- Email: [email protected]
Relevance of Water Quality to Broiler and Turkey Performance
Relevance of Water Quality to Broiler and Turkey Performance T. LIONEL BARTON University of Arkansas, Fayetteville, Arkansas 72701 tively correlated with livability and calcium and nitrate were positively correlated with condemnation. The bacterial results showed no significant difference between top and bottom producers for either Pseudomonas or Escherichia coli. In the turkey study, calcium, magnesium, bicarbonate, hardness, and aggressive index were beneficial to feed conversion. Phosphate and ammonia were detrimental to feed conversion. Calcium, magnesium, dissolved oxygen, zinc, hardness, and aggressive index were all positively correlated with adjusted body weight. Magnesium was negatively correlated with livability. Magnesium and aggressive index were positively correlated with condemnation and potassium, zinc, nitrate, and phosphate were negatively correlated with condemnation. Although fewer farms were involved in the turkey study, the results generally support the results of the broiler study.
(Key words: water quality, nitrate, hardness, dissolved oxygen, aggressive index) 1996 Poultry Science 75:854-856
INTRODUCTION Water quality is an important consideration in the performance of both broilers and turkeys. We have been interested in the mineral content of water and its effect on performance of broilers and turkeys for a number of years. We have conducted a broiler study involving 300 farms and a turkey study involving 100 farms to determine the effect that water quality has on performance.
BROILER STUDY This study was conducted in cooperation with three integrated poultry companies, each having at least two locations in the state. Water was tested from 300 broiler farms in the state (100 samples from each company). Twenty-five top producers and 25 bottom producers were selected at each location by each company. Attempts were made to eliminate poor producers that had obviously inadequate housing or poor management
Received for publication August 10, 1994. Accepted for publication April 1, 1996.
practices. Performance data for the previous year were obtained from the integrated company to use in this study. Performance criteria were feed conversion, body weight, livability, and condemnation. Because different companies produce broilers to different ages and different market weights, these data were adjusted to a 49-d body weight and 49-d feed conversion by linear regression. The livability and condemnation information were not adjusted. One company produced a significant number of all male broilers. These data were converted to mixed-sex weights and feed conversion by least squares procedures. The water samples were collected either at the well source or at point of entry into the broiler house. The mineral analysis samples were collected in quart plastic bottles and transported to the University of Arkansas Diagnostic Laboratory for analysis. Samples were assayed for items listed in Table 1 according to procedures listed in Standard Methods for the Examination of Water and Waste Water (1980). Bacterial samples were collected either by the author or by selected personnel within these companies. The procedure for collecting these samples was to flame the water faucet with a propane torch, run a small amount of water to cool the faucet, and collect the sample in a sterile 854
Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 6, 2015
ABSTRACT Water was tested from 300 broiler farms in Arkansas in cooperation with three integrated poultry companies, each having at least two locations in the state. The turkey study was conducted in cooperation with three integrated turkey companies with samples from 100 turkey farms, although the numbers were not equal among companies. Performance criteria collected were body weight, feed conversion, livability, and condemnation. In the overall analysis in the broiler study, nitrate had a detrimental effect on performance. Calcium was negatively correlated with adjusted conversion; i.e., conversion improved as calcium increased. Magnesium was positively correlated with adjusted conversion, or had an adverse effect on conversion. Dissolved oxygen, bicarbonate, hardness, calcium, and magnesium were all positively correlated with adjusted weight but nitrate was negatively correlated with adjusted weight. Calcium and potassium were nega-
855
SYMPOSIUM: WATER QUALITY AND POULTRY PRODUCTION TABLE 1. Analyses conducted on water samples Cations
Anions
Other
Ammonia Calcium Magnesium Sodium Potassium Iron Zinc Manganese Copper
Phosphate Nitrate Sulfate Bicarbonate Carbonate Chloride Nitrite
pH Electrical conductivity Hardness Dissolved oxygen
TURKEY STUDY This study was conducted in cooperation with three integrated turkey companies. Water was tested from 100 turkey farms in the state, although the numbers were not equal among companies. Performance criteria were the same as in the broiler study. The performance data were adjusted to 130-d age and sex. The same minerals were analyzed as in the broiler study. In addition, an aggressive index (AI) (which indicates corrosiveness of water) was calculated for each sample and analyzed in this study. Bacterial samples were not collected in this study.
BROILER STUDY RESULTS In the overall analysis, nitrate (quantity) was the only mineral that had a significant effect on performance. Higher nitrate levels had a detrimental effect on
The bacterial results were analyzed on a positive or negative basis using chi-square procedures. No significant differences were found between top and bottom producers for either Pseudomonas or E. coli. The number of positive samples by company and by location is shown for top and bottom producers in Table 3. Company 1 and Company 3 producers were tested at the headquarters location only, whereas Company 2 producers were tested at both locations. Company 2 had been testing farms for quite some time and chlorinating those found to be positive, which may explain why they had only about one-half the number of positive farms as Companies 1 and 3.
TURKEY STUDY RESULTS Significant correlations in the turkey study are also shown in Table 2. Calcium, magnesium, bicarbonate, hardness, and aggressive index were beneficial to feed conversion. Phosphate and ammonia were detrimental to feed conversion. Calcium, magnesium, dissolved oxygen, zinc, hardness, and aggressive index were all positively correlated with adjusted body weight. Magnesium was negatively correlated with livability. Magnesium and AI were positively correlated with condemnation, whereas potassium, zinc, nitrate, and phosphate were negatively correlated with c o n d e m n a t i o n .
TABLE 2. Simple correlations with performance data Broiler study
Turkey study
Performance data
Positive correlation
Negative correlation
Positive correlation
Negative correlation
Adjusted conversion
Mg
Ca
P04, NH3
Ca, Mg, H C 0 3 Hardness 2 , AI
Adjusted weight
DO, H C 0 3 / Hardness, Ca (0.09) Mg
N03
Ca, Mg, DO, Zn, Hardness, AI
Adjusted livability Adjusted condemnation
Ca, K Ca, N 0 3
Mg, AI
Mg K, Zn, N 0 3 , P 0 4
Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 6, 2015
container. None of these samples were collected inside broiler houses in which birds were present. (It is the author's opinion that an aseptic sample cannot be collected inside a broiler house when birds are present.) These bacterial water samples were transported to laboratories of two companies involved in this study and plated on the same day of collection. Company 1 and Company 3 producers were tested at the headquarters location only, but Company 2 producers were tested at both locations. The samples were cultured for Escherichia coli using MacConkey's medium and for Pseudomonas using Cetrimide agar. One-half milliliter of water was used for each plate.
performance. Simple correlation coefficients that were significant (P < 0.05) are shown in Table 2 for the overall analysis. Calcium was negatively correlated with adjusted conversion, which means that as calcium increased, conversion decreased. That is, conversion improved as calcium increased. Magnesium was positively correlated with adjusted conversion, or had an adverse effect on conversion. Dissolved oxygen, bicarbonate, hardness, calcium (P < 0.09), and magnesium all were positively correlated with adjusted weight but nitrate was negatively correlated with adjusted weight. Calcium and potassium were negatively correlated with livability, indicating that livability decreased as calcium or potassium increased. This detrimental effect of calcium on livability is opposite to the beneficial effect of calcium on adjusted conversion and adjusted weight. One explanation could be that calcium is interfering with water vaccination and thus affecting livability. Calcium and nitrate were positively correlated with condemnation.
856
BARTON TABLE 3. Number of positive bacterial samples Pseudomonas Company 1 2 2
3
1
Location 0 3 4
5
Top producer
Escherichia coli Top producer 6
2 0
Bottom producer 4 1 4 2
0
Bottom producer 9 4 7
2
2
9
10
Although fewer farms were involved in this study, the results generally support the results of the broiler study.
DISCUSSION
Water samples from 300 broiler farms in Arkansas were analyzed for mineral content and the mineral content correlated with body weight, feed conversion, livability, and mortality. Twenty-five top and 25 bottom producers were selected at each of two locations by three integrated broiler firms (100 samples from each company). Bacterial samples were collected from 200 of these farms and cultured for Pseudomonas and E. coli. No differences were found between top and bottom producers related to bacterial contamination. Dissolved oxygen, bicarbonate, hardness, calcium (P < 0.09), and magnesium were all positively correlated with growth rate, whereas nitrate was negatively correlated with growth rate. Calcium was negatively correlated with feed conversion and magnesium was positively correlated with feed conversion. This result means that feed conversion improved as calcium increased and got worse as magnesium increased. Calcium and potassium were negatively correlated with livability. The detrimental effect of these minerals may be related to interference during water vaccination, thus affecting livability. Interference during vaccination might also explain the positive correlation of calcium and nitrate with condemnation. We suggest testing the water for mineral content through State Diagnostic Laboratories. In addition, we recommend testing the water for bacterial contamination at the same time. Chlorination will be a cheaper first treatment in trying to correct water quality related performance problems. We also ask producers whether they are on a rural water system. In some cases, we suspect that high chlorine levels are interfering with field vaccinations. Water quality for broiler and turkey production will become increasingly important and hopefully will stimulate more research in future years.
ACKNOWLEDGMENT This research was supported in part by a grant from the Southeastern Poultry and Egg Association.
REFERENCES American Public Health Association, 1980. Standard Methods for the Examination of Water and Waste Water. 15th ed. American Public Health Association, Washington, DC.
Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 6, 2015
An examination of the correlation analysis in Table 2 for the broiler and turkey studies indicate some similarities as well as some differences. Calcium was negatively correlated with feed conversion in both studies. Higher calcium levels meant lower feed conversions or better feed conversions. Magnesium, bicarbonate, and hardness were also associated with improved feed conversions in the turkey study. However, magnesium was associated with poor feed conversions in the broiler study. The AI was also associated with improved feed conversions in the turkey study and is an indicator of the corrosivity in water. The AI is derived from the actual pH, calcium hardness, and total alkalinity. The AI is not a quantitative measure of corrosion but is a general indicator of the tendency for corrosion to occur. An AI of 12 or above indicates unaggressive (not corrosive) water. Aggressive index values of 10 to 12 indicate the water is moderately aggressive. Aggressive index values below 10 indicate extremely aggressive (corrosive) water. A higher AI value negatively correlated with feed conversion means better feed conversion with less aggressive water. The same situation occurred with adjusted weight in that less aggressive water was associated with higher growth rate. Adjusted weight was also positively correlated with dissolved oxygen, calcium, magnesium, and hardness in both studies. Additionally, bicarbonate was positively correlated with weight in the broiler study, whereas zinc and aggressive index were correlated in the turkey study. Nitrate was negatively correlated with adjusted weight in the broiler study but not in the turkey study. Calcium and potassium were negatively correlated with livability in the broiler study whereas magnesium was negatively correlated with livability in the turkey study. Condemnation was positively correlated with calcium and nitrate in the broiler study and with magnesium and AI in the turkey study. It is unclear why less aggressive water affected condemnation although it was beneficial to weight and feed conversion. It is also puzzling as to why potassium, zinc, nitrate, and phosphate were negatively correlated with condemnation in the turkey study.
SUMMARY
(Key words: water quality, nitrate, hardness, dissolved oxygen, aggressive index) 1996 Poultry Science 75:854-856
INTRODUCTION Water quality is an important consideration in the performance of both broilers and turkeys. We have been interested in the mineral content of water and its effect on performance of broilers and turkeys for a number of years. We have conducted a broiler study involving 300 farms and a turkey study involving 100 farms to determine the effect that water quality has on performance.
BROILER STUDY This study was conducted in cooperation with three integrated poultry companies, each having at least two locations in the state. Water was tested from 300 broiler farms in the state (100 samples from each company). Twenty-five top producers and 25 bottom producers were selected at each location by each company. Attempts were made to eliminate poor producers that had obviously inadequate housing or poor management
Received for publication August 10, 1994. Accepted for publication April 1, 1996.
practices. Performance data for the previous year were obtained from the integrated company to use in this study. Performance criteria were feed conversion, body weight, livability, and condemnation. Because different companies produce broilers to different ages and different market weights, these data were adjusted to a 49-d body weight and 49-d feed conversion by linear regression. The livability and condemnation information were not adjusted. One company produced a significant number of all male broilers. These data were converted to mixed-sex weights and feed conversion by least squares procedures. The water samples were collected either at the well source or at point of entry into the broiler house. The mineral analysis samples were collected in quart plastic bottles and transported to the University of Arkansas Diagnostic Laboratory for analysis. Samples were assayed for items listed in Table 1 according to procedures listed in Standard Methods for the Examination of Water and Waste Water (1980). Bacterial samples were collected either by the author or by selected personnel within these companies. The procedure for collecting these samples was to flame the water faucet with a propane torch, run a small amount of water to cool the faucet, and collect the sample in a sterile 854
Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 6, 2015
ABSTRACT Water was tested from 300 broiler farms in Arkansas in cooperation with three integrated poultry companies, each having at least two locations in the state. The turkey study was conducted in cooperation with three integrated turkey companies with samples from 100 turkey farms, although the numbers were not equal among companies. Performance criteria collected were body weight, feed conversion, livability, and condemnation. In the overall analysis in the broiler study, nitrate had a detrimental effect on performance. Calcium was negatively correlated with adjusted conversion; i.e., conversion improved as calcium increased. Magnesium was positively correlated with adjusted conversion, or had an adverse effect on conversion. Dissolved oxygen, bicarbonate, hardness, calcium, and magnesium were all positively correlated with adjusted weight but nitrate was negatively correlated with adjusted weight. Calcium and potassium were nega-
855
SYMPOSIUM: WATER QUALITY AND POULTRY PRODUCTION TABLE 1. Analyses conducted on water samples Cations
Anions
Other
Ammonia Calcium Magnesium Sodium Potassium Iron Zinc Manganese Copper
Phosphate Nitrate Sulfate Bicarbonate Carbonate Chloride Nitrite
pH Electrical conductivity Hardness Dissolved oxygen
TURKEY STUDY This study was conducted in cooperation with three integrated turkey companies. Water was tested from 100 turkey farms in the state, although the numbers were not equal among companies. Performance criteria were the same as in the broiler study. The performance data were adjusted to 130-d age and sex. The same minerals were analyzed as in the broiler study. In addition, an aggressive index (AI) (which indicates corrosiveness of water) was calculated for each sample and analyzed in this study. Bacterial samples were not collected in this study.
BROILER STUDY RESULTS In the overall analysis, nitrate (quantity) was the only mineral that had a significant effect on performance. Higher nitrate levels had a detrimental effect on
The bacterial results were analyzed on a positive or negative basis using chi-square procedures. No significant differences were found between top and bottom producers for either Pseudomonas or E. coli. The number of positive samples by company and by location is shown for top and bottom producers in Table 3. Company 1 and Company 3 producers were tested at the headquarters location only, whereas Company 2 producers were tested at both locations. Company 2 had been testing farms for quite some time and chlorinating those found to be positive, which may explain why they had only about one-half the number of positive farms as Companies 1 and 3.
TURKEY STUDY RESULTS Significant correlations in the turkey study are also shown in Table 2. Calcium, magnesium, bicarbonate, hardness, and aggressive index were beneficial to feed conversion. Phosphate and ammonia were detrimental to feed conversion. Calcium, magnesium, dissolved oxygen, zinc, hardness, and aggressive index were all positively correlated with adjusted body weight. Magnesium was negatively correlated with livability. Magnesium and AI were positively correlated with condemnation, whereas potassium, zinc, nitrate, and phosphate were negatively correlated with c o n d e m n a t i o n .
TABLE 2. Simple correlations with performance data Broiler study
Turkey study
Performance data
Positive correlation
Negative correlation
Positive correlation
Negative correlation
Adjusted conversion
Mg
Ca
P04, NH3
Ca, Mg, H C 0 3 Hardness 2 , AI
Adjusted weight
DO, H C 0 3 / Hardness, Ca (0.09) Mg
N03
Ca, Mg, DO, Zn, Hardness, AI
Adjusted livability Adjusted condemnation
Ca, K Ca, N 0 3
Mg, AI
Mg K, Zn, N 0 3 , P 0 4
Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 6, 2015
container. None of these samples were collected inside broiler houses in which birds were present. (It is the author's opinion that an aseptic sample cannot be collected inside a broiler house when birds are present.) These bacterial water samples were transported to laboratories of two companies involved in this study and plated on the same day of collection. Company 1 and Company 3 producers were tested at the headquarters location only, but Company 2 producers were tested at both locations. The samples were cultured for Escherichia coli using MacConkey's medium and for Pseudomonas using Cetrimide agar. One-half milliliter of water was used for each plate.
performance. Simple correlation coefficients that were significant (P < 0.05) are shown in Table 2 for the overall analysis. Calcium was negatively correlated with adjusted conversion, which means that as calcium increased, conversion decreased. That is, conversion improved as calcium increased. Magnesium was positively correlated with adjusted conversion, or had an adverse effect on conversion. Dissolved oxygen, bicarbonate, hardness, calcium (P < 0.09), and magnesium all were positively correlated with adjusted weight but nitrate was negatively correlated with adjusted weight. Calcium and potassium were negatively correlated with livability, indicating that livability decreased as calcium or potassium increased. This detrimental effect of calcium on livability is opposite to the beneficial effect of calcium on adjusted conversion and adjusted weight. One explanation could be that calcium is interfering with water vaccination and thus affecting livability. Calcium and nitrate were positively correlated with condemnation.
856
BARTON TABLE 3. Number of positive bacterial samples Pseudomonas Company 1 2 2
3
1
Location 0 3 4
5
Top producer
Escherichia coli Top producer 6
2 0
Bottom producer 4 1 4 2
0
Bottom producer 9 4 7
2
2
9
10
Although fewer farms were involved in this study, the results generally support the results of the broiler study.
DISCUSSION
Water samples from 300 broiler farms in Arkansas were analyzed for mineral content and the mineral content correlated with body weight, feed conversion, livability, and mortality. Twenty-five top and 25 bottom producers were selected at each of two locations by three integrated broiler firms (100 samples from each company). Bacterial samples were collected from 200 of these farms and cultured for Pseudomonas and E. coli. No differences were found between top and bottom producers related to bacterial contamination. Dissolved oxygen, bicarbonate, hardness, calcium (P < 0.09), and magnesium were all positively correlated with growth rate, whereas nitrate was negatively correlated with growth rate. Calcium was negatively correlated with feed conversion and magnesium was positively correlated with feed conversion. This result means that feed conversion improved as calcium increased and got worse as magnesium increased. Calcium and potassium were negatively correlated with livability. The detrimental effect of these minerals may be related to interference during water vaccination, thus affecting livability. Interference during vaccination might also explain the positive correlation of calcium and nitrate with condemnation. We suggest testing the water for mineral content through State Diagnostic Laboratories. In addition, we recommend testing the water for bacterial contamination at the same time. Chlorination will be a cheaper first treatment in trying to correct water quality related performance problems. We also ask producers whether they are on a rural water system. In some cases, we suspect that high chlorine levels are interfering with field vaccinations. Water quality for broiler and turkey production will become increasingly important and hopefully will stimulate more research in future years.
ACKNOWLEDGMENT This research was supported in part by a grant from the Southeastern Poultry and Egg Association.
REFERENCES American Public Health Association, 1980. Standard Methods for the Examination of Water and Waste Water. 15th ed. American Public Health Association, Washington, DC.
Downloaded from http://ps.oxfordjournals.org/ at University of Manitoba on June 6, 2015
An examination of the correlation analysis in Table 2 for the broiler and turkey studies indicate some similarities as well as some differences. Calcium was negatively correlated with feed conversion in both studies. Higher calcium levels meant lower feed conversions or better feed conversions. Magnesium, bicarbonate, and hardness were also associated with improved feed conversions in the turkey study. However, magnesium was associated with poor feed conversions in the broiler study. The AI was also associated with improved feed conversions in the turkey study and is an indicator of the corrosivity in water. The AI is derived from the actual pH, calcium hardness, and total alkalinity. The AI is not a quantitative measure of corrosion but is a general indicator of the tendency for corrosion to occur. An AI of 12 or above indicates unaggressive (not corrosive) water. Aggressive index values of 10 to 12 indicate the water is moderately aggressive. Aggressive index values below 10 indicate extremely aggressive (corrosive) water. A higher AI value negatively correlated with feed conversion means better feed conversion with less aggressive water. The same situation occurred with adjusted weight in that less aggressive water was associated with higher growth rate. Adjusted weight was also positively correlated with dissolved oxygen, calcium, magnesium, and hardness in both studies. Additionally, bicarbonate was positively correlated with weight in the broiler study, whereas zinc and aggressive index were correlated in the turkey study. Nitrate was negatively correlated with adjusted weight in the broiler study but not in the turkey study. Calcium and potassium were negatively correlated with livability in the broiler study whereas magnesium was negatively correlated with livability in the turkey study. Condemnation was positively correlated with calcium and nitrate in the broiler study and with magnesium and AI in the turkey study. It is unclear why less aggressive water affected condemnation although it was beneficial to weight and feed conversion. It is also puzzling as to why potassium, zinc, nitrate, and phosphate were negatively correlated with condemnation in the turkey study.
SUMMARY