BioCycle December 2009, Vol. 50, No. 12, p. 28
Authors develop model to predict the likelihood of compost phytotoxicity, providing producers and end users a tool to make informed compost and soil management decisions.
Danielle Aslam and Jean VanderGheynst

PHYTOTOXICITY of compost related to the stability of the product is an ongoing consideration for producers and end users. Questions to consider when addressing this issue include: What is the impact of letting the material stabilize prior to using it as a soil amendment? How much of an impact will an adjustment in amendment rate have on phytotoxicity? How can compost stability measurements be used to make management decisions to prevent phytotoxicity? To answer these questions, tools for measuring and predicting phytotoxicity from stability estimates are needed.
Phytotoxicity refers to the “intoxication of living plants by substances present in the growth medium, when these substances are taken up and accumulated in plant tissue” (Araujo and Monteiro, 2005). Many methods have been utilized to measure compost phytotoxicity including: electrical conductivity, pH, respiration, plant assays, nitrogen species, humification ratio, C/N, reactive carbon, cation exchange capacity, hemicellulose and cellulose, physical properties and optical density.
However, there is no universally accepted method for measuring compost phytotoxicity. This is due in part to the multiple factors that can contribute to phytotoxicity, including volatile fatty acids (e.g., acetic), ammonia, pH, heavy metals, CO2, trace elements, phenol, salinity, ethylene oxide, nitrogen immobilization and anaerobic conditions.
One goal of our laboratory research at the University of California at Davis was to develop a set of tools to help compost producers and end users evaluate phytotoxicity-related management decisions. Based on this research, a model was developed to predict the likelihood that compost will be phytotoxic depending on its amendment rate, time in the field before planting and stability.

COMPOSTS STUDIED
For the results to apply to a variety of combinations of soil type, compost type and amendment levels, the experiments used both sandy loam and potting soils, each amended with 5 percent and 50 percent (by volume) food waste and green waste composts. There was also a soil control for each soil type. Food waste and green waste compost samples were obtained from Jepsen Prairie Organics in Dixon, California in December 2004.
At the time the study was completed, the facility received food waste from residential and commercial sectors including restaurants, delis, markets, coffee shops, hotels and bakeries. Jepsen Prairie Organics composted the food waste under conditions of intermittent aeration in large, aerated plastic silage bags for 30 days and then in windrows, which were turned and watered twice a week for an additional 30 days. Green waste from surrounding areas was composted by itself in windrows for 60 days.

PLANT ASSAYS
Plant assays have long been used in evaluating compost phytotoxicity. These employ a variety of species, planting substrates and phytotoxicity evaluation methods. Plant assays are especially beneficial because they can measure the compounded effects of various phytotoxic factors. However, they are not always consistently or sufficiently sensitive and can be time consuming.
The plant assay method selected for this research project used direct seeding into compost and soil mixtures for phytotoxicity experimentation. This method was chosen over the more common water extract method based on the difficulty of obtaining water extracts from the composts, as well as to simulate agricultural conditions. This assay was used to select a sensitive seed species; garden cress, radish, Chinese cabbage and lettuce were evaluated. The seed evaluation assays used 50 percent food waste compost in sandy loam soil to simulate horticultural amendment, and 5 percent food waste compost in sandy loam soil to simulate field amendment (Figure 1).
Compost literature has often shown food waste to be phytotoxic in higher concentrations. Thus, it would be expected that the food waste compost would be more phytotoxic at the 50 percent amendment rate than at the 5 percent amendment. This was true for all the seeds examined (Figure 2). However, we wanted to select a seed that also would be sensitive to low levels of compost amendment. Because cress was the only seed tested that showed a difference in phytotoxicity between the 5 percent amendment rate and soil control, it was identified as the most sensitive seed species (Figure 2).
Additional experiments showed that this assay was indeed sensitive to phytotoxic compounds, however it was still subject to variability, which pointed to the need for a less variable phytotoxicity indicator. Furthermore, cress is a tiny seed that blends in well with soil and compost, so it was time consuming to find seeds and determine whether they had germinated. (More information about the plant assay research can be found in a paper in Environmental Engineering Science (2008). Other journal papers on this research are cited in the references.)

READILY MEASURABLE COMPOST PROPERTY
Relating this assay to a readily measurable compost property could reduce the effort needed to evaluate compost phytotoxicity. Because respiration is often measured by compost producers and end users to determine compost stability, it was decided to test whether or not it could be related to compost assay phytotoxicity.
Previous research in composting suggests that respiration rate is related to compost phytotoxicity. As compost feedstocks proceed through the decomposition process, the demand for O2 decreases. This results in a decrease in respiration rate. A lower demand for O2 reduces the likelihood that anaerobic conditions will develop in compost or in compost-amended soil. Anaerobic conditions in compost-amended soil can lead to the production of phytotoxic compounds, such as volatile fatty acids. As compost decomposes further, these volatile fatty acids are often degraded by microorganisms and it is less likely that they will form again due to the lower demand for O2.
Respiration rates from compost samples were measured using respirometers (Figure 3). Respiration rate data (Figure 4a) were used to estimate compost stability and degradable carbon in the compost (Figure 4b). For the food waste and green waste compost investigated in this study, total degradable carbon in the compost was approximately 370 mg CO2/g total solids (TS) compost and 100 mg CO2/g TS compost, respectively. Respiration rate measurements were also collected on aerated soil and compost mixtures incubated at various temperatures between 20°C and 45°C and were used to determine remaining degradable carbon in the composts over time. To measure phytotoxicity, samples of the incubated compost and soil mixtures were taken every few days and examined using the direct-seeding cress assay method described previously.
Next, phytotoxicity in the plant assay was compared to the respiration measurements and estimates of remaining degradable carbon of the compost mixtures. The strongest and most useful correlation existed between cress seed germination and the remaining degradable compost carbon in the soil (Figure 5). A sigmoidal equation was used to represent this correlation.
The remaining degradable compost carbon is the amount of carbon available for microorganisms to degrade. It is quantified by subtracting the amount of carbon decomposed during stabilization from the amount of degradable carbon present when compost was first amended to the soil. The amount of compost carbon decomposition is a function of soil temperature, time and the amount of compost amended.
A simple model was developed and validated that predicts degradable compost carbon remaining in soil as a function of time: where k is a rate constant that varies with temperature. The model is independent of soil and compost type.

APPLYING MODELS TO SOIL MANAGEMENT DECISIONS
The phytotoxicity and compost decomposition models allow compost producers to tailor production to achieve a certain level of degradable carbon in their product and allow end users to adjust amendment rates and soil stabilization times to minimize phytotoxicity. When using the developed models, the appropriate phytotoxicity threshold should be selected based on crop tolerance. Figure 6 provides phytotoxicity as a function of stabilization time for various amendment rates of degradable compost carbon in soil. The curves in the figure were generated using the correlation in Figure 5 and the first-order kinetic equation. The figure assumes a soil temperature of 25°C.
A compost amendment rate of 5 g CO2-C/L soil would be equivalent to amending soil with approximately 4 percent (v/v) of the food waste compost investigated in this research (Figure 4), which had a bulk density of 350 g TS/L. A 4 percent (v/v) amendment rate of this compost would be equal to approximately 4 to 5 dry tons/acre assuming an incorporation depth of 2 to 4 inches. At this amendment rate, initial phytotoxicity would be 38 percent; a stabilization time at 25°C would need to be approximately 35 days to reduce phytotoxicity to 10 percent.
A compost amendment rate of 0.5 g CO2-C/L soil would be equivalent to amending soil with approximately 1 to 2 percent (v/v) of the green waste compost investigated in this research (Figure 4), which had a dry weight bulk density of 340 g TS/L. A 1 to 2 percent (v/v) amendment rate would be equal to approximately 1.5 dry tons/acre assuming an incorporation depth of 2 to 4 inches. At a phytotoxicity threshold of 10 percent, no stabilization time would be needed to prevent phytotoxicity.
Amendment levels greater than 10 g CO2-C/L soil might be encountered in compost-based potting mixes. An amendment level of 10 g CO2-C/L soil would be equivalent to a potting mix containing approximately 25 percent (v/v) of the green waste compost investigated in this research. The initial phytotoxicity of the green waste potting mix would be about 55 percent and would require approximately 40 days of stabilization at 25°C to reduce the phytotoxicity to 10 percent.
The remaining degradable carbon in compost amended to soil can be used to predict the likelihood of compost stability-related phytotoxicity using the approach described earlier and in Figure 6. It may not be able to predict phytotoxicity resulting from persistent conditions such as salts and herbicides. However, additional experiments on these composts showed that both cress germination and remaining degradable carbon of compost amended soils were well-correlated with the amount of total dissolved salts, measured by electrical conductivity. The dissolved salts contributing to electrical conductivity and corresponding phytotoxicity in the composts could have been from a variety of sources including inorganic salts associated with the soils and composts, organics acids, ammonium and nitrate. Organic acids are common in immature food waste composts and could very well have contributed to the elevated electrical conductivity levels of the food waste compost studied here.
Although all possible causes of compost phytotoxicity are not directly related to compost stability measurements, this research concluded they may be indirectly related. Thus, we believe that our model, based on the remaining degradable carbon of compost amended soils, will be useful for compost producers and end users in predicting the likelihood of compost phytotoxicity, allowing them to make informed compost and soil management decisions.

Danielle Aslam is with the California Integrated Waste Management Board, Local Assistance & Market Development-South Branch in Long Beach, California. Jean VanderGheynst is with the University of California, Davis’ Department of Biological & Agricultural Engineering.

REFERENCES
Araujo, A.S.F., Monteiro, R.T.R. (2005) Plant bioassays to assess toxicity of textile sludge compost. Scientia Agricola, 62(3): 286-290.
Aslam, D.N., VanderGheynst, J.S. (2008) Predicting phytotoxicity of compost-amended soil from compost stability measurements. Environmental Engineering Science, 25(1):81-90.
Aslam, D.N., VanderGheynst, J.S., and Rumsey, T.R. (2008) Development of models for predicting carbon mineralization and associated phytotoxicity in compost-amended soil. Bioresource Technology,99:8735-8741.
Aslam, D.N., Horwath, W., and VanderGheynst, J.S. (2008) Comparison of several maturity indicators for estimating phytotoxicity of compost-amended soil. Waste Management, 28:2070-2076.