconsin (Program on Agricultural Technology Studies, 1997; Jackson-Smith and Nevius, 1998; Saam et al., 2005). These differences were attributed to three reasons. First, the fragmented pattern of land ownership and rental in this area forces the producers to haul manure greater distances on local roads with increased volumes of commuter traffic. It is not simply a matter of distance, but the degree of impedance on these roads that can force the decision to distribute manure elsewhere (Cabot et al., 2004). Second, some fields may receive less manure simply because of field conditions at the time when manure must be applied. Concerns regarding long-term implications of soil compaction were cited by our producers as a reason why manure was not placed at certain locations at certain times in the crop cycle. Finally, in some cases, producers also reported that spring weight restrictions on local roads prevented hauling manure to more distant fields, and this temporarily reduced the amount of land area they could access for distribution. Consequently, this group of producers made strategic manure management decisions stemming from constraints at different spatial and temporal scales, from long-term tenure relations to daily traffic patterns, and from avoiding portions of their fields to losing entire fields to suburban development.

Tactical Decisions
Several factors act at the tactical level of management to create differences between the application goals stated by producers versus what they may have actually accomplished. The first and most prevalent example of this involves the land rental arrangements that many producers use to maintain an adequate land base. It was common for most of the producers interviewed to develop informal and formal agreements with local area landowners to rent land for crop production. These parcels also receive manure from the renting operation. In the area where these data were gathered, the lease lengths of these arrangements were becoming shorter in certain cases because of competition between developers and producers for rented parcels. The ratio of owned-to-operated land, for instance, ranged from 0.75 to 1.00 for the small and mid-sized producers, but was much smaller for the larger operations (FS-1/3 = 0.49, FS-2/4 = 0.27). Interviews revealed that these larger operations placed a high value on cooperative relationships with neighbors who may be active or retired producers and will allow them access to land for crop production and manure spreading. This dynamic land rental situation may make nutrient management planning and manure distribution decision increasingly difficult from year to year.

Second, many operators stated that seasonal labor and time limitations (i.e., planting and harvesting periods) often affected the manure distribution process. This limitation made some producers reluctant to formally develop and implement a nutrient management plan as they felt these plans did not recognize the episodic nature of labor demands on their farm. The operators of larger AFOs also indicated that they have little control over scheduling the services of a custom hauler as many AFOs with storage request these services during similar periods (preplant and postharvest). Producers hauling manure daily also stated that their decision to spread near their barns was frequently a result of shortages of on-farm labor and/or time. This is consistent with the findings of Dittrich (1993), who performed an analysis of nutrient management guidelines for this area and found that these recommendations tended to underemphasize the nutrients available in manure and thereby indirectly supported the "dumping" of manure in a time-efficient manner. Comments by producers also indicated that management decisions may have to change in response to annual weather patterns. Producers stated that "bad" weather or climatic conditions influenced their decision to overapply on closer fields while avoiding more distant fields. The above examples of tactical level management factors are not commonly incorporated within rationally based nutrient plans that are designed to balance nutrient inputs and exports. Nonetheless, these factors can influence the actual distribution of manures on the agricultural landscape as producers make decisions that are not necessarily anticipated when a plan is written.

Operational Decisions
It was also evident that the operational level of decision-making may also have been a source of divergence between planned and actual outcomes in manure distribution. Based on the interviews, it seemed likely that operational or "day-to-day" decisions would make it difficult for producers to accurately report their own management decisions. First, the preference to haul only full loads of manure to their fields was cited by producers as a fundamental attitude toward managing manure. Further discussion with producers revealed how this attitude could easily result in overapplication of manure on portions of a field. The area of a field (A) is not always uniformly divisible by the number of loads (x) delivered by the spreader at the rate recommended by the nutrient management plan. A fully loaded manure spreader (solid, liquid, or both), for instance, could technically cover A by applying x loads at a planned rate (r). To apply a discreet number of spreader loads, a field should therefore accommodate A/x loads when applying at rate r. Problems arise for the producer when the actual r, which is governed by the speed of the tractor or the consistency of the manure, is not commensurate with the nutrient management plan or recommended application rates. Overapplication then occurs when the producer cannot distribute A/x loads, but instead finishes a field with a partial load still in the spreader, or not enough manure to finish the field. Rather than hauling a partial load back to the barn, the farmers managing manure on a daily-haul basis reported overapplying to portions of the field. Similarly, the producers did not prefer to haul partial loads to a field to provide uniform coverage, so portions of their fields were left without manure applied.

This full-load principle also applied when ditches, grass waterways, or buffers were present on a field. The effort to avoid these locations can result in a fragmented and complex area in which manure is to be spread. A simplifying strategy reported by our operators was to overapply or underapply on portions of their fields to drive their tractors in continuous and logical patterns. When the remaining "spreadable" field area coincided with areas that frequently flooded, the producers were left with an issue of how to credit and distribute manure to land that is vulnerable to runoff generation.

A second operational-level management decision that may potentially be affecting manure distribution was the long-standing belief that eroded knolls or side-slopes need extra manure to build or maintain the soil fertility. This traditional belief, combined with an awareness of weather conditions, was a factor in operational decisions regarding where manure was to be spread within a field. This tradition, however, counters the recommendation that producers should not apply manure to sloping lands, especially if runoff-producing events are likely. Producers also reported avoiding certain parts of fields in spreading manure because it was either "too slick" or soil compaction would occur.

Finally, producers cited the uncertainty associated in the application rate found on their spreaders. Only five of the nine operators actually knew the capacity in tons or gallons of their manure spreaders, and none recalled the actual spreading rates. Several made remarks that confirmed research (Davis, 1998; Richard and Hanna, 1999) and criticisms (Nowak et al., 1998) by others, who note the wide variation in the consistency of manure application rates using box-spreaders and side-slingers. This situation was made more complicated by the fact that some producers cited the rough ride when applying manure on previously plowed ground as a factor that forced them to drive slower, which in turn resulted in higher application rates.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
After the Federal Water Pollution Control Act (FWPCA) of 1972 and the Clean Water Act (CWA) of 1977 were both passed, the National Pollutant Discharge Elimination System (NPDES) came into effect as the official permitting process for regulating AFOs. To determine the risk that an AFO poses to the environment, the NPDES permitting process primarily focuses on the quantity of manure generated on the operation. This is why large "feedlots," or concentrated animal feeding operations (CAFOs) as they are now labeled, are explicitly targeted as likely point sources of pollution and permitted through the NPDES in the same fashion as municipal wastewater treatment plans or industrial dischargers. What constitutes a "large" farm or CAFO is an issue that is still debated. Central to this debate is the use of AU thresholds for legally promulgating the definition of CAFOs. When the CWA was written, these thresholds were not determined in a scientific fashion, but instead based on the number of operations that the USEPA could realistically regulate. As of 2003, the USEPA has eliminated the use of the generic AU thresholds in favor of thresholds based on animal species (68 FR 7176), but this change has not affected the determination of environmental risk as a function of manure generated.

The purpose of this discussion was to emphasize that the distribution phase of the manure management system on AFOs is often overshadowed by the manure generation and storage phase. This discussion further emphasizes that the manure distribution phase is an important potential precursor to nonpoint-source pollution. Since P delivery is becoming recognized as a problem that is often localized to critical source areas (Gburek and Sharpley, 1998), it is especially important to minimize both soil P accumulation and manure mismanagement at the field and subfield scales. The risk of nonpoint-source P loss from AFOs is therefore dependent not on the number of animals or fields managed, but instead by the number and extent of areas where soil P surpluses and manure mismanagement may be occurring. With regard to soil P surpluses, although larger AFOs generally manage more fields, this study found that within a group of nine large, mid-sized, and small operations, there was a basically homogeneous distribution of fields with STP surpluses. The influence of the manure distribution process, as embedded in the PI ratings for individual fields, however, was significantly different among the fields managed by these AFOs, but this significance did not appear to be related to either herd size or animal density.

The sensitivity of the field-scale risk assessment to the manure management rating is noteworthy, since manure management tends to be more complex than existing quantitative measures can realistically capture. The preceding discussion indicates that reliable quantification of manure management remains difficult because of the discrepancies between how producers plan to distribute manure versus what they actually do in the field. During interviews, producers were very cognizant of constraints affecting them at the strategic, tactical, and operational levels of decision-making and management. These constraints confounded simple mass-balance approaches to manure management. It is clear that additional research on the variability in manure distribution among producers needs to be accomplished if better quantifications of this activity are to be developed.

The value of educational activities for improving manure management will be determined by the extent that recommendations address the underlying constraints that have shaped actual distribution decisions. For example, education may have little impact on an operator's preference to haul only full loads, as this preference is largely a function of equipment design and field layout. However, a relatively easy educational objective would be to encourage more producers to learn the capacity and nutrient content of their full spreaders or tankers. Understanding the constraints to diminishing the gap between planned and actual outcomes should be the first objective of educational programs for manure management.

Precision conservation as defined by Berry et al. (2003) may also be a positive step toward rectifying the discrepancies between plans versus what happens in reality. The objective of precision conservation should be to increase the spatial and temporal congruency between operator management actions and the resiliency or buffering capacity of specific biophysical settings. The need to achieve this objective should be evident by the points discussed in this paper. Widespread applicability of precision conservation may be years in the future, but as a research tool it could be used to analyze the sensitivity of PI ratings to the spatial and temporal dimensions of the decisions guiding manure distribution. Until then, the nutrient management policy process needs to pay greater attention to the factors that shape the actual decisions made by AFOs when distributing manure in agricultural landscapes.

	ACKNOWLEDGMENTS

 
This project was partially funded by Water and Watersheds Grant R-82801001 from the U.S. Environmental Protection Agency's Science to Achieve Results (STAR) program and Grant 2001-04610 from the USDA-CSREES-IFAFS program. We also received partial support from the NSF-Long Term Ecological Research, Northern Temperate Lakes program. We are grateful to Professor Francis Pierce (Washington State University), who aided in the development of the soil sampling strategy. We would also like to recognize the assistance of Dr. Kathryn Brasier (Pennsylvania State University), Dr. Bruce Kahn, and Dr. Marieke Heemskerk, along with graduate students Larry Cutforth, Jeff Maxted, Karen Schaepe, and finally undergraduates Jason Burkhard, Brittany Futterman, Aaron Meier, Chris Scharenbroch, and Chris Sell. We also received valuable help from Kevin Connors, Laurie Lambert, and Steve Ottelien with the Dane County Land Conservation Department. Lastly, we appreciate the continued cooperation with the producers in our area of study and crop consultant Dave Buss.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Beegle, D.B., O.T. Carton, and J.S. Bailey. 2000. Nutrient management planning: Justification, theory, practice. J. Environ. Qual. 29:72–79.[Abstract/Free Full Text]
• Beegle, D.B., and L.E. Lanyon. 1994. Understanding the nutrient management process. J. Soil Water Conserv. 49(supplement):23–30.
• Beegle, D.B., L.E. Lanyon, and D.D. Lingenfelter. 1997. Nutrient management legislation in Pennsylvania: A summary of the final regulations. Agronomy Facts 40. Coop. Ext. Serv., College of Agric. Sci., Pennsylvania State Univ., University Park.
• Berry, J.K., J.A. Delgado, R. Khosla, and F. Pierce. 2003. Precision conservation for environmental sustainability. J. Soil Water Conserv. 58(6):332–339.
• Bohrnstedt, G.W. 1983. Measurement. p. 69–121. In P. Rossi, J. Wright, and A. Anderson (ed.) Handbook of survey research. Academic Press, Orlando, FL.
• Bouma, J. 1997. Precision agriculture: Introduction to the spatial and temporal variability of environmental quality. p. 5–17. In A. Stein and J. Bouma (ed.) Spatial and temporal variability of environmental quality. John Wiley & Sons, Chichester, UK.
• Bray, R.H., and L.T. Kurtz. 1945. Determination of total, organic, and available forms of phosphorus in soils. Soil Sci. Soc. Am. J. 59:39–45.
• Cabot, P.E., S.K. Bowen, and P. Nowak. 2004. Manure management in urbanizing settings. J. Soil Water Conserv. 59(6):235–243.
• Correll, D.L. 1998. The role of phosphorus in the eutrophication of receiving waters: A review. J. Environ. Qual. 27:261–266.[Abstract/Free Full Text]
• Davis, J. 1998. Manure spreader uniformity and calibration methods. p. 157–162. In Animal production systems and the environment. Proc. Int. Conf. on Odor Water Quality, Nutrient Management and Socio-Economic Issues, Des Moines, IA. 19–22 July 1998. Iowa State Univ., Ames.
• Dillman, D. 2000. Mail and internet surveys: The tailored design method. 2nd ed. John Wiley & Sons, New York.
• Dittrich, M. 1993. Corn fertility and manure management: Agricultural extension and bulletin information, 1938–1991. M.S. thesis. Univ. of Wisconsin, Madison.
• ESRI. 2002. ArcView 3.3 GIS. ESRI, Redlands, CA.
• Gburek, W.J., and A.N. Sharpley. 1998. Hydrologic controls on phosphorus loss from upland agricultural watersheds. J. Environ. Qual. 27:267–277.[Abstract/Free Full Text]
• Gollehon, N., M. Caswell, M. Ribaudo, R. Kellog, C. Lander, and D. Letson. 2001. Confined animal production and manure nutrients. Agric. Info. Bull. 771. USDA Econ. Res. Serv., Washington, DC.
• Jackson-Smith, D., M. Nevius, and B. Barham. 1995. Manure management in Wisconsin. Res. Rep. 1. Program on Agric. Technol. Studies, Madison, WI.
• Jackson-Smith, D.B., and M. Nevius. 1998. Do larger farms pose greater environmental risks? Livestock intensity and manure management on Wisconsin dairy farms. p. 383–388. In Animal production systems and the environment. Proc. Int. Conf. on Odor Water Quality, Nutrient Management and Socio-Economic Issues, Des Moines, IA. 19–22 July 1998. Iowa State Univ., Ames.
• Kelling, K.A., L.G. Bundy, S.M. Combs, and J.B. Peters. 1998. Soil test recommendations for field, vegetable and fruit crops. Univ. of Wisconsin Ext., Madison.
• Lemunyon, J.L., and R.G. Gilbert. 1993. The concept and need for a phosphorus assessment tool. J. Prod. Agric. 6:483–486.
• MidWest Plan Service. 2000. Manure characteristics: Manure management systems series. MWPS-18, Section 1, Table 6. MWPS, Ames, IA.
• Norris, P.E., and A.F. Seidl. 2001. federal water quality policy and animal confinement operations. In J.L. Outlaw and E.G. Smith (ed.) The 2002 Farm Bill: Policy options and consequences. Publ. 2001-01. Farm Foundation, Oak Brook, IL.
• Nowak, P.J., F. Madison, and R. Shepard. 1998. Farmers and manure management: A critical analysis. p. 1–32. In J. Hatfield and B. Stewart (ed.) Animal waste utilization: Effective use of manure as a soil resource. Ann Arbor Press, Chelsea, MI.
• Natural Resources Conservation Service. 2002. Conservation practice standard: Nutrient management. Code 590. NRCS, Washington, DC.
• Parry, R. 1998. Agricultural phosphorus and water quality: A U.S. Environmental Protection Agency perspective. J. Environ. Qual. 27:258–261.[Abstract/Free Full Text]
• Program on Agricultural Technology Studies. 1997. Manure management on Wisconsin livestock farms. Wisconsin Family Farm Facts 6. PATS, Madison, WI.
• Red Hen Systems. 1999. FarmGPS. Red Hen Systems, Ft. Collins, CO.
• Richard, T.L., and M. Hanna. 1999. Solid manure application: Toward a sophisticated spreader. p. 79–82. In Proc. 11th Annual Integrated Crop Management Conf., Ames, IA. 1–2 Dec. 1998. Iowa State Univ., Ames.
• Rossi, P.H., J.D. Wright, and A.B. Anderson. 1983. Handbook of survey research. Academic Press, New York.
• Saam, H., J.M. Powell, D.B. Jackson-Smith, W.L. Bland, and J.L. Posner. 2005. Use of animal density to estimate nutrient recycling ability of Wisconsin dairy farms. Agric. Syst. (in press).
• Sharpley, A.N., T.C. Daniel, J.T. Sims, and D.H. Pote. 1996. Determining environmentally sound soil phosphorus levels. J. Soil Water Conserv. 51:160–166.
• Sharpley, A.N., R.H. Foy, and P.J.A. Withers. 2000. Practical and innovative measures for the control of agricultural phosphorus losses to water: An overview. J. Environ. Qual. 29:1–9.
• Sharpley, A.N., J.L. Weld, D.B. Beegle, P.A. Klienman, W.J. Gburek, P.A. Moore, and G. Mullins. 2003. Development of phosphorus indices for nutrient management planning strategies in the United States. J. Soil Water Conserv. 58:137–152.
• Shepard, R. 2000. Nitrogen and phosphorus management on Wisconsin farms: Lessons learned for agricultural water quality programs. J. Soil Water Conserv. 55(1):63–68.
• SPSS. 2002. SPSS Version 11.5. SPSS, Chicago, IL.
• USDA and USEPA. 1999. Unified national strategy for animal feeding operations. USDA and USEPA, Washington, DC.
• USEPA. 1998. National water quality inventory: 1996 report to Congress. EPA 841-S-00-001. Office of Water, Washington, DC.
• USEPA. 2000. National water quality inventory: 2000 report to Congress. EPA 841-R-02-001. Office of Water, Washington, DC.
• Wischmeier, W.H., and D.D. Smith. 1978. Predicting rainfall erosion losses. Agric. Handb. 537. USDA Agric. Res. Serv., Washington, DC.
• Wollenhaupt, N.C., R.P. Wolkowski, and M.K. Clayton. 1994. Mapping soil test phosphorus and potassium for variable-rate fertilizer application. J. Prod. Agric. 7:441–448.

Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2005 34: ix. [Full Text]  

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Cabot, P. E. Articles by Nowak, P. Search for Related Content PubMed PubMed Citation Articles by Cabot, P. E. Articles by Nowak, P. Agricola Articles by Cabot, P. E. Articles by Nowak, P. Related Collections Animal Waste Nutrient Management Phosphorus