| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Dairy and Animal Science, and
Department of Agricultural & Biological Engineering, The Pennsylvania State University, University Park 16802
2 Corresponding author: ajh{at}psu.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT REFERENCES |
|---|
Key Words: ammonia emission • dairy heifer • yeast culture • forage:concentrate
Environmental policies and regulations are having an increasing impact on animal production. Dairy and beef cattle emit approximately 50% of the NH3 released to the environment from agricultural sources in the United States (Battye et al., 1994). Currently, USDA estimates the US dairy replacement population (heifers of BW 
227 kg) at approximately 4 million, or 4% of the total dairy and beef cattle inventory (USDA, National Agricultural Statistics Service, 2007; based on 6 reports from 2005 to 2007). Dietary manipulation has been proven to affect the composition of manure excreted and NH3 emissions (James et al., 1999; Misselbrook et al., 2005). Three trials were conducted to assess the effect of manipulating the ratio of forage to concentrate (F:C) on NH3 volatilization from manure produced by growing dairy heifers
In 2 growth studies, 48 total heifers (experiment 1: 24 animals, 158.9 ± 21.4 d of age and 159 ± 3.3 kg of initial BW; experiment 2: 24 animals, 246.5 ± 50.2 d of age and 227.9 ± 27.1 kg of initial BW) were cared for according to a protocol approved by The Pennsylvania State University Institutional Animal Care and Use Committee. Heifers from the university herd were randomly assigned to 4 different pens and blocked according to age and sire in the first experiment and according to BW in the second. Pens of 6 heifers and 5 or 7 heifers in the first and second experiments, respectively, were then randomly assigned to either a high-concentrate (HC) or low-concentrate (LC) treatment by using a complete block design. Heifers remained in the same treatment group for 154 and 133 d in experiments 1 and 2, respectively. They were housed in a naturally ventilated barn with free access to water and were bedded with sand and sawdust. Heifers were fed twice daily at 0700 and 1900 h in the first trial and once a day at 0800 h in the second. Animals had individual access to feed by using the Calan Broadbent Feeding System (American Calan, Northwood, NH). Nutrient composition of feedstuffs and TMR was analyzed according to the procedures used by Moody et al. (2007). Animal BW was measured weekly approximately 1 h before feeding on 2 consecutive days per week.
Dairy heifers housed in naturally ventilated free-stall barns released NH3 to the environment from the deposition of manure on the floor. Scraping occurred once a day or every other day. A practical method to measure ammonia released to the environment is the use of portable flux chambers (Blanes-Vidal et al., 2007).
In Experiments 1 and 2, ammonia volatilization was measured from the manure accumulated on the surface of the barn floor by using a portable flux chamber. On the first day of data collection, manure was scraped from the pens, and then approximately 3 h later, NH3 emission data collection commenced. During the next 36 h, manure was not scraped, but flux chamber data collection occurred. This allowed manure from the heifers in each pen to accumulate on the barn floor. Five periods of 36 h each were measured for both experiments 1 and 2. For experiment 1, NH3 concentration was determined by using a portable non-steady-state flux chamber, multigas analyzer technique (Blanes-Vidal et al., 2007). This flux chamber method did not have internal air recirculation, and further testing revealed that to estimate emissions from the barn accurately, a correction factor was needed. Measured emission rates were multiplied by a factor of 7.9 to correct for the exclusion of wind velocity (Blanes-Vidal et al., 2007). During experiment 1, 9 locations were selected in the scraped portion of the pen that minimized gravity-flow manure or urine cross-contamination from other pens, feed, and bedding.
Experiment 2 used an improved airflow recirculation portable flux chamber to offer a more uniform internal air velocity (Wheeler et al., 2007b). Again, only the scraped portion of the pen was used for gathering emission data, but the flux chamber locations were more uniformly located across the pen. Liquid cross-contamination from other pens was prevented by constructing dams between pens on the sloped floor. For experiments 1 and 2, each pen was divided into 9 symmetric sampling squares that were measured at 8 different times throughout the 36-h sampling period. In the first experiment, the area of the alley of all 4 pens was 17.28 m2, and in the second experiment two 17.28 m2 and two 20.18 m2 pens were used.
In a companion digestibility trial (experiment 3), 8 Holstein heifers were selected from 2 different age groups (4 heifers/group), referred to as young (288 ± 4.5 d of age; 234 ± 15.3 kg BW) and old (410 ± 2.2 d of age; 409 ± 20.3 kg of BW). Dry matter intakes were estimated by using NRC (2001) guidelines and were formulated to attain 0.22 Mcal of ME/kg of empty BW0.75 and provide 800 g/d of ADG (Moody et al., 2007). Heifers were individually housed in tie stalls in an environmentally controlled barn with continuous access to fresh water. Treatments in experiment 3 were arranged as a 2 x 2 factorial: HC and LC with or without yeast culture (YC; Yea-sacc1026, Alltech Inc., Nicholas-ville, KY), which was top-dressed at the time of feeding (1 g of YC/kg of intake, as-fed basis). Treatments were administered according to a split-plot Latin square design, with heifer age as the whole plot and treatment as the subplot, and were fed for four 21-d periods (17 d of adaptation, 4 d of total collection). Ingredients and nutrient composition of treatment rations are given in Table 1
for the 3 experiments.
|
Ammonia emissions in experiment 3 were analyzed by using 6 benchtop, steady-state (dynamic) flux chambers under controlled conditions (Wheeler et al., 2007a). In brief, the dynamic chamber setup used a photoacoustic sensor (model 1412, Innova, Ballerup, Denmark), to monitor gas concentrations every 20 min for a 24-h period. Data from samples of 5 different heifers were collected in 2 separate 24-h periods (5 heifers/24-h period). A repetition of a random fifth heifer was used as an emission control. Five of the 3.8-L glass jars contained 200 g of manure slurry (feces and urine) combined at the initiation of the NH3 emission test, representing the ratio of feces to urine excreted. Total feces and urine excretion from the first 2 d of each period were used to determine the feces- to-urine ratio (wet basis) for each heifer. The sixth jar contained distilled water as a control, a check for cross- contamination of sampling lines and for determining background NH3 levels. The non-steady-state flux chamber used during measurements in experiment 1 was a first-generation design following an EPA protocol, but during use and follow-up testing it proved to underestimate the true emission rate. A correction factor was developed to better express measured emission rate to true emission rate. During experiment 2, the second-generation flux chamber that was developed to more accurately determine the true emission rate was used; thus, no correction factor was needed. For experiment 3, a steady-state, multichamber flux chamber was used that monitored emission rates from the moment the urine and feces were mixed in the chambers up to 24 h of data collection.
All statistical analysis were conducted in SAS (SAS Institute, Cary, NC) with the MIXED procedure. Because pen was considered the experimental unit, both growth trials were analyzed as a completely randomized design, with the model:
 |
where µ is the overall mean; Hi is the fixed effect of period (i = 1 to 5); Rj is the fixed effect of treatment (j = 1 to 2); Pk(R)j is the random effect of pen within treatment (k = 1 to 4); Tl is the fixed effect of time (l = 1 to 8); Lm is the fixed effect of location (m = 1 to 9); (R x T)jl is the effect of the interaction of Rj and Tl; (R x L) jm is the effect of the interaction of Rj and Lm; (T x L)lm is the effect of the interaction of Tl and Lm; (R x T x L) jlm is the effect of the interaction of Rj, Tl, and Lm; and eijklm is residual error.
Heifers were randomized across pens and pens were randomized across diets. Main effects and interactions were tested by using the residual error. Eight time points were measured (1, 4, 7, 10, 25, 28, 31, 34 h) after the last scraping of the pens. For the first experiment, location was not included in the model because the same location was sampled at every time point. All dependent variables of the digestibility experiment were analyzed as a 4 x 4 Latin square design. A split-plot design was used, with age as the whole plot and diet treatment as the subplot. Sources of variation associated with fixed design effects of period and fixed treatment effects of age category, F:C and YC addition, and their interaction with period as the repeated effect were used. Heifer within age was included as a random effect and repeated measurements were analyzed by using the first-order autoregressive covariance structure (Littell et al., 1998). Residual variances were assumed to be normally distributed, and all data are presented as the least squares means. For all data, treatment effects were considered significant when P < 0.05.
Ingredients and nutrient composition of treatment rations are given in Table 1
for the 3 experiments. All experiments used corn silage as the sole forage source, and diets used in the second and third experiment were the same. In experiment 1, manure from pens with heifers receiving LC diets had numerically greater NH3 volatilization than did manure from HC heifers (314.0 vs. 174.4; SE = 36.1 µg/cm2 per min; P = 0.11).
In the second experiment, there was no difference between LC or HC heifers in NH3 emissions (322 vs. 290; SE = 92 µg of NH3/cm2 per min; P = 0.81). In the first experiment, the sampled part of the pen avoided areas closest to the feed bunk or the adjacent pen, which resulted in lower emissions than in the second experiment, in which the whole pen was sampled. It is important to note that the whole-pen measurement is more representative but that the same emission pattern was followed in both experiments. There was a time effect in both experiments, and emissions were higher after heifers were fed (P < 0.01; Figures 1 and 2). Possible explanations could be that the mixture of feed with slurry increased NH3 emissions from the pen area. In addition, greater amounts of manure were produced while heifers were standing by the feed bunk; thus, emission rate was greater at this time for both treatments. The tendency of the HC diets to reduce NH3 volatilization in both of the experiments is partially explained by the lower DMI of the HC heifers (5.7 vs. 4.8 kg; SE = 0.10; P < 0.01 and 5.5 vs. 5.3 kg; SE = 0.46; P = 0.09).
|
|
presents the results for the digestibility study, in which NH3 emissions from manure slurries were greater on a unit of manure basis for the HC diets (P < 0.01). Nitrogen intake was not different among treatments (Table 2). However, NFC content was greater for the HC diets (Table 1). Increasing dietary NFC has been shown to affect microbial N metabolism by increasing starch fermentation in the large intestine (Clark et al., 1992). As a result, more blood urea N can be recycled into microbial N, shifting N excretion from urine to feces (Knowlton et al., 1998). According to this, less urea would be readily available for volatilization with the use of HC diets. However, in the present study, changes in the ratio of feces to urine (P < 0.01) resulted from a reduced fecal excretion by heifers consuming the HC diet. Thus, more urinary urea per unit of manure (Table 2; P < 0.01) was available to be hydrolyzed by the urease present in feces (Misselbrook et al., 2005). It is important to note that water intake did not have an effect on the greater feces:urine of the HC group, because it was similar between animals fed the HC or LC diet (Table 2). As hypothesized from the results for the first 2 experiments, less total manure (g/d) was excreted by heifers fed the HC diets (Moody et al., 2007) as a result of a reduction in DMI (Table 2). In addition, total NH3 emissions were not different when reported for a 24-h period (19.67 vs. 19.07, and 11.94 vs. 11.52 g of NH3/d; P = 0.44) for HC and LC in the 2 age groups. This is particularly important on a farm level, because it shows that there is no difference in NH3 released to the environment on a daily basis when feeding HC diets with the advantage of a reduced quantity of manure to be handled. Moreover, these results are in agreement with Powell et al. (2008). These researchers used group animal chambers to monitor NH3 emissions of dairy heifers of approximately 400 kg.; emissions ranged from 10.1 to 21 g of NH3/d, depending on the season or bedding used. This demonstrates that simulation and laboratory models can be used to asses the effects of dietary and management changes in animal husbandry. However, caution should be taken when interpreting these results, because dietary changes can have an impact on NH3 released to the environment and different forage-based diets have reported higher emissions (James et al., 1999).
|
In summary, NH3 emissions from HC diets were not different from LC diets when measured from the barn floor in experiments 1 and 2. Ammonia emissions were greater when they were measured while heifers were eating, and the same emission pattern was followed in both growth studies. In the digestibility study, emissions from slurries of manure from heifers fed the HC diet showed greater emission rates per unit of manure. The fact that HC diets resulted in less DMI and a subsequent reduction in fecal output affected the feces to urine ratio. The shift in this ratio affected NH3 emissions on a per-gram basis, whereas the total manure produced affected daily emissions. Cumulative daily emissions were not different between diets, which concurs with the results found in experiments 1 and 2 and with the YC addition.
|
FOOTNOTES |
|---|
Received for publication March 13, 2008. Accepted for publication July 2, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. J. Lascano, G. I. Zanton, F. X. Suarez-Mena, and A. J. Heinrichs Effect of limit feeding high- and low-concentrate diets with Saccharomyces cerevisiae on digestibility and on dairy heifer growth and first-lactation performance J Dairy Sci, October 1, 2009; 92(10): 5100 - 5110. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. I. Zanton and A. J. Heinrichs Digestion and nitrogen utilization in dairy heifers limit-fed a low or high forage ration at four levels of nitrogen intake J Dairy Sci, May 1, 2009; 92(5): 2078 - 2094. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and low-concentrate (
) diets in experiment 1 over a 36-h period. Arrows indicate feeding times at 2 and 26 h after scraping. Bars on the graph indicate standard errors. **P < 0.05; *P < 0.10.