This is the submitted version of the following article: Young, BJ, et al. Development of phytotoxicity indexes and their correlation with ecotoxicological, stability and physicochemical parameters during passive composting of poultry manure in Waste management (Ed. Elsevier), vol. 54 (Aug. 2016), p. 101-109, which has been published in final form at: DOI 10.1016/j.wasman.2016.05.001 © 2016. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/ 1 Development of phytotoxicity indexes and their correlation with ecotoxicological, stability and 2 physicochemical parameters during passive composting of poultry manure 3 4 Brian Jonathan Young(1), Pedro Federico Rizzo(1), Nicolás Iván Riera(1), Virginia Della Torre(1), Valeria 5 Alejandra López(1), Cecilia Denisse Molina(1), Florencia Estefanía Fernández(1), Diana Cristina Crespo(1), 6 Raquel Barrena(2), Dimitrios Komilis*(2), Antoni Sánchez(2) 7 (1) 8 Laboratorio de Transformación de Residuos, Instituto de Microbiología y Zoología Agrícola 9 (IMYZA), Instituto Nacional de Tecnología Agropecuaria (INTA), Las Cabañas y De los Reseros s/n, 10 Hurlingham, Buenos Aires, Argentina. (2) 11 Departamento de Ingeniería Química, Universitat Autònoma de Barcelona, 08193 Bellaterra 12 (Cerdanyola del Vallès), Spain. 13 14 *Corresponding author. E-mail address: Dimitrios.Komilis@uab.cat; Phone number: +34 93-581-4796. 15 1 16 Abstract 17 18 Both raw and composted poultry manure is applied as soil amendment. The aims of this study were: 1) to 19 develop phytotoxicity indexes for organic wastes and composts, and 2) to assess the correlation among 20 phytotoxicity indexes, ecotoxicological endpoints and stability and physicochemical parameters during 21 passive composting of poultry manure. Six 2-m3 composting piles were constructed and four parameter 22 groups (physicochemical and microbiological parameters, ecotoxicological endpoints, and biological 23 activity) were determined at four sampling times during 92 days. Extracts were used to carry out acute 24 toxicity tests on Daphnia magna, Lactuca sativa and Raphanus sativus. Composting decreased average 25 toxicity 22.8% for the 3 species and D. magna was the most sensitive species. The static respiration 26 index decreased from 1.12 to 0.46 mg O2 g OM h-1 whilst organic matter reduced by 64.1% at the end of 27 the process. Escherichia coli colonies remained higher than values recommended by international 28 guidelines. The D. magna immobilization test allowed the assessment of possible leachate or run-off 29 toxicity. The new phytotoxicity indexes (RGIC0.8 and GIC80%), proposed in this study, as well as salinity, 30 proved to be good maturity indicators. Hence, these phytotoxicity indexes could be implemented in 31 monitoring strategies as useful ecotoxicological tools. Multivariate analyses demonstrated positive 32 correlations between ecotoxicological endpoints (low toxicity) and biological activity (stability). These 33 two parameter groups were associated at the final sampling time and showed negative correlations with 34 several physicochemical parameters (organic and inorganic contents). The final poultry manure compost 35 was rendered stable, but immature and, thus, unsuitable for soil amending. 36 37 Keywords: Daphnia magna, Lactuca sativa, maturity, phytotoxicity, Raphanus sativus, stability. 38 2 39 1. Introduction 40 Poultry production is a quantitatively important agro industry worldwide. World production of eggs has 41 increased 24.98% between 2001 and 2011, according to statistics from Food and Agriculture 42 Organization of the United Nations. During the same period, egg production in Argentina soared 101.7% 43 (FAOSTAT, 2013). Consequently, a large volume of waste (poultry manure) is an inevitable side effect 44 of this increase. 45 Commonly, raw poultry manure is applied to farmland as organic amendment (Bolan et al., 2010). This 46 waste contains nutrients (N, P, K), heavy metals (As, Pb, Zn, Ni, Cd, Cu, Mn), xanthophylls, antibiotics, 47 antiprotozoals, antioxidants, mold inhibitors, probiotics, polychlorinated phenols, tetrachlorodibenzo-p- 48 dioxin and hormones (Frank et al., 1988; Jackson et al., 2003) that may impact negatively the ecosystem 49 through leaching and runoff. These adverse effects on ecosystem can be circumvented with low-cost 50 composting. Composting minimizes the concentration of phytotoxic substances, controls the spread of 51 pathogens, improves storage and handling of waste, and reduces unpleasant odors (Edwards and Daniel, 52 1992). The quality of the compost may have either a positive or negative impact on both soil fertility and 53 plant health. Nutrients loss depends on several factors such as aeration, moisture content, temperature 54 and carbon-to-nitrogen (C:N) ratio (Ogunwande et al., 2008). The initial C:N ratio is the most widely 55 used parameter for deciding composting methodology. Poultry manure contains high nitrogen content. 56 Therefore, the degradation process may be improved adding carbon-rich materials (Petric et al., 2009). 57 High initial C:N ratio causes longer composting time (Tuomela et al., 2000), whereas low initial C:N 58 ratio generates higher emission of volatile gases and leachates (Tiquia and Tam, 2000b). Co-composting 59 of poultry manure with other agricultural wastes improves the physicochemical characteristics and 60 reduces the phytotoxicity (Rizzo et al., 2013). Composting may be aerated by passive or active systems 61 (Ogunwande, 2011). Passive aeration is an effective inexpensive treatment system for co-composting 62 poultry manure, poultry litter and sawdust, according to Ogunwande and Osunade (2011), and is more 3 63 cost-effective than active aeration systems in terms of initial capital investment, operation, maintenance, 64 and operator training costs (Solano et al., 2001). 65 The use of toxicity tests on aquatic and terrestrial organisms allows the integral assessment of the waste 66 before disposal. Several authors have reported acute toxicity on several organisms exposed to raw (Gupta 67 and Kelly, 1992; Gupta et al., 1997) and composted poultry manure (Komilis and Tziouvaras, 2009). D. 68 magna has demonstrated a good sensitivity to assess toxicity of landfill leachate (Matejczyk et al., 2011; 69 Pablos et al. 2011), hazardous wastes (Pablos et al., 2009) and municipal solid waste leachate (Isidori et 70 al., 2003; Bortolotto et al., 2009). Germination index (GI) is the phytotoxicity index used commonly to 71 assess toxicity from complex solid samples, such as waste or compost. However, there is a lack of 72 ecotoxicological tools in the literature when a material demonstrates high toxicity. In addition, integral 73 monitoring strategies have not previously been used to study a passive aeration composting of poultry 74 manure with low quantities of carbon-rich materials. The aims of this study were: 1) to develop 75 phytotoxicity indexes for waste or compost samples, and 2) to assess the correlation between 76 ecotoxicological endpoints with both stability and physicochemical parameters during passive aeration 77 composting of poultry manure. A seed toxicity test was used to assess effects on a terrestrial plant since 78 our objective was to develop the phytotoxicity indexes. On the other hand, D. magna test was selected to 79 assess possible leachate or run-off toxicity, since it is a standard toxicity test widely used in monitoring 80 programs of different kind of samples. 81 82 2. Materials and methods 83 84 2.1 Composting 85 2.1.1 Experiment 86 Poultry manure was collected in an automatized farm of the Zucami® type, located in Hurlingham, 87 Argentina. Six 2-m3 (1m x1m x1.2m) composting piles were built mixing poultry manure with dry grass 4 88 (7:3 v/v) in an experimental field of the National Institute of Agricultural Technology (INTA), 89 Hurlingham, Argentina. Composting piles had an initial C:N ratio of 24.6 ± 3.6:1 and moisture content of 90 70.6 ± 3.2% wb. In addition, wood chips were added as bulking agent. A static pile with a passive 91 aeration system and V-shaped pipe configuration was used as composting method, according to 92 Ogunwande (2011). A pipe with 35-mm diameter perforations was used, as recommended by 93 Ogunwande and Osunade (2011). 94 Experimental design consisted of a completely randomized statistical design with 6 composting piles 95 (n=6) and repeated measurements. Each composting pile was the experimental unit (n = 6 piles). 96 Sampling was done by quartering from each pile at days 0, 14, 56 and 92 (n = 24), according to 97 standardized specifications (USDA and USCC, 2001).Three sub-samples were taken from each 98 composting pile and at each sampling time and were kept at 4 ºC until analysis. 99 100 2.1.2 Physicochemical and microbiological characterization 101 The following physicochemical parameters were evaluated: ambient and pile temperature, moisture 102 content (MC), organic matter (OM), total organic carbon (TOC), total Kjeldhal nitrogen (TN), C:N ratio, 103 soluble organic carbon (SC), total phosphorous (TP), soluble phosphorous (SP), major cations, metals, 104 pH, and electrical conductivity (EC), according to standard methods (USDA and USCC, 2001). The 105 major cations (Ca, Mg, K, Na) and metals (Zn, Mn, Cu) were quantified using an atomic absorption 106 spectrophotometer (Varian model 220 A). The percentages of OM and TN losses were calculated using 107 Eq. 1 and 2, according to Paredes et al. (2000). 108  X (100 - X 2 )  109 OM loss (%) = 100 - 100 ×  1  Equation 1  X 2 (100 - X1 )  X N  110 TN loss (%) = 100 - 100 ×  1 2  Equation 2  X 2 N1  111 5 112 where N1 and N2 are the initial and final TN percentages, and X1 and X2 are the initial and final ash 113 percentages, respectively. 114 Commercial kits (Rida Count®) were used for microbiological characterization to determine total 115 coliforms, E. coli and Salmonella spp. (CFU g-1) provided by R-Biopharm AG. 116 Biological activity was measured using the static respiration index (SRI) (Iannoti et al., 1993; USDA 117 and USCC, 2001). This technique is a static respiration stability assessment method which is performed 118 in mesophilic temperatures (37 °C) with sealed 500 mL flasks. An electrochemical dissolved O2 119 electrode is placed in the headspace of the flask and records the O2 air concentration drops within the 120 flask. Oxygen uptake rate (OUR) is finally expressed in mg O2 g-1 OM h-1 and is calculated via the slope 121 of the O2 concentration drop. The SRI is the maximum averaged OUR calculated during a 24 hour period 122 (after the initial lag time). 123 124 2.2 Toxicity tests 125 In order to simulate the mixture of water-extractable substances present in leachate or runoff, aqueous 126 extracts were prepared mixing a dry sample with deionized water (1:10 w/v). These extracts were stirred 127 at room temperature (23±2 ºC), according to a procedure described by Tiquia et al. (1996). 128 129 2.2.1 Organisms 130 Two species of plants and an aquatic crustacean were used as test organisms. A non-chemically treated 131 seed lot of lettuce (L. sativa variety “Carilauquen INTA”) and radish (R. sativus variety “Puntas 132 blancas”) were obtained from the experimental stations of the INTA, located in La Consulta and San 133 Juan cities, Argentina, respectively. Seeds were kept in a dry environment at 4 ºC. 134 The aquatic crustacean D. magna was reared in a laboratory of ecotoxicology (IMYZA, INTA). The 135 population of daphnid was fed 3-4 times per week with a mixture of several species of algae, under 6 136 controlled conditions (23 ± 2 ºC and 16L:8D). Dechlorinated and aerated tap water (pH = 8.1 ± 0.3; EC = 137 642 ± 24 µS cm-1) was used as culture medium. 138 139 2.2.2 Seed germination and root elongation test 140 Seed germination and root elongation tests were carried out at 22 ± 2 ºC in darkness for 120-h, according 141 to standardized protocols (Sobrero and Ronco, 2004). Experimental design consisted of 10 treatments 142 (i.e. 9 different concentrations of the extracts and a control group) per composting pile and per sampling 143 time (n = 240) using triplicates. The extract concentrations used in the tests ranged from 0.5% to 100% 144 v/v (0.5, 1; 5, 10, 20, 40, 60, 80, and 100%); deionized water was used as a negative control and zinc 145 chloride solutions as positive control. A total of 15 tests were conducted using lettuce and 18 tests using 146 radish. Fifteen seeds of each the species (radish and lettuce) were exposed to 4-mL of each of the nine 147 extract concentrations and control water in 90-mm diameter Petri dishes lined with filter paper (Munktell 148 AB Box 300, SE-790 20 GRYCKSBO, Sweden). A total of 10800 seeds of each species were used in 149 these experiments. The quality controls used were percentage of germination over 90%, coefficient of 150 variation for root elongation below 30%, in negative controls, whilst Zn (zinc chloride) was used as a 151 reference toxic in positive controls. The zinc chloride concentrations at each positive control were: 18.75, 152 37.5, 75, 150, 300 mg Zn L-1. 153 Toxicity endpoints assessed were seed germination and root elongation (Inhibition concentration 50 154 [IC50, no-observed-effect concentration [NOEC], lowest-observed-effect concentration [LOEC], relative 155 growth index [RGI], and germination index [GI]). Alterations in germination and normal development of 156 seedlings were recorded. The root elongation length was used to calculate RGI (Eq. 3) (Alvarenga et al., 157 2007). RGI values between 0 and 0.8 are categorized as inhibition of root elongation (I), values greater 158 than 0.8 and less than 1.2 as no-significant-effect (NSE), and values greater than 1.2 as stimulation of 159 root elongation (S) (Young et al., 2012). The number of germinated seeds and root elongation length 7 160 were used to calculate GI (Eq. 4) (Zucconi et al., 1981). GI values lower than 80% were considered to 161 indicate inhibition (Tiquia et al., 1996). RLS 162 RGI = Equation 3 RLC RLS GSS 163 GI % = × ×100 Equation 4 RLC GSC 164 where RLS is the radicle length of the sample, RLC is the radicle length of the control, GSS is the 165 number of germinated seeds in the sample and GSC is the number of germinated seeds in the control. 166 Two phytotoxicity indexes (RGIC0.8 and GIC80%) are proposed herein to assess the maturity of 167 composted manure. RGIC0.8 estimates the lowest extract concentration to get an inhibition of root 168 elongation (RGI = 0.8). GIC80% estimates the lowest extract concentration to get a response of 80% in 169 GI. The validation process of these new phytotoxicity indexes was conducted using published and 170 unpublished data of our group from several types of samples. Phytotoxicity indexes were applied to data 171 of four samples of compost and two samples of effluents. The poultry manure derived compost (PMC) 172 was produced after a period of 12 weeks, according to Rizzo et al. (2013). Poultry manure was mixed 173 with corn bare cobs, sawdust and shavings. Composting piles were manually turned. The poultry litter 174 and horse manure derived compost (PLHMC) was composted in an experimental field of the INTA after 175 a period of 16 weeks, according to Riera et al. (2014). Poultry litter contained a mixture composed by 176 poultry manure, feathers, spilled feed, and bedding material. Active aeration composting was obtained in 177 manually turned bins. The municipal solid waste derived compost (MSW1) was obtained from a 178 composting facility in Trenque Lauquen (Argentina). Organic fraction of MSW was separated at home 179 and then composted in the plant for 16 weeks. Active aeration composting was conducted in manually 180 and mechanically turned piles. Other municipal solid waste derived compost (MSW2) was obtained from 181 a composting facility in Metropolitan Area of Buenos Aires (Argentina). Organic fraction of MSW was 182 manually and mechanically separated within the plant. Active aeration composting was obtained in 183 manually turned bins after a period of 11 weeks. The samples of the untreated and treated effluent were 184 collected in the treatment system from an anaerobic bioreactor, according to Young et al. (2012). The 8 185 anaerobic bioreactor was loaded daily with 35 kg of cereal residues and 125 L of treated effluents. 186 Untreated and treated effluents were obtained from the inflow into the first treatment pond and the 187 recirculated flow to the bioreactor respectively. 188 Values of the phytotoxicity indexes (RGIC0.8 and GIC80%) were differentiated into two categories 189 according to the toxicity effects observed: 190 - Inhibitory effects: ≤ 100% 191 - Non-inhibitory effects: > 100% 192 193 2.2.3 D. magna immobilization test 194 The Daphnia immobilization test was used to assess the acute toxicity from composting extracts 195 (USEPA, 1996). Toxicity tests were carried out by triplicate. Experimental design consisted in 10 196 treatments for each composting pile and sampling time (n = 240). Extract concentrations used in the tests 197 ranged from 0.1% to 80% v/v (0.1, 1, 4, 8, 15, 25, 40, 60, and 80%), and a negative control. Ten neonates 198 less than 24-h of hatching were exposed during 48-h in a static system, containing 30-mL of each of nine 199 the extract concentration or control water. A total of 7200 daphnids were used in these experiments. 200 Toxicity endpoints assessed were effective concentration 50 (EC50), NOEC, and LOEC. The quality 201 controls used were immobilization under 10% in negative controls and Cr (potassium dichromate) as 202 reference toxic in positive controls. The potassium dichromate concentrations at each positive control 203 were: 0.1, 0.2, 0.3, 0.4 and 0.5 mg Cr L-1. 204 205 2.3 Statistical analysis 206 The temporal variation of parameters was assessed by one-way ANOVA. When the F values of the 207 ANOVA were significant (p<0.05), means were compared by the Tukey’s pair wise test. The influence 208 of physicochemical parameters on the ecotoxicological endpoints and biological activity was also 9 209 assessed by multivariate statistical procedures, such as principal component analysis (PCA) and 210 correlation analysis (Pearson correlation coefficient). 211 212 3. Results and discussion 213 214 3.1 Composting 215 3.1.1 Physicochemical characterization 216 The variation of the ambient and pile temperature profiles showed a similar tendency from 40 days (Fig. 217 1). As was reported by other authors, two main phases can be seen in the temperature profile of 218 composting piles. The average maximum temperature of the piles ranged between 40 and 46°C and 219 lasted for five days, as shown in Fig. 1 (top). However, some piles reached a maximum of 60.5° C. The 220 maturation phase started from day 40, when the temperature of the piles was similar to ambient 221 temperature. Passive aeration systems reach lower temperatures than active aeration systems (Barrington 222 et al., 2003). Silva et al. (2009) reported that co-composting of poultry manure with low quantities of 223 carbon-rich materials (80:20 ratio) reached a maximum pile temperature lower than 40ºC. However, 224 Ogunwande and Osunade (2011) compared three passive aeration composting of poultry manure that 225 reached a thermophilic phase above 42 °C that lasted for approximately 20 days. However, this longer 226 thermophilic phase could be due the initial composition of the composting piles. Ogunwande and 227 Osunade (2011) evaluated composting with sawdust, poultry manure and litter. Poultry manure is 228 characterized by a high relative density, whereas sawdust and poultry litter are materials with low density 229 that could improve the total porosity of the mix. In this study, the initial composition had a high 230 proportion of poultry manure (70%) which could have affected the porosity and thus oxygen diffusion. 231 Both the organic and inorganic content decreased (OM = 34.8% and EC = 54.5%; p<0.05) during the 232 biodegradation period. Other parameters showed a significant decrease as well, such as MC, TOC, SC, 233 TN, Ca and Mg (Table 1). The MC was kept above 60% by manual irrigation. The pH was remained 10 234 slightly alkaline from day 14, then increased and the final pH was less than 9. This increase could be 235 attributable to proteolysis and high microbial activity during the first days of composting (Bustamante et 236 al., 2008). Authors reported similar pH values using passive and active aeration systems (Ogunwande 237 and Osunade, 2011; Rizzo et al., 2013). The high initial values of EC could be associated to the poultry 238 diet (Bolan et al., 2010). Although EC decreased, the compost obtained had restrictions in use due to a 239 high EC final value. Active aeration systems may reach a higher decrease of EC due to salt loss by 240 higher leaching (Rizzo et al., 2013). Such high EC values in poultry manure compost were found by 241 Komilis and Tziouvaras (2009). 242 The highest losses of OM and TN were registered during the first 14 days (Fig. 1-middle). The 243 cumulative losses of OM and TN at day 92 were of 64.1 ± 2.1 % and 68.1 ± 10.4% respectively. Also, a 244 positive correlation between OM and EC (R2 = 0.77) was found. TN loss was associated with the gradual 245 increase of pH during the first 14 days (52.1%), which could increase the volatilization of N-NH3+. 246 Ogunwande (2011) compared three passive aeration systems and reported a TN loss of 38.1% until day 247 14, lower than those found in this study. Tiquia and Tam (2002) reported similar losses of TN (58%) 248 using a forced-aeration system for composting of poultry litter. On the other hand, Parkinson et al. (2004) 249 found a higher TN loss in active aeration system than in passive aeration system. Both moderate 250 temperatures and presence of microbial groups that increase and / or maintain the pool of N, such as N- 251 fixing and nitrifying bacteria (Paredes et al., 1996). It could have caused a decrease in TN loss during the 252 mesophilic phase. 253 254 3.1.2 Stability and microbiological contents 255 The highest pile temperature and biological activity (SRI = 1.12 mg O2 g OM h-1) were recorded during 256 the first 14 days (Fig. 1). The SRI showed a negative correlation with SC and Mn (R2: 0.92 and 0.76 257 respectively; Table 3), whereas it showed a slight positive correlation with TP (R2: 0.65; p<0.05). Low 258 values of SRI at end of the process (day 92) indicated that biological stability (SRI ≤ 0.5 mg O2 g-1 OM 11 259 h-1) was reached. Salmonella spp. was not detected during composting. However, the high counts of total 260 coliforms and E. coli observed in all piles and at all sampling times suggest that these pathogens survived 261 the short thermophilic phase of the process, which indicates the low quality of the derived compost. 262 263 3.2 Toxicity tests 264 3.2.1 Quality controls 265 Results of the toxicity tests were acceptable according to the criteria established by the quality controls. 266 In the seed tests, the coefficients of variation between the averages of root length in the negative controls 267 were 19.6 and 23.3% for lettuce and radish, respectively, lower than that recommended in the test 268 protocols. The IC50 average values of root elongation in the positive controls were 55.4 ± 16.9 (n = 15) 269 and 82.6 ± 15.1 (n = 18) mg L-1 of Zn for lettuce and radish respectively. On the other hand, the average 270 value of immobilized neonates of D. magna in the negative controls was 2.2%, lower than that 271 recommended in the test protocols. The EC50 average value in the positive controls was 0.30 ± 0.07 (n = 272 21) mg L-1 of Cr. 273 274 3.2.2 Exposure to extracts 275 Toxicity tests carried out on terrestrial plant species (lettuce and radish) allowed determining the quality 276 of the compost as a soil amendment, whereas on the aquatic organism (daphnid) allowed determining the 277 potential toxicity of leachates or runoff. The three organisms exposed to aqueous extracts showed acute 278 toxicity in all samples. Ecotoxicological endpoints of the test organisms at each sampling time can be 279 found in Table 2. The average EC50 or IC50 of the 3 species was 8.29 ± 0.35% (n=18) in the initial 280 sampling (day 0) and 31.12 ± 10.99% (n=18) in the final sampling (day 92). Composting reduced the 281 average toxicity by 22.8% for the 3 species. The sensitivity of the organisms measured in terms of EC50 282 or IC50 was highest for daphnid, followed by lettuce and then radish. Rizzo et al. (2013) also found 283 lettuce to be more sensitive to adverse effects than radish. Endpoints of immobilization (D. magna) and 12 284 root elongation (L. sativa and R. sativus) exhibited toxic response in all samples. Delgado et al. (2013) 285 reported high mortality on daphnid exposed to poultry manure leachates. Root elongation was an 286 endpoint with most sensitivity than seed germination for the both plant species, as reported by Fuentes et 287 al. (2004). Seed germination exhibited no toxic response in 17 and 33% of the samples for LOEC, NOEC 288 and IC50 respectively at day14 and 56 for lettuce and at day 0 for radish. In addition, this endpoint 289 exhibited no toxic response from day 14 for radish. Several authors have reported the genotoxicity of 290 leachate landfill and compost extracts on terrestrial plants and bacteria (Cabrera et al., 1999; De Simone 291 et al., 2005; Kwasniewska et al., 2012). Gupta and Kelly (1992) demonstrated that poultry litter leachate 292 may induce mutagenicity using the Ames test. Further studies could focus on assessing the capability of 293 composting to reduce the genotoxicity of poultry manure. 294 295 Insert Figure 1 296 297 298 299 Ins 300 301 3.2.3 Phytotoxicity indexes 302 RGI and GI have been used to assess the toxicity of composting samples (Tiquia and Tam, 2000a; 303 Tiquia, 2010). The results obtained in the present study are similar to those reported by other authors. GI 304 values in the 100% extract concentration were zero in 62.5% of the samples for lettuce (average GI value 305 = 3.88%; n= 24) and in 16.6% of samples for radish (average GI value =21.07%; n= 24). Komilis and 306 Tziouvaras (2009), for example, found GI values between 0 and 6% in extract concentration of 100% 307 (raw extract) of poultry manure derived compost using radish, lettuce, pepper (Capsicum spp.), spinach 308 (Spinacia oleracea), tomato (Lycopersicon esculentum), cress (Lepidium sativum) and cucumber 309 (Cucumis sativus). If raw extract inhibits germination completely, RGI and GI lose their value as 310 indexes. For this reason, Komilis and Tziouvaras (2009) had excluded GI data of poultry manure derived 311 compost from correlation analysis. An alternative experimental strategy was proposed by Morel et al. 312 (1985), who determined GI using three aqueous extract concentrations (10, 20 and 40% w/v) (Silva et al., 13 313 2009). However, this methodology cannot be used with any type of sample because the concentrations 314 depend on the toxicity degree. Therefore, we propose to use RGIC0.8 and GIC80% as cut-off values to 315 indicate the lowest concentration that induces inhibitory effects. Also, values lower than or equal to 316 100% indicate any toxicity degree from a sample or immaturity of the compost, whereas values greater 317 than 100% indicate a non-toxicity degree from a sample or maturity of the compost. These new indexes 318 allow the comparison between samples with different toxicity degrees, such as EC50, IC50 or LC50, which 319 are commonly used in ecotoxicology. The use of several types of samples allowed to analyze the 320 robustness of the indexes during the validation process (Table 3). Mature compost (MSW2) and treated 321 effluent showed non-inhibitory effects, whereas immature compost (PMC, PLHMC and MSW1) and 322 untreated effluent showed inhibitory effects. 323 The RGIC0.8 values showed an increase between the initial and final sampling time for lettuce from 0.31 324 to 30.50%, and for radish from 0.06 to 52.74% (Fig. 1-bottom). The GIC80% values showed an increase 325 between the initial and final sampling time for lettuce from 0.42 to 54.34%, and for radish from 0.06 to 326 54.76% (Fig. 1-bottom). These values indicate that composting reduced toxicity. However, the maximum 327 values of RGIC0.8 and GIC80% were lower than 100%. Therefore, the composting piles of this study did 328 not reach full maturity. Further studies could incorporate these phytotoxicity indexes to assess several 329 types of samples, such as effluents, surface water or solid waste extracts. 330 331 3.2.4 Correlations 332 Multivariate analysis can suggest a relationship between toxicity and the inorganic and organic content. 333 An association between physicochemical parameters and ecotoxicological endpoints, including the initial 334 and final sampling times (Fig. 2) was detected after applying Principal Components Analysis (PCA). The 335 results of this analysis account for 67.6% of the variability of the data matrix. PCA showed two clear 336 groups of parameters associated to each sampling time. High values of EC, carbon content (TOC and 337 SC), TN, MC, OM and some cations (mainly Mn, Mg and Ca) were associated with the initial time, 14 338 whereas high values of the ecotoxicological endpoints (low toxicity), pH and TP were associated with 339 the final time. 340 341 Insert Figure 2 342 343 A correlation analysis was carried out between physicochemical and ecotoxicological parameters (Table 344 4). The EC50 for daphnid showed a negative correlation with SC, Mn and Ca (R2: 0.73, 0.72, and 0.69 345 respectively). Also, correlations were observed between phytotoxicity endpoints of lettuce and 346 physicochemical parameters. The highest R2 values were obtained for IC50, NOEC and LOEC of seed 347 germination on lettuce. Pablos et al. (2011) suggested a relationship between electrical conductivity and 348 increasing toxicity. Specifically, authors associated conductivity with the inhibition of root elongation on 349 seeds of lettuce (Young et al., 2012). The IC50 showed a negative correlation with SC, Mn and TN (R2: 350 0.86, 0.79, 0.71 respectively), whereas showed a positive correlation with both SRI and TP (R2: 0.85 and 351 0.70 respectively). The lack of strong correlation between maturity and stability indexes was also 352 observed in Oviedo et al. (2015) as well as in Komilis and Tziouvaras (2009). Komilis and Tziouvaras 353 (2009) reported negative correlations between GI of cress and both TOC and TN. However, we found a 354 negative correlation between GIC80% of lettuce and Mn (R2: 0.69). Both lettuce (R2: 0.79) and radish (R2: 355 0.98) was obtained a positive correlation between RGIC0.8 and GIC80%. The lower correlation between 356 these phytotoxicity indexes for lettuce could be attributable to a higher inhibition of seed germination. 357 358 5. Conclusions 359 1. The proposed monitoring strategy demonstrated the low effectiveness of passive aeration systems 360 to treat poultry manure that is present in high percentages in composting piles (>70%). 361 2. Although the values of SRI, C:N ratio and OM loss indicated compost stability, E. coli colonies 362 remained higher than the limits recommended by international guidelines. 15 363 3. The D. magna endpoints allowed the assessment of possible leachate or run-off toxicity, which 364 showed positive correlations with phytotoxicity endpoints. 365 4. Multivariate analyses demonstrated positive correlations between ecotoxicological endpoints 366 (low toxicity) and biological activity (stability). A PCA demonstrated that these two parameter 367 groups were associated with final sampling time and showed negative correlations with several 368 other physicochemical parameters (organic and inorganic contents). The latter were associated to 369 initial sampling time. 370 5. The RGIC0.8 and GIC80% indexes and salinity indicated that the compost did not reach maturity. 371 As a result, the final compost was not considered suitable for use as a soil amendment. 372 6. The newly proposed phytotoxicity indexes (RGIC0.8 and GIC80%) could be used to assess toxicity 373 from complex samples or could be implemented in monitoring strategies as useful 374 ecotoxicological tools. 375 376 Acknowledgements 377 The study was funded by the Project PNNAT 1128042 from Instituto Nacional de Tecnología 378 Agropecuaria, Argentina. Authors are grateful to Dr. Carlos Greco for comments on the manuscript. 379 Dimitrios Komilis thanks Techniospring for the financial support. 380 381 References 382 Alvarenga, P., Palma, P., Goncalves, A.P., Fernandes, R.M., Cunha-Queda, A.C., Duarte, E., Vallini, G., 383 2007. Evaluation of chemical and ecotoxicological characteristics of biodegradable organic residues 384 for application to agricultural land. Environment international 33, 505-513. 385 Barrington, S., Choinière, D., Trigui, M., Knight, W., 2003. Compost convective airflow under passive 386 aeration. Bioresour Technol 86, 259-266. 16 387 Bolan, N.S., Szogi, A., Chuasavathi, T., Seshadri, B., Rothrock, M., Panneerselvam, P., 2010. Uses and 388 management of poultry litter. World Poultry Sci J 66, 673-698. 389 Bortolotto, T., Bertoldo, J.B., Zanette da Silveira, F., Defaveri, T.M., Silvano, J., Tröger Pich, C., 2009. 390 Evaluation of the toxic and genotoxic potential of landfill leachates using bioassays. Environ Toxicol 391 Pharmacol 28, 288-293. 392 Bustamante, M.A., Paredes, C., Marhuenda-Egea, F.C., Pérez-Espinosa, A., Bernal, M.P., Moral, R., 393 2008. Co-composting of distillery wastes with animal manures: Carbon and nitrogen transformations 394 in the evaluation of compost stability. Chemosphere 72, 551-557. 395 Cabrera, G. L., Rodriguez, D. M. G., Maruri, A. B., 1999. Genotoxicity of the extracts from the compost 396 of the organic and the total municipal garbage using three plant bioassays. Mutat Res 426, 201-206. 397 De Simone, C., Angelucci, R., Errichetti, M. F., Marconi, S., Rossi, M., Selvi, S., 2005. A statistical 398 approach to evaluate compost genotoxicity. Biol fert soils 41, 9-14. 399 Delgado, M., Miralles de Imperial, R., Alonso, F., Rodríguez, C., Martín, J.V., 2013. Ecotoxicity 400 bioassays on leachates from poultry manure. Bull Environ Contam Toxicol 90, 401-404.. 401 Edwards, D.R., Daniel, T.C., 1992. Environmental impacts of on-farm poultry waste disposal — A 402 review. Bioresour Technol 41, 9-33. 403 FAOSTAT, 2013. Statistics of Food and Agricultural commodities production. Food and Agriculture 404 Organization of the United Nations. Available at: (verified August 2015). 405 Frank, R., Stonefield, K.I., Luyken, H., 1988. Monitoring wood shaving litter and animal products for 406 polychlorophenols residues, Ontario, Canada, 1978–1986. Bulletin of environmental contamination 407 and toxicology 40, 468-474. 408 Fuentes, A., Lloréns, M., Sáez, J., Aguilar, M.I., Ortuño, J.F., Meseguer, V.F., 2004. Phytotoxicity and 409 heavy metals speciation of stabilised sewage sludges. Journal of hazardous materials 108, 161-169. 410 Gupta, G., Borowiec, J., Okoh, J., 1997. Toxicity identification of poultry litter aqueous leachate. Poultry 411 science 76, 1364-1367. 17 412 Gupta, G., Kelly, P., 1992. Poultry litter toxicity comparison from various bioassays. J Environ Sci 413 Health, Part A 27, 1083-1093. 414 Iannotti, D.A., Pang, T., Toth, B.L., Elwell, D.I., Keener, H.M., Hoitink, H.A., 1993. A quantitative 415 respirometric method for monitoring compost stability. Compost Science and Utilization, 1, 52-65. 416 Isidori, M., Lavorgna, M., Nardelli, A., Parrella, A., 2003. Toxicity identification evaluation of leachates 417 from municipal solid waste landfills: a multispecies approach. Chemosphere 52, 85–94. 418 Jackson, B.P., Bertsch, P.M., Cabrera, M.L., Camberato, J.J., Seaman, J.C., Wood, C.W., 2003. Trace 419 element speciation in poultry litter. Journal of environmental quality 32, 535-540. 420 Komilis, D.P., Tziouvaras, I.S., 2009. A statistical analysis to assess the maturity and stability of six 421 composts. Waste management (New York, N.Y.) 29, 1504-1513. 422 Kwasniewska, J., Nałęcz-Jawecki, G., Skrzypczak, A., Płaza, G.A., Matejczyk, M., 2012. An assessment 423 of the genotoxic effects of landfill leachates using bacterial and plant tests. Ecotoxicol Environ Saf 424 75, 55-62. 425 Matejczyk, M., Płaza, G.A., Nałęcz-Jawecki, G., Ulfig, K., Markowska-Szczupak, A., 2011. Estimation 426 of the environmental risk posed by landfills using chemical, microbiological and ecotoxicological 427 testing of leachates. Chemosphere 82, 1017-1023. 428 Morel, J., Colin, F., Germon, J., Godin, P., Juste, C., 1985. Methods for the evaluation of the maturity of 429 municipal refuse compost. in: Gasser, J. (Ed.). Composting of agricultural and other wastes. Elsevier 430 Applied Science. 431 Ogunwande, G.A., 2011. Feasibility study of a V-shaped pipe for passive aeration composting. Waste 432 management & research : the journal of the International Solid Wastes and Public Cleansing 433 Association, ISWA 29, 240-248. 434 Ogunwande, G.A., Osunade, J.A., 2011. Passive aeration composting of chicken litter: effects of aeration 435 pipe orientation and perforation size on losses of compost elements. Journal of environmental 436 management 92, 85-91. 18 437 Ogunwande, G.A., Osunade, J.A., Adekalu, K.O., Ogunjimi, L.A., 2008. Nitrogen loss in chicken litter 438 compost as affected by carbon to nitrogen ratio and turning frequency. Bioresour Technol 99, 7495- 439 7503. 440 Oviedo, E., Torres, P., Marmolejo, L., Hoyos, L., Gonzales., S., Barrena, R., Komilis, D., Sánchez, A., 441 2015. Stability and maturity of biowaste composts derived by small municipalities: Correlation 442 among physical, chemical and biological indices. Waste Manag 44, 63-71. 443 Pablos, M.V., Fernández, C., Babín, M.M., Navas, J.M., Carbonell, G., Martini, F., García-Hortigüela, 444 P., Tarazona, J.V., 2009. Use of a novel battery of bioassays for the biological characterisation of 445 hazardous wastes. Ecotoxicol Environ Saf 72, 1594–1600. 446 Pablos, M.V., Martini, F., Fernandez, C., Babin, M., Herraez, I., Miranda, J., Martinez, J., Carbonell, G., 447 San-Segundo, L., García-Hortigüela, P., 2011. Correlation between physicochemical and 448 ecotoxicological approaches to estimate landfill leachates toxicity. Waste Manage 31, 1841-1847. 449 Paredes, C., Bernal, M.P., Cegarra, J., Roig, A., Navarro, A.F., 1996. Nitrogen tranformation during the 450 composting of different organic wastes. in: Van Cleemput, O., Hofman, G., Vermoesen, A. (Eds.). 451 Progress in Nitrogen Cycling Studies. Springer Netherlands, pp. 121-125. 452 Paredes, C., Roig, A., Bernal, M.P., Sánchez-Monedero, M.A., Cegarra, J., 2000. Evolution of organic 453 matter and nitrogen during co-composting of olive mill wastewater with solid organic wastes. Biol 454 Fert Soils 32, 222-227. 455 Parkinson, R., Gibbs, P., Burchett, S., Misselbrook, T., 2004. Effect of turning regime and seasonal 456 weather conditions on nitrogen and phosphorus losses during aerobic composting of cattle manure. 457 Bioresour Technol 91, 171-178. 458 Petric, I., Šestan, A., Šestan, I., 2009. Influence of initial moisture content on the composting of poultry 459 manure with wheat straw. Biosyst Eng 104, 125-134. 19 460 Riera, N., Torre, V.D., Rizzo, P.F., Butti, M., Bressan, F., Zarate, N., Weigandt, C.,Crespo, D., 2014. 461 Composting process evaluation of two mixtures of poultry manures. Rev FCA UNCUYO 46, 195- 462 203. 463 Rizzo, P., Torre, V., Riera, N., Crespo, D., Barrena, R., Sánchez, A., 2013. Co-composting of poultry 464 manure with other agricultural wastes: process performance and compost horticultural use. J Mater 465 Cycles Waste Manage, 1-9. 466 SENASA, 2011. Manual of Procedures for registering products in the National Register of Fertilizers, 467 Dressings, Substrates, Protectors, Conditioners and Raw Materials (In Spanish). Resolution Nº 468 264/11. Argentina. Available at (verified August 2015). 469 Silva, M.E., Lemos, L.T., Cunha-Queda, A.C., Nunes, O.C., 2009. Co-composting of poultry manure 470 with low quantities of carbon-rich materials. Waste Manage Res 27, 119-128. 471 Sobrero, C., Ronco, A.E., 2004. Ensayo de toxicidad aguda con semillas de L. sativa. in: Castillo 472 Morales, G. (Ed.). Ensayos toxicológicos y métodos de evaluación de calidad de aguas: 473 Estandarización, intercalibración, resultados y aplicaciones, Mexico, pp. 71-79. 474 Solano, M.L., Iriarte, F., Ciria, P., Negro, M.J., 2001. SE—Structure and Environment: Performance 475 Characteristics of Three Aeration Systems in the Composting of Sheep Manure and Straw. J Agr Eng 476 Res 79, 317-329. 477 Tiquia, S.M., 2010. Reduction of compost phytotoxicity during the process of decomposition. 478 Chemosphere 79, 506-512. 479 Tiquia, S.M., Tam, N.F.Y., 2000a. Co-composting of spent pig litter and sludge with forced-aeration. 480 Bioresour Technol 72, 1-7. 481 Tiquia, S.M., Tam, N.F.Y., 2000b. Fate of nitrogen during composting of chicken litter. Environmental 482 pollution (Barking, Essex : 1987) 110, 535-541. 483 Tiquia, S.M., Tam, N.F.Y., 2002. Characterization and composting of poultry litter in forced-aeration 484 piles. Process Biochem 37, 869-880. 20 485 Tiquia, S.M., Tam, N.F.Y., Hodgkiss, I.J., 1996. Effects of composting on phytotoxicity of spent pig- 486 manure sawdust litter. Environmental pollution (Barking, Essex : 1987) 93, 249-256. 487 Tuomela, M., Vikman, M., Hatakka, A., Itävaara, M., 2000. Biodegradation of lignin in a compost 488 environment: a review. Bioresour Technol 72, 169-183. 489 USDA, USCC, 2001. Test Methods for the Examination of Composting and Compost. Houston: Edaphos 490 International, Department of Agriculture and Composting Council, USA. 491 USEPA, 1996. Ecological Effects Test Guidelines OPPTS 850.1010 Aquatic Invertebrate Acute Toxicity 492 Test, Freshwater Daphnids, EPA 712-C-96-114 Environmental Protection Agency, USA. 493 WRAP, 2011. Guidelines for the specification of quality compost for use in growing media., Waste & 494 Resources Action Programme. United Kingdom. Available at: (verified August 495 2015). 496 Young, B.J., Riera, N.I., Beily, M.E., Bres, P.A., Crespo, D.C., Ronco, A.E., 2012. Toxicity of the 497 effluent from an anaerobic bioreactor treating cereal residues on Lactuca sativa. Ecotoxicol Environ 498 Saf 76, 182-186. 499 Zucconi, F., Pera, A., Forte, M., De Bertoldi, M., 1981. Evaluating toxicity of immature compost. 500 Biocycle 22, 54-57. 501 502 21 503 Table 1. Mean (± SD) physicochemical and microbiological parameters of the six composting piles at each sampling time and limit values of 504 final composts. Target value or Parameter 0-d 14-d 56-d 92-d Reference range / upper limit pH 6.8 ± 0.3 a 7.7 ± 0.3 b 7.6 ± 0.4 b 8.2 ± 0.4 b 6-8/9 WRAP (2011) EC (mS cm-1) 17.6 ± 1.9 a 13.3 ± 5.5 ab 10.8 ± 2.9 b 8.0 ± 1.8 b < 0.6 / 1.5 WRAP (2011) C:N ratio 24.6 ± 3.6 a 31.6 ± 3.1 a 23.4 ± 3.6 a 24.8 ± 3.7 a 20:1 SENASA (2011) OM (%) 70.1 ± 1.3 a 62.1 ± 3.0 b 51.1 ± 6.8 c 45.7 ± 2.7 c ≥ 15 SENASA (2011) MC (%) 70.6 ± 3.2 a 67.8 ± 3.8 a 64.6 ± 4.9 ab 60.4 ± 4.2 b 35 - 40 / 50 WRAP (2011) TOC (%) 35.0 ± 0.7 a 31.0 ± 1.5 b 25.5 ± 3.4 c 22.9 ± 1.3 c - - TN (%) 1.4 ± 0.2 a 1.0 ± 0.1 b 1.1 ± 0.2 b 0.9 ± 0.1 b NPK ≥ 6 SENASA (2011) TP (mg g-1) 20.3 ± 3.1 a n.d. n.d. 24.8 ± 3.0 a NPK ≥ 6% SENASA (2011) SC (%) 2.5 ± 0.3 a n.d. n.d. 1.2 ± 0.3 b - - SP (mg g-1) 0.5 ± 0.1 a n.d. n.d. 0.5 ± 0.1 a NPK ≥ 6% SENASA (2011) Ca (mg L-1) 79.8 ± 15.4 a n.d. n.d. 31.5 ± 40.8 b > 1% SENASA (2011) Mg (mg L-1) 125.2 ± 30.3 a n.d. n.d. 70.9 ± 47.5 b > 0.05 % SENASA (2011) K (mg L-1) 1636.6 ± 165.9 a n.d. n.d. 1695.3 ± 861.3 a NPK ≥ 6% SENASA (2011) Na (mg L-1) 472.9 ± 39.9 a n.d. n.d. 468.9 ± 139.7 a < 100 / 150 WRAP (2011) Zn (mg L-1) 0.9 ± 0.5 a n.d. n.d. 0.6 ± 0.6 a < 150 / 400 WRAP (2011) Mn (mg L-1) 2.0 ± 0.5 a n.d. n.d. 1.1 ± 0.3 b - - Cu (mg L-1) 1.5 ± 1.3 a n.d. n.d. 0.9 ± 0.7 a < 50 / 100 WRAP (2011) Total coliforms 7.5x106 ± 8.0x106 ± 1.0x107a n.d. n.d. - - (CFU) 9.3x106a 7.5x106 ± Absent / 1000 WRAP (2011) E. coli (CFU) 1.9x107 ± 1.1x107a n.d. n.d. 9.1x106a Salmonella spp. Absent n.d. n.d. Absent Absent / Zero WRAP (2011) 505 Different letters indicate significant differences (p<0.05) among sampling times. 506 EC = Electrical conductivity; OM = Organic matter; MC = Moisture content; TOC = Total organic carbon; TN = Total Kjeldahl nitrogen; SC = 507 Soluble carbon; TP = Total phosphorous; SP = Soluble phosphorous; n.d. = no data 22 508 Table 2. Mean (95% CI) ecotoxicological endpoints of the test organisms at each sampling time. Endpoint 0-d 14-d 56-d 92-d Lettuce Seed germination IC50 (%) 23.5 [15.6-31.5] a 69.5 [61.5-77.4] **b 60.0 [49.2-70.9] **bc 46.8 [44.4-49.3] c LOEC (%) 28.3 [17.7-39.0] a 70.0 [54.0-86.0] *b 64.0 [50.6-77.4] *b 50.0 [41.2-58.8] ab NOEC (%) 14.2 [8.8-19.5] a 50.0 [34.0-66.0] *b 44.0 [30.6-57.4] *b 30.0 [21.2-38.8] ab Root elongation IC50 (%) 8.8 [3.6-14.0] a 45.4 [26.5-64.4] b 55.0 [39.8-70.2] b 32.9 [24.7-41.1] ab LOEC (%) 7.8 [1.9-13.6] a 25.2 [10.7-41.3] a 38.3 [16.6-60.1] a 28.3 [17.7-39.0] a NOEC (%) 3.6 [0.5-6.7] a 12.3 [3.9-20.7] a 24.2 [6.9-41.4] a 14.2 [8.8-19.5] a Radish Seed germination IC50 (%) 63.3 [47.7-78.8]** n.t. n.t. n.t. LOEC (%) 88.0 [73.7-100.0]* n.t. n.t. n.t. NOEC (%) 68.0 [53.7-82.3]* n.t. n.t. n.t. Root elongation IC50 (%) 9.7 [4.4-15.0] a 51.8 [34.4-69.2] b 54.1 [39.5-68.7] b 41.6 [20.6-62.7] ab LOEC (%) 12.0 [0.0-24.3] a 49.0 [21.2-76.8] a 54.2 [33.9-74.5] a 41.0 [11.6-70.4] a NOEC (%) 5.5 [0.0-11.9] a 32.2 [8.1-56.3] a 36.8 [21.4-52.3] a 26.9 [3.0-50.8] a Daphnid EC50 (%) 6.7 [4.3-9.0] a 26.5 [16.6-36.4] ab 29.3 [20.1-38.6] b 28.5 [14.8-42.2] b LOEC (%) 9.0 [5.0-13.0] a 27.6 [17.0-38.2] a 33.3 [16.4-50.2] a 28.2 [12.4-43.9] a NOEC (%) 4.3 [1.8-6.8] a 16.8 [9.8-23.8] a 21.0 [9.0-33.0] a 17.3 [6.4-28.3] a 509 Different letters indicate significant differences (p<0.05) among sampling times. 510 * A total of 83% of the samples exhibiting a toxic response. 511 ** A total of 67% of the samples exhibiting a toxic response. 512 n.t.: no toxicity response. 513 23 514 Table 3. Values of the new phytotoxicity indexes on several types of samples to validate the proposed 515 methodology. Type of sample Waste origin Treatment Seed RGIC0.8 GIC80% Reference Immature compost Active Lettuce 19.67 16.11 Poultry manure composting Rizzo et al. (2013) (PMC) Radish 34.40 21.77 Immature compost Active Lettuce 35.99 30.88 Poultry manure Riera et al. (2014) (PLHMC) composting Radish 24.63 20.32 Immature compost Organic Active Lettuce 40.61 45.56 Unpublished data (MSW1) fraction of MSW composting Radish 87.88 73.45 Mature compost Organic Active composting Lettuce 108.76 104.19 Unpublished data (MSW2) fraction of MSW Untreated effluent Anaerobic 18.02 24.77 Cereal residues biodigestion Lettuce Young et al. (2012) Treated effluent 111.35 121.57 516 PMC: poultry manure derived compost. PLHMC: poultry litter and horse manure derived compost. 517 MSW1 and MSW2: municipal solid waste derived compost. 24 518 Table 4. Pearson correlation coefficients among various parameters measured at four sampling times at six composting piles (n = 24). Physico-chemical parameters Stability D. magna L. sativa (lettuce) R. sativus (radish) OM or NOEC or NOEC or NOEC or NOEC or Ash SC Ca Mg Na Mn Cu SRI IC50 s.g. GRIC0.8 GIC80% GRIC0.8 GIC80% TOC LOEC LOEC r.e. LOEC s.g. LOEC r.e. pH n.s. n.s. -0.75* n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. EC -0.80** 0.77** 0.83** n.s. n.s. n.s. 0.79* n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. Ash -0.78** n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. OM or TOC 0.92** n.s. n.s. n.s. 0.78* n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. Physico- TN 0.87** n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. chemical parameters SC 0.77* n.s. n.s. 0.78* n.s. -0.92** n.s. n.s. -0.86** n.s. n.s. n.s. n.s. n.s. n.s. Ca 0.94** n.s. 0.75* n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. K 0.97** n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. Zn n.s. 0.84** n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. Mn n.s. -0.76* n.s. n.s. -0.79* -0.91** n.s. n.s. n.s. n.s. n.s. Stability SRI n.s. n.s. 0.85** n.s. n.s. n.s. n.s. n.s. n.s. D. magna EC50 0.90** n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. IC50 r.e. 0.85** n.s. 0.78** 0.90** 0.82** n.s. n.s. n.s. NOEC or L. sativa LOEC r.e. n.s. n.s. 0.84** 0.91** n.s. n.s. n.s. (lettuce) IC50 s.g. 0.83** n.s. n.s. n.s. n.s. n.s. GRIC0.8 0.79** n.s. n.s. n.s. IC50 r.e. 0.89** 0.81** 0.81** R. sativus NOEC or (radish) LOEC r.e. 0.87** 0.87** GRIC0.8 0.98** 519 Significant parameters are only shown (* p<0.01 and ** p<0.001). 520 n.s.: not significant; r.e.: root elongation; s.g.: seed germination; OM and TOC are shown together because they have the same correlation 521 coefficient values; NOEC and LOEC are shown together because they have the same correlation coefficient value. 25 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 Figure 1. 548 26 549 550 551 552 553 554 555 556 Figure 2. 557 558 559 560 561 562 27 563 Figure captions 564 565 566 Figure 1. (Top): Average temperature of ambient air and of the inside of the composting piles (average 567 based on n = 6) error bars demonstrate standard deviations), (Middle): Average cumulative losses 568 (±standard error) of the OM and TN (%) and average SRI (mg O2 g-1 OM h-1) during the composting 569 period; (Bottom) Average values (±standard error) of the phytotoxicity indexes measured during the 570 composting period. 571 572 573 Figure 2. Principal components analysis (PCA) shows the association between physicochemical 574 parameters and ecotoxicological endpoints with respect to the first two components. Data of IC50 of 575 lettuce and radish are only shown for root elongation. Black dots indicate each composting pile and 576 sampling time. 577 578 579 28