«University of California Division of Agriculture and Natural Resources Committee of Experts on Dairy Manure Management September 2003 February 2004, ...»
Under the supervision of Thomas Harter, Van der Schans (2001) confirmed the leaching rates with a calibrated, site-specific groundwater model. Harter et al. (2001) and Van der Schans (2001) also implemented a field mass balance for a field in Merced County that received frequent manure applications and compared the leaching estimates from the mass balance with those obtained from groundwater quality measurements downgradient of the field. They estimated that, on the average, the N input to the forage production field amounted to approximately 1,200 kg ha-1 yr-1, of which 900, 280, and 28 kg ha-1 yr-1 came from dairy manure, chemical fertilizers, and irrigation water plus atmospheric deposition, respectively. A total of 540 kg ha-1 yr-1 of the 1,200 kg N ha-1 yr-1 input (45%) may be accounted for by N in the harvested corn and fall-planted wheat. The mass balance approach suggests that 660 kg ha-1 yr-1 are lost to leaching, volatilization, or denitrification. Monitoring well measurements on the other hand indicate that approximately 600 to 700 kg ha-1 yr-1 N was recharged to the shallow groundwater (55%).
Hence, groundwater nitrate accounted for all of the losses, meaning that N was completely mineralized and no denitrification occurred. Does this mean that volatilization and denitrification losses are near zero? The authors point out that the estimate for the manure N input (900 kg ha-1 yr-1) is not very precise. Higher amounts may have been applied, which would imply that some denitrification and/or volatilization losses have indeed occurred.
Harter et al. (2001) installed a series of monitoring wells immediately downgradient from manure application fields to assess the condition of shallow groundwater and its response to improved nutrient management that began in 1998. By matching dairy wastewater nutrient application to crop nitrogen uptake during the growing seasons, they showed that nitrate concentrations in the groundwater have been steadily declining since the experiment began due to the improved balance between N input and crop N uptake.
Comparison of mass balance analysis and groundwater quality also confirmed that in this case a) all organic nitrogen mineralized and b) that denitrification and volatilization during the application and in the unsaturated zone of the fine sandy to loamy sandy soils were negligible.
Harter, T., H. Davis, M.C. Mathews and R.D. Meyer. 2002. Shallow groundwater quality on dairy farms with irrigated forage crops. J. of Contaminant Hydrology 55 (3-4), pp.
Van der Schans, Martin. 2001. Nitrogen leaching from irrigated dairy farms in Merced County, California. Thesis report, Sub-Department of Water Resources, Wageningen University, The Netherlands. 58 pages.
Abstract Manure nutrient management, long neglected and considered merely a waste disposal operation, is one of the key elements of sustainable dairy farming and the corner-stone of recently enacted local, state, and federal animal farming regulations in California. A four-year field trial of liquid dairy manure nutrient management in a flood irrigated corn-winter grain double crop rotation was implemented. Throughout the study period, water and nitrogen applications were metered, soil profile ammonia and nitrate measurements were taken, and shallow groundwater quality was measured immediately downgradient of the field trial. Here, we use inverse modeling of the unsaturated zone flow and transport processes to determine field scale soil hydraulic properties, mineralization, nitrification, and denitrification rates, and the linear sorption coefficients for organic nitrogen and ammonia. The retardation coefficient for organic nitrogen sorption is very high, although soil profile ammonia predictions are not very sensitive to retardation coefficients, R, above 10. The retardation coefficient for ammonia sorption (R = 3) is obtained by fitting modeled to measured plant nutrient uptake, which is very sensitive to ammonia sorption. The best fit transformation rates are kmin=1.6 d-1 for summer-time mineralization, knit=0.1 d-1 for summer-time nitrification, and kden=0. Despite the extensive soil profile data collected, results are relatively insensitive to mineralization rates.
The model is insensitive to organic N transformation that results in half-lifes much smaller than 2 months. Organic N half-lifes ranging from less than 1 day to 2 months all give reasonable results. However, the inverse model is much more sensitive to ammonia nitrification rates. The ammonia half-life is found to be one week. Unlike soil ammonia and nitrate profiles, predicted nitrate fluxes to groundwater are sensitive to the percent organic N available for mineralization and net denitrification. Only under complete mineralization and with negligible denitrification can the model explain the elevated groundwater nitrate concentrations. The relatively high mineralization and nitrification rates are thought to be Report to the Committee of Consultants, December 2002 Nakamura et al., Committee of Consultants Report December 2002 caused by the low organic carbon content in the soil profile, the high temperatures, the high dissolved oxygen content of the irrigation water, and the strongly oxic conditions throughout the unsaturated zone. The same conditions also explain why denitrification is found to be minimal.
Keywords: nitrogen, mineralization, nitrification, sorption, vadose zone modeling, manure, dairy, fertilization
Nutrient management is a key issue regarding the environmental sustainability of dairy operations. In California and elsewhere, manure nutrient management has become the corner stone of the permit process for dairy (and other animal) farming at the local, state, and federal level. Proper nutrient management is key to successfully growing a crop on manure fertilizer without negatively affecting the environment (groundwater, surface water, air). The goal of proper nutrient management is to balance the amount of nutrients applied in a field with the amount of nutrients used by the crop (Meisinger and Randall, 1991). Optimal nutrient
application practices that meet crop needs while sustaining environmental quality rely on:
knowing the amount of nutrients in the manure, knowing the amount and distribution of nutrients applied to the field, knowing the crop nutrient needs and uptake processes, and understanding the fate of nutrients once they are incorporated into field soil.
The focus of this report is the fate of soil nitrogen: In California, dairies are the largest animal farming industry. Most of California’s dairy herd is located in valleys and basins, often far from surface water (e.g., Tulare basin in the Central Valley), but generally overlying alluvial aquifer systems that are more or less vulnerable to nonpoint source pollution. Of the major nutrients, nitrogen is of particular interest with respect to groundwater due to its potential for nitrate contamination. In contrast, the other two major nutrients (potassium and phosphorus) are not commonly found to occur at concentrations that would limit the beneficial uses of groundwater. Phosphorus becomes a major nutrient management issue only in areas with surface water runoff to streams or indirect runoff via tile drainage or via shallow groundwater to nearby streams. Potassium contributes to the overall salinity of groundwater.
Manure nitrogen is primarily available as organic nitrogen and ammonium-nitrogen (ammonium-N). In both, solid and liquid manure, nitrate-nitrogen (nitrate-N), nitrite-N, and other forms of inorganic nitrogen are generally available at negligible and often unmeasurable amounts. After the application of manure to field crops, organic nitrogen begins to mineralize to ammonium-N, which in turn converts to nitrate-N if soils are sufficiently aerated. It is generally thought, that only a fraction of the organic nitrogen will mineralize over time, while the remainder becomes incorporated into the soil nitrogen pool (bound into soil organic matter).
Knowledge of the amount and timing of organic nitrogen mineralization is critical in planning proper nutrient management. Mineralization rates not only affect the timing of nutrient applications. The total amount of mineralization is directly related to the sizing of the dairy herd given a limited acreage for manure fertilizer applications. The larger the amount of Nakamura et al., Committee of Consultants Report December 2002 organic N that becomes permanently embedded in the soil organic N pool, the larger the amount of manure that can be applied and the larger the herd size sustainable. In well-aerated soils, ammonia nitrification is usually complete and under California climatic conditions generally occurs within days of mineralization to ammonia. Only the nitrate-N is mobile enough to leach below the root zone, through the deeper vadose zone, and into shallow groundwater in significant quantities, where excess moisture leaches out of the soil profile.
Available literature for quantification of the nitrogen cycling processes in soils are climate-, soil-, and crop-specific. Little data are currently available for the typical cropping systems and soils occurring in the San Joaquin Valley, California, where two-thirds of the California dairy herd is located. Moreover, much of the existing literature relies on extrapolation of soil profile and plant uptake data with little or no control through deep vadose zone or shallow groundwater quality data. The objective of this report is to characterize mineralization, nitrification, and denitrification processes that occur in the vadose zone below a crop fertilized with liquid dairy manure. A four year field trial was established in a corn and winter grain rotation on a dairy farm located on relatively permeable, sandy soils near Hilmar, Merced County, California. The field, “Bun1”, was managed to optimize the use of liquid dairy manure, while minimizing the use of commercial fertilizer. We use inverse modeling of unsaturated zone flow and transport processes to estimate mineralization, nitrification, and denitrification rates from root zone nitrate-N and ammonium-N profiles collected at various time intervals and from groundwater nitrate measurements observed during the four year trial.
H.2 Model Description The modeling period is from April 28th, 1998 to September 28th, 2001. We use HYDRUS-1D (ver.2) (Simunek et al., 1998).
H.2.1 Water transport in vadose zone
The governing water transport equation is expressed by Richards' equation:
: volumetric water content (cm cm-3), h : pressure head (cmH2O), K : unsaturated hydraulic conductivity (cm h-1), z : vertical coordinate (cm).
The relationship between h and is formulated by van Genuchten's equation:
H.2.2 The fate of nitrogen We consider organic-nitrogen (Org-N), ammonium-nitrogen (NH4-N), and nitrate-nitrogen (NO3-N). The transformations of nitrogen are simulated using HYDRUS-1D including mineralization (Org-N -- NH4-N), nitrification (NH4-N -- NO3-N), and denitrification (NO3-N -- gaseous nitrogen).
The governing equations describing the transport and fate of nitrogen are expressed as follows,
c1, c2, c3 : concentrations of soluble Org-N, soluble NH4-N, and soluble NO3-N (mgN cm-3), s1, s2 : concentrations of adsorbed Org-N and adsorbed NH4-N (mgN g-1), : bulk density (1.55g cm-3), D1, D2, D3 : dispersion coefficients of soluble Org-N, soluble NH4-N, soluble NO3-N (cm2h-1), kmin, knit, kden : the first-order reaction coefficients of mineralization, Nakamura et al., Committee of Consultants Report December 2002 nitrification, and denitrification (h-1) The dispersion coefficient is formulated by
DL: longitudinal dispersivity (cm), |q| : the absolute value of Darcy flux (cm h-1), Dw :
molecular diffusion coefficient in free water (cm2h-1), : tortuosity (-) Dw of Org-N, NH4-N, and NO3-N are set up to be zero because the nitrogen fluxes by diffusion are negligible compared to the dispersive transport. DL is assumed to be 20cm.
For lack of better data, we assume that the first-order coefficients of soluble nitrogen and adsorbed nitrogen are identical. The isotherms of Org-N and NH4-N are formulated by a
linear adsorption isotherm:
Mineralization, nitrification, and denitrification are all linked to microbial activity, which is strongly temperature dependent. Lower soil temperature during the winter months significantly reduce microbial activity and, hence, transformation rates. We assumed that the soil surface temperature is 0ºC during the periods from November 1st to March 31st every year and 20ºC during the rest periods and calculated the heat transport equation, T T T C Cw q (10) t z z z C : volumetric heat capacity of soil (J K-1cm-3), T : soil temperature, : thermal conductivity of soil (W cm-1K-1), Cw: volumetric heat capacity of water (J K-1cm-3) where is set to be a relatively high value so that soil temperature profile changes rapidly at the two switching dates with an accompanying rapid switch in reaction coefficients. Transformation rates are maximal during the summer and set to be negligible during the winter (when the model assumes soil temperatures of 0ºC). These conditions only approximate actual conditions, but are considered appropriate in light of the lack of data on actual transformation rates during the winter. Not enough data existed to calibrate summer and winter time transformation rates separately.
Nakamura et al., Committee of Consultants Report December 2002 H.2.3 Root Uptake of Water and Soluble NH4-N and NO3-N Feddes' model is used to simulate root water uptake (Simunek et al., 1998). The model parameters are provided in the database in HYDRUS-1D for the vegetative period of corn and small grains. Furthermore, we consider the root growth modeled by Verhulst-Pearl's logistic
L0 : the initial root depth (set to be 0.01cm for both crops), r : the growth rate (h-1) The growth rate r is calculated based on the assumption that 50% of the rooting depth will be reached after 50% of the growing season has elapsed.
The root distribution function is expressed by an exponential function with a maximum at the soil surface.