«University of California Division of Agriculture and Natural Resources Committee of Experts on Dairy Manure Management September 2003 February 2004, ...»
Only manure deposited in areas associated with milking (i.e., sprinkler pens, holding areas and milking parlor) are cleaned with water and enters retention ponds at dairies without flush concrete lane systems. For these systems, 8 to 19% of the total manure produced daily enters the liquid collection system (Table 3-1). Based on the total dissolved solids (TDS) concentration of dairy manure and based on the amounts recovered in the wash water, an estimated 11% of manure produced in a dry lot dairy may be in the liquid stream (Chang et al., 1974). When lactating cows are housed in dry lots where the feed lanes are cleaned by flushing, an estimated 21 to 48% of the total manure produced enters the liquid stream, accounting for manure deposited during the milking and on feed aprons. In freestall dairies, 42 to 100% of the manure may be collected in the liquid stream. Management at a given facility may be such that 100% of manure is collected as liquid during some months and markedly less manure is collected as liquid in other months.
Bedding material is used to produce soft and comfortable freestalls for cows to rest on. The sources of the bedding material vary. Coarse solids separated from liquid dairy manure, dried corral solids, almond shells, rice hulls, sawdust and straw have all been used, separately or in combination. Bedding will be transferred from the freestalls to the adjacent lane by cow activities. The quantity of bedding material varies among dairies. Bedding contributes to total solids and N loads of the dairy waste stream. In one freestall dairy bedding was quantified to be 26 lb/cow/day (11.8 kg/head/day) (Meyer et al., 2004). Thus for a freestall flush dairy, it is reasonable to suggest that all of the manure excreted by the cows, plus the daily replacement bedding material, can be collected in a liquid form.
Table 3-1: Summary of estimated residence time for milking cows on concrete surfaces and estimated percent of manure collected as a liquid1.
3.3.2 Solid Forms Manure not collected as a liquid or slurry is collected as a solid. Additionally, solids can be removed from the liquid stream. Solid separation occurs primarily through mechanical or gravity separation (including weeping walls, screens, conveyor scraper) or by settling basins. Van Horn et al. (1998), in a study of Florida dairies, indicated that the stationary mechanical separators typically remove 20 to 30% of the organic matter from the liquid waste stream.
In California, mechanical and gravity separation systems have been evaluated (Meyer et al., 2003). Mechanical screens measured in California have removed 5% to 15% of total solids (TS) in a dilute ( 2%) TS stream. The efficiency of solid removal in gravity settling basins, measured at weekly intervals, ranged from less than 25% to more than 65% (ibid.). Short-circuiting appeared to affect the performance. As the basins accumulate sediments, more and more water found a short exit path and circumvented the basin’s ability to reduce the flow velocity and settle out more suspended solids. Weeping walls consistently removed an average of more than 50% of the total solids (Meyer et al., 2004).
3.3.3 Salt and Nutrient Removal in Solid Separation Systems
Typically, most of the minerals (salts) and N remain in the liquid stream because they are either dissolved in the liquid or associated with the finest particle sizes (Figures 3-1 and 3-2, Table 3Fine particles are not removed from the liquid stream by solid separation (Meyer et al., 2004).
Van Horn et al. (1998) estimated that 5 to 20% of the N may be removed by screening when the mechanical solid separator is used in dairies in Florida. Chang and Rible (1975) size classified samples of freshly collected dairy manure. Their data also showed that the N contents of the waste material were size classified, with the highest N content found in the smallest size fraction (Table 3-2). More than 50% of the N in the liquefied dairy manure was in the 0.053 mm size fraction and would not be removed unless the solid separation process was effective to remove the very fine particles. Mechanical solid separators typically remove particles that are 0.5 - 1 mm in size. Chang and Rible’s data suggested that 25 to 30% of the N could be expected to separate in the process.
Figure 3-1: Particle size distribution in the feces and urine of four milking cows that were fed alfalfa based total mixed rations. Fecal dry matter can represent 88% to 95% of daily dry matter.
However, in a recent California study it was found that the N and macro-element nutrient content (salts) of influent and effluent samples did not significantly differ in a gravity flow separation system where approximately 50% of total solids were removed (Meyer et al., 2004). Soluble N and salts, in addition to fine and very fine suspended particles remained in solution (in the liquid manure) and were removed with solids only to the extent that water was removed in solids. For nutrient management purposes, documentation of mass of removal and nutrient content is necessary to obtain site-specific nutrient removal credits. Without such site-specific measurements, nutrient removal estimates based on total solids removal are inherently unreliable and should not be used.
Particle size greater than (microns) Figure 3-2: Nitrogen distribution among particle sizes in four milking cows that were fed alfalfa based total mixed rations. Most of the nitrogen in fecal matter is associated with the smallest particle sizes. Note that fecal N represents only 35% to 50% of the total daily N excretion (the remainder being in urine). Shown here are results for total manure (feces + urine).
3.4 Surface runoff There is no method available to estimate nutrient runoff from corral surfaces. Runoff coefficients used by the Natural Resources Conservation Service (NRCS) are small for dairies in the San Joaquin Valley. Following the method in the NRCS Agricultural Waste Management Field Office Technical Guide liquid storage needs can be estimated. Monthly rainfall and runoff coefficients were multiplied and then summed across the winter months to estimate potential runoff between Lodi and Bakersfield, suggesting runoff of 2.5 to 8 inches of rain (December through March). This allows one to estimate runoff volume. However the nutrient content of the associated runoff is unknown. Standard practices throughout California include removal of manure from corrals prior to the rainy season.
Van Horn et al. (1998) estimated that approximately 1% of the excreted N on Florida dairies may be lost through surface runoff. The frequency, intensity and amount of precipitation in the major dairy regions in California are considerably lower than in Florida. The magnitude of N and salt runoff to liquid retention systems is therefore assumed to be negligible.
3.5 Summary Manure distribution patterns vary depending on the facility infrastructure and operational and managerial decisions. Most likely, the manure distribution pattern will not be the same in any two dairies. Even within the same dairy, the deposition pattern shifts as operational parameters change. The amount of manure collected in liquid will vary from 8% (only manure excreted in the milking parlor) to 100% (manure from animals always living on concrete). More intensively managed systems (freestalls) will collect 42 to 100% of manure on a daily, monthly, or yearly basis.
There are no specific data to quantify the nutrients carried from the corral to the retention pond as a result of rain runoff. Data from Florida suggest this number is minimal.
Solids removal from mechanical and gravity flow separation systems are quite poor (under 25%) to good (consistently 50%). However, nutrient removal cannot be described based on a percent of solids removed. Soluble nutrients and salts predominantly remain in the liquid system.
Chapter 4 – Atmospheric N losses from Liquid Dairy Manure prior to Land Application
4.1 Introduction Atmospheric nitrogen losses from dairy manure can be significant and their quantification is important to improve management of N application to crops at agronomic rates. Atmospheric nitrogen losses occur as soon as feces and urine are mixed upon excretion, and continue throughout the material handling processes of collection, storage, treatment, distribution and utilization. The magnitude of atmospheric N loss from each process differs depending on the chemical and physical composition of manure and is influenced by environmental variables.
Atmospheric N losses are primarily due to volatilization of ammonia, but also include smaller amounts of nitrous oxide and nitrogen gas. The Committee investigated ranges of atmospheric ammonium nitrogen volatilization specifically from liquid dairy manure in freestalls, flushlanes, and storage lagoons (“animal production area”). The chapter is divided into two sections: The first section (section 4.2) provides a discussion of methods to measure site-specific atmospheric N losses. The second section (section 4.3) applies the recommended approach to California dairies based on data collected by members of the committee. This chapter does not cover atmospheric losses that occur during or after land application of manure. These latter type of losses are discussed in chapter 5.
4.2 Estimating atmospheric N losses at existing dairies.
The National Research Council convened an Ad Hoc Committee on Air Emission from Animal Feeding Operations (AFOs) to review and evaluate the scientific basis for estimating the emissions from existing operations (NRC, 2003). The current approach of estimating air emissions using “emission factors” is based on measuring emissions from several facilities to obtain an average emission per production unit. A critical requirement for estimating appropriate emission factors is a statistically representative survey of emissions from a class of AFOs over several iterations of the time period to be represented. The NRC (2003) committee concluded that the existing emission factors for AFOs are generally inadequate and inappropriate because of the limited number of measurements on which they are based, as well as wide emission variability among AFOs. The NRC (2003) committee also concluded that it is impractical to individually assess atmospheric nitrogen (N) losses in the animal production area via direct measurement of volatilization on existing operations, except for research purposes.
The NRC (2003) report recommended replacing the currently used “emission factor” approach with process-based modeling to estimate site specific emissions. Process-based modeling involves analysis of the farm enterprise through a mass-balance constrained study of the physicochemical nitrogen dynamics and mass transfers among its component parts (see, for example, Appendix B). The amount of N lost from one farm component affects the amount that can be lost from subsequent components. For example, if ammonia (NH3) volatilizes from the freestalls, it cannot volatilize again from the waste storage lagoon unless additional ammonia has become available through mineralization of organic N. Moreover, transformations that occur in one of the farm components might affect emissions and further transformations in other components.
Process-based modeling uses nutritional mass-balance calculations for excretion estimation and physico-chemical, dynamic process representation to represent mass flows and losses of major elements (e.g., N, carbon [C] and sulfur [S]) within the animal production area as well as in the land application (crop production) area. To assess atmospheric N losses from existing dairies, site-specific process-based models should be used in conjunction with mass balance plans as suggested by the NRC (2003) recommendations.
Whole-farm and individual farm component mass balance estimates provide aggregated estimates of all N losses including atmospheric losses, runoff, and leaching. An N balance (i.e., farm imports and exports) for each farm component can be calculated and predicted by using software described by Dou et al. (1996) as recommended by NRC (2003). The N balance is computed from information about the total amount and N contents in imported crop and feed and in exported milk and animals. Farm-component specific N balances can be obtained by considering milk production and N excretion (see chapter 2), N distribution among farm components (see chapter 3), and N application for crop production (see chapter 5). The difference between imports and exports (i.e., N balance) in each component provides a sitespecific upper limit for the maximum potential atmospheric N emissions to the environment (the loss term also includes runoff and leaching losses). Some of the records needed for such mass balance analyses are normally maintained by dairies for other (i.e., tax and management) reasons. However, for proper estimation numerous records of export/import quantities of N containing materials (i.e., animals, milk, feeds, crops) as well as their actual N contents are required and are not commonly collected on California dairies.
Actual atmospheric N losses are difficult to obtain from mass balance analyses alone as the mass balance analysis does not distinguish atmospheric, runoff, and leaching N losses within the various farm components. Hence, the mass balance approach can only be used as an upper limit check for process-based models that estimate site-specific atmospheric N losses.
4.3 Potential atmospheric N losses from liquid manure in California dairies 4.3.1 Approximate bounds of atmospheric N losses from the animal production area N to P Ratio Method: On freestall dairies, estimates of net atmospheric N losses in the animal
production area can be obtained by measuring changes in the nitrogen to phosphorus ratio: