«Final Report of the Cold Compost Project Prepared by The Cornell Waste Management Institute Ithaca, NY Ellen Z. Harrison Director May 2004 Cold ...»
Enterococci – Both laboratories reported Enterococci as MPN/g dry weight in the later round of sampling, so these data were used to perform an ANOVA.
ANOVA revealed that Laboratory 1 reported significantly higher Enterococci counts than Laboratory 2 at a 95% confidence level. Laboratory 1 averaged log 3.707 (or 5093 MPN/g) and Laboratory 2 averaged log 2.979 (953 MPN/g).
Fecal Coliforms – For all sampling, both labs reported results in similar units, MPN/g dry weight. Fecal coliform is also the only microbe tested both in the early and late samples.
ANOVA revealed that Laboratory 1 reported significantly higher fecal coliform counts than Laboratory 2 at a confidence level of 95%. Laboratory 1 averaged log 3.562 (3648 MPN/g) and Laboratory 2 averaged 2.839 (690 MPN/g).
Salmonella – Both laboratories reported Salmonella as MPN/g dry weight in the later round of sampling, so these data were used to perform an ANOVA.
ANOVA revealed that Laboratory 2 reported significantly higher counts of Salmonella than Laboratory 2 at a confidence level of 95%. Laboratory 1 averaged 0.524 MPN/g Salmonella and Laboratory 2 averaged 4.867 MPN/g Salmonella.
We found that differences between labs for E. coli, Enterococci, fecal coliform, total coliform, and Enterococci were all statistically significant - the variation in samples between laboratories was high for each of the microorganisms. This led us to ask the question of whether using either laboratory’s data individually would be feasible. Again, we used ANOVA methods to look at each dataset individually and examine consistency of variation within and between samples. In
Cornell Waste Management InstituteCold Compost Project – Final Report
this instance, neither laboratory 1 nor laboratory 2 displayed a significant difference in variation within or between samples. While both were consistent when looked at separately, a final decision was made, based on knowledge that laboratory 2 had worked extensively with compost testing methodologies while laboratory 1 had not, to use only the dataset from laboratory 2 for the remaining analyses performed in this study.
Relation of Compost Physical Parameters to Microbial Concentration Researchers asked the question: Do physical parameters of small-scale compost piles influence the concentration of microbes? Multiple regression analysis was performed to derive a prediction equation for each type of microbe looked at in this study. It should be noted that because ANOVA found significant differences between labs when results were compared, and other reasons outlined in the discussion section, only laboratory 2 data were used in the regression analysis. This resulted in a small dataset – 18 samples in all. See Appendix F for more detailed results. No statistically significant relationship was found between the physical parameters and microbial concentrations.
Salmonella Y = – 7.812 – 0.188*X1 – 0.126*X2 – 0.098*X3 + 1.978*X4 + 0.144*X5 + 2.048*X6 – 0.269*X7 R2 = 0.628 Significance = 0.101 Clostridium log(Y) = 31.836 + 0.289*X1 – 1.136*X2 – 0.173*X3 – 9.959*X4 + 0.036*X5 – 0.109*X6 – 0.497*X7 R2 = 0.369 Significance = 0.582 Fecal Coliform log(Y) = – 1.294 + 0.014*X1 – 0.209*X2 + 0.016*X3 – 0.986*X4 + 0.050*X5 + 0.631*X6 – 0.079*X7 Cornell Waste Management Institute Cold Compost Project – Final Report R2 = 0.296 Significance = 0.744 Total Coliform log(Y) = – 18.260 – 0.489*X1 + 1.337*X2 + 0.017*X3 + 12.554*X4 + 0.018*X5 + 0.453*X6 + 0.128*X7 R2 = 0.568 Significance = 0.177 E. coli log(Y) = 7.670 + 0.147*X1 – 0.687*X2 – 0.054*X3 – 5.719*X4 + 0.088*X5 + 0.638*X6 – 0.078*X7 R2 = 0.422 Significance = 0.459 Enterococci log(Y) = – 21.316 – 0.352*X1 + 1.026*X2 + 0.062*X3 + 8.813*X4 + 0.032*X5 + 0.756*X6 + 0.147*X7 R2 = 0.368 Significance = 0.584 Fecal Strep log(Y) = – 18.623 – 0.354*X1 + 1.210*X2 + 0.136*X3 + 11.555*X4 – 0.067*X5 – 0.091*X6 + 0.159*X7 R2 = 0.372 Significance = 0.576 Correlations between Physical Compost Parameters and Microbial Concentrations Results of multiple regression analysis were obtained using data from laboratory 2 for the reasons stated above. OM, CN ratio, density, TKN, moisture, pH, and salts were considered as independent variables and the microbial concentrations were considered the dependent variable.
Of the seven microbes examined, only for Salmonella did the multiple regression show a significance level close to the cut-off of 0.1, or 90% (0.101).
TKN and pH were the independent variables with which Salmonella concentrations were most correlated. Higher pH and higher nitrogen was associated with higher levels of Salmonella.
Looking at slopes (bx) of each independent variable in the prediction equation for Salmonella, total nitrogen (b = +1.978) and pH (b = -2.048) have the strongest influence, while all other physical compost characteristics have b values near zero. While no significant correlation between physical compost characteristics and the other microbes was found, an examination of those analyses show that among the parameters, pH and TKN showed the strongest relationship, Cornell Waste Management Institute Cold Compost Project – Final Report for all of the different types of bacteria. This suggests that pH and TKN are important influences on microbial populations, with higher pH and TKN correlated with larger microbial populations.
Relationship among Microbes The data were analyzed to determine if the different microbes were correlated. The only correlations found were for microbes that were part of the same set of organisms. Thus total coliform was correlated with fecal coliform, for example. However no relation was found between the non-related microbes. This means that none of the microbes could be considered an appropriate indicator of general hygienic status.
Relationship of Compost Practices and Microbial Concentration To look at the relationship between microbial populations and management practices, including addition of pre and post consumer food waste, meat scrap addition, and turning, a series of independent samples t-tests were performed. Data for each microbe were grouped according to management practice and compared to look for differences.
The only significant difference found was between samples submitted to laboratory 1 when sorted according to whether meat scrap was added to the compost piles. Results showed that piles where no meat scrap was added actually had significantly higher E. coli counts than piles where meat scrap was added (log 3.48 and log 2.28, respectively). For detailed results, see Appendix G.
Discussion An unanticipated, but important, finding that came out of this study is that methodology for the analysis of composts is not standardized and is an important factor. As the results of ANOVA demonstrate, there were significant differences between laboratory 1 and laboratory 2. For all microbes measured, with the exception of total coliform, significant differences were found in results from the two labs. We suspect that this resulted from differences in methodology.
The discovery of the significant differences in results between laboratories, the change in reporting units for laboratory 1 midway through the project, and the greater consistency of results from laboratory 2 led us to use only those results in our further analysis. Thus the dataset Cornell Waste Management Institute Cold Compost Project – Final Report was half of what had been anticipated. While statistical analysis was performed on the dataset, the small sample size limits the accuracy and power of results obtained.
Results indicate that none of the microbes examined in this study are reliable indicators of compost hygiene in small-scale settings. However, the United States Environmental Protection Agency (USEPA) regulates Salmonella and fecal coliform concentrations in composted sewage sludges. While small-scale composts are in no way regulated, the figures provided by USEPA can be used as a benchmark to examine their hygienic quality compared to a set of established and frequently used criteria.
The limit set by USEPA for Salmonella spp. is less than 3 Most Probable Number (MPN)/4 grams of solid. The limit for fecal coliform concentration is less than 1000 MPN/gram of solid.
The USEPA regulations state that a sludge compost need only pass either Salmonella or fecal coliform to be suitable for use.
Among the early samples analyzed by laboratory 1, 5 of 32 had greater than 3 MPN/4 g, and 9 of 32 exceeded 1000 MPN. But most composts passed either Salmonella or fecal coliform.
Laboratory 2 did not report any samples in MPN/4 g, but each of the 18 samples analyzed fell below the level of detection, meaning Salmonella levels were still very low. 7 of 18 samples tested by laboratory 2 exceeded fecal coliform limits, but most composts fared well. Overall, 60% of the compost samples fell below 1000MPN/g of fecal coliform.
The finding of bacteria in the home-scale compost systems is not surprising since most systems are not highly managed. Consider that among the microbial groups tested in this study – total coliform and fecal coliform – disease-causing organisms posing a risk to humans represent only a small fraction of these. Add to this the fact that even within a genus such as Salmonella, there are multiple species, and even sub-species, and only a select few are pathogenic. In this study, all Salmonella were measured, but this doesn't say anything definitive about health risk.
Cornell Waste Management Institute Cold Compost Project – Final Report Another factor to consider is “infectious dose.” Even a pathogenic organism does not cause disease unless sufficient numbers are present. The dose that may cause disease will also vary with the susceptibility of the exposed person.
Background levels of microorganisms have been documented in a number of studies because of their importance in storm water contamination, land use practices, and other topics (Van Donsel et al 1967, Faust 1982, and Geldreich et al 1962). Background levels of microbes are an important factor that was not closely examined in this study, but are nonetheless important in understanding compost hygiene.
A 2002 study of large-scale composting facilities, sponsored by the Nordic Council of Ministers, examined several composting facilities taking in household waste, defined by the researchers as including meat scrap, soft yard waste and shredded biodegradable household items. While much larger in scale than sites examined in this current study, some important observations were made relating to the sanitization of composts made from household sources of material.
Researchers found consistently high concentrations of E. coli and Enterococcis in the end products of household waste composts that were actively composted for shorter periods of time, compared to those that composted longer. As a result, it was recommended that when high concentrations of coliforms are present in raw materials, more effective methods of thermophilic composting, and time, are needed to ensure pathogen reduction (Christensen et al., 2002).
The Nordic Council paper also suggests that fecal coliforms and Enterobacteriaceae may not be highly reliable indicators of pathogen reduction mainly because both represent very heterogeneous groups of organisms. For example, “fecal” coliforms found in raw materials of household based composts were in fact fecal in origin, whereas "fecal" coliforms in finished end product were not. The study authors support this statement by pointing out that E. coli, a known fecal coliform, was high in unfinished compost, but low or undetectable in finished products, while fecal coliforms were consistently high. In the case of Enterobacteriaceae, non-fecal species of this group are known to grow on decomposing plant matter found in finished composts (Christensen et al., 2002).
Cornell Waste Management InstituteCold Compost Project – Final Report
Conclusion Based on the results of this study, a review of current literature, and common sense, the following guidelines are suggested for use in small-scale compost settings to minimize any potential health risks (refer to Appendix H for a fact sheet on compost hygiene for small-scale systems). Small-scale on-site compost systems provide many environmental benefits. When good hygiene practices are used, the relative health risks are low.
1. Avoid certain inputs to the compost pile such as raw poultry or meat wastes, pet feces, and plate scrapings from people who are ill.
2. Consider managing your composting system to ensure that it gets and stays hot long enough to reduce pathogens. There are methods available for small-scale compost piles.
3. Practice good personal hygiene when handling compost. Proper personal sanitation is the most effective method for controlling the impact of any pathogens that may be in the compost. Wash hands after handling compost and/or use gloves. If the compost is particularly dusty, watering is an option.
4. Persons with weakened immune systems or medical conditions that compromise the body’s ability to fight infection should use caution when handling compost.
5. If possible, allow composts that are produced in a small-scale setting to age for at least a year before use.
Cornell Waste Management InstituteCold Compost Project – Final Report
References Bollen, G.J. (1990) Composting of Agricultural and Other Wastes. (J.R. Gasser, Ed., Elsevier Appli. Pub. Amsterdam,) [incomplete reference] Christensen, K.K., M. Carlsbaek, E. Norgaard, K.H. Warberg, O. Venelampi, and M. Brogger.
(2002) Supervision of the sanitary quality of composting in the Nordic countries: evaluation of 16 full-scale facilities. Nordic Council of Ministers, Environment TemaNord 2002: 567.
Cooperband, L.R. and J.H. Middleton. (1996) Changes in Chemical, Physical, and Biological Properties of Passively-aerated Cocomposted Poultry Litter and Municipal Solid Waste Compost. Compost Science and Utilization 4(4): 24-34.
Droffner, M.L. and W.F. Brinton. (1994) Evidence for the Prominence of Different Well Characterized Gram Negative Mesophilic Bacteria in the Thermophilic (50-70C) Environment of Composts. (Abs.) American Society for Microbiology Q-266, Las Vegas, Nevada.
Droffner, M.L. and W.F. Brinton. (1995) Survival of E. coli and Salmonella populations in aerobic thermophilic composts as measured with DNA gene probes. Zbl. Hyg. 197, 387/?