Introduction
Composting the organic fraction from municipal solid waste (MSW) is a sustainable way to manage and recycle urban waste and is a potentially viable source of high quality fertilizer. Indeed, several national and international directives encourage the maximal exploitation of MSW, with the aim of reducing the environmental impact and the loss of organic resources derived from urban refuse dumping (The Ninety-fourth United States Congress, 1976; Council of the European Union, 1999). Composted MSW is obtained by biological degradation of the organic substrates under aerobic conditions. The safe use of composted MSW in agriculture can be guaranteed if the organic matter source is properly sorted at the origin. Thus, good source management restricts the levels of heavy metals and other contaminants, producing a compost that meets quality standards for organic agriculture (Hargreaves et al., Reference Hargreaves, Adl and Warman2008).
Agricultural use of compost from the MSW organic fraction provides an interesting alternative to mineral fertilizers from a double standpoint: it is compatible with the increasing demand of agricultural commodities produced under sustainable practices, and it may lead to very positive changes in the soil. Concerning the latter, some reports have described how the continued use of compost can lead to favorable soil changes including increased moisture, organic matter and bulk density and the appearance of beneficial microorganisms (Elherradi et al., Reference Elherradi, Soudi, Chiang and Elkacemi2005; Martínez-Blanco et al., Reference Martínez-Blanco, Muñoz, Antón and Rieradevall2009), which ultimately result in better fruit quality and production rates. However, other reports have demonstrated that under certain conditions, compost fertilization induces lower production yields in comparison with mineral fertilization (Martínez-Blanco et al., Reference Martínez-Blanco, Muñoz, Antón and Rieradevall2011; Seufert et al., Reference Seufert, Ramankutty and Foley2012). The detrimental effects may be due to the fact that nutrients in mineral fertilizer are very soluble and easily available to plants, while compost nutrients are in more complex forms and therefore less available.
Thus, the use of MSW compost in tomato production (Solanum lycopersicum L.) may result in changes not only in the production yield, but also in fruit growth and fruit quality. At the same time, the tomato phenolic profile may be affected (Treutter, Reference Treutter2010). Tomato phenolics deserve special attention for human health given that these compounds with antioxidant properties have been implicated in metabolic processes that may be important in reducing cancer and cardiovascular disease risk (Nijveldt et al., Reference Nijveldt, van Nood, van Hoorn, Boelens, van Norren and van Leeuwen2001; Crozier et al., Reference Crozier, Jaganath and Clifford2009). For example, the high life expectancy of some Mediterranean countries has been linked to diets rich in fruits and vegetables, of which the tomato is a very common ingredient (Zamora-Ros et al., Reference Zamora-Ros, Serafini, Estruch, Lamuela-Raventós, Martínez-González, Salas-Salvadó, Fiol, Lapetra, Arós, Covas and Andres-Lacueva2013). It has been demonstrated that tomato plants accumulate phenolic compounds as a defense mechanism under certain stress conditions, such as low N availability (Løvdal et al., Reference Løvdal, Olsen, Slimestad, Verheul and Lillo2010). An increase in tomato phenolic content has also been observed after organic amendment, such as compost fertilization (Mitchell et al., Reference Mitchell, Hong, Koh, Barrett, Bryant, Denison and Kaffka2007; Brandt et al., Reference Brandt, Leifert, Sanderson and Seal2011). Carotenoids, on the contrary, seem not to be affected by limited nutrient conditions (Brandt and Mølgaard, Reference Brandt and Mølgaard2001; Caris-Veyrat et al., Reference Caris-Veyrat, Amiot, Tyssandier, Grasselly, Buret, Mikolajczak, Guilland, Bouteloup-Demange and Borel2004).
This work studied the effect of fertilizing with compost from the MSW organic fraction on tomato yield, fruit quality and phenolic content, as part of an integrated strategy for the reutilization of MSW. The composting plant and the experimental field were located near Barcelona (Spain). The tested plants were of the ‘Penjar’ varietal type, which yields small-sized tomatoes with long shelf life that are very popular and widely consumed along the Spanish Mediterranean coast.
Materials and Methods
Field design and growth conditions
The experimental tomato plants were obtained from commercial seeds of the variety Palamós, ‘Penjar’ varietal type (Semillas Fitó, Barcelona, Spain). Plants were cultivated in an open-field system located in the Maresme region, Catalonia (northeast Spain, 41°38′27″N 2°43′00″E, 10 m elevation). This region that borders the Mediterranean Sea is prolific in terms of horticultural production. The climate is typically Mediterranean, with an average temperature of 18.7°C and total rainfall equivalent to 165 mm3 between March and August (Servei Meteorològic de Catalunya, 2015).
Plants were spaced 0.7–2.0 m apart, giving a density of 7142 plants·ha−1. The soil of the experimental garden was a sandy loam. There was minimal usage of pesticides and herbicides, following the procedures accepted by the environmentally friendly regulations of the Integrated Cultivation Management (The Spanish Ministry of Agriculture, Fisheries and Food, 2002). Growing conditions were in accord with ‘low input agricultural management’ (Viaux, Reference Viaux, Biala, Terres, Pointereau and Paracchini2008). Following the regional standard conditions for ‘Penjar’ tomato cultivation, a plastic mulch layer was installed, plants were not staked and water was extracted from local wells and supplied via drip irrigation. Every treatment plot received 190 mm m−2 of water, in accordance with FAO guidelines (Allen et al., Reference Allen, Pereira, Raes and Smith1998) for cultures under plastic mulch and with specific water needs calculated on the basis of data obtained from tensiometers buried 30–60 cm underground.
Fertilization
Three different fertilization strategies were tested as follows: (i) M, mineral fertilizer only, (ii) C, MSW compost only and (iii) C + M, a combination of MSW compost and mineral fertilizer. The experimental design consisted of nine random blocks with three treatments and three repetitions each. Potassium nitrate (KNO3) was used as a mineral fertilizer and delivered with the water supply. Compost for the C and C + M treatments was obtained from an industrial composting plant (Metrocompost, Castelldefels, Spain), which processed the MSW from an area near Barcelona with approximately 30,000 inhabitants. The raw compost material, which was mainly food waste, was sorted from the organic fraction of MSW. Pruning waste from parks and gardens was also added in a 1:1 v/v ratio in order to acting as a bulking agent (Martínez-Blanco et al., Reference Martínez-Blanco, Colón, Gabarrell, Font, Sánchez, Artola and Rieradevall2010), generating porosity and promoting liquid retention. This facility used the most typical technology for indoor plant composting in Spain, including in-vessel decomposition and a curing phase in turned windrows (Martínez-Blanco et al., Reference Martínez-Blanco, Muñoz, Antón and Rieradevall2011). The quality of the compost, including its microbiological and chemical safety (Table 1) met required standards (Quirós et al., Reference Quirós, Villalba, Gabarrell and Muñoz2015).
Chemical and microbiological information on the compost used in the previous and current seasons of the study (no replicates of analysis, only for estimation purposes).
The fertilizer doses in each treatment were calculated considering all possible sources of N, including the mineral nitrogen (N) initially contained in the soil (Table 2) and the N addition during the study (Table 3). The sampling for determining the initial soil moisture and N content in each plot, prior to the study, was done at planting time at three different points and three soil depth intervals (0–20, 20–40 and 40–60 cm). Soil moisture content was determined gravimetrically, after drying triplicate samples at 105°C for 48 h. The soil extractable mineral N (NH4 +-N, NO3 −-N, NO2 −-N), prior to the study, was determined following the method ISO 14256-2 (International Organization for Standardization, 2005), by extraction with potassium chloride solution, and compost N was estimated following the Kjeldahl method (Bremner, Reference Bremner1960). The soil bulk density was determined in order to express data as kg ha−1. For the first layer (0–20 cm), the bulk density was measured from undisturbed soil core samples. For the deeper layers, the bulk density was calculated from a pedotransfer function (Saxton et al., Reference Saxton, Rawls, Romberger and Papendick1986) (Table 2).
Soil moisture and extractable mineral N prior to the study (mean values ± standard deviation; n = 3).
1 Average (NH4 +-N, NO3 −-N, NO2 −-N) of three samples and three N measurements per sample.
Estimated N inputs 1 to the fertilization treatments (kg N·ha−1).
1 Theoretical values, partially calculated from the fertilizer formulation and dose (mineral N applied), and the predicted rate of mineralization from the compost of the previous season (compost N applied). No replicates can arise from such estimations; therefore, these data must only be interpreted for estimation purposes.
2 Includes N from irrigation water (50 kg ha−1 in all treatments) and mineral fertilizer (KNO3).
3 N input from compost applied in the current and previous seasons.
4 The estimated total N is the addition of the total soil N content at 0–60 cm prior to the study (Table 2) and the mineral and compost N inputs.
Concerning the mineral N inputs during the study (Table 3), NO3 −N concentrations in the irrigation water were determined by the copperized Cd reduction colorimetric method described by Keeney and Nelson (Reference Keeney, Nelson, Page, Miller and Keeney1982). All treatments were provided with 50 kg ha−1 of mineral N, including C and C + M, due to the presence of mineral N in the irrigation water obtained from local wells. The amount of mineral N from fertilizer was calculated from its formulation. The estimated N inputs are given in Table 3. No replicates for this calculation could be performed.
The use of compost during previous growing seasons in the C and C + M parcels was an additional source of available N, due to the slow mineralization of the compost N forms (Amlinger et al., Reference Amlinger, Götz, Dreher, Geszti and Weissteiner2003; Hargreaves et al., Reference Hargreaves, Adl and Warman2008). According to these authors, the additional contribution in the C and C + M treatments could be estimated as 12% of the N not mineralized in the previous season of application. Hence, the total N contribution was estimated to be 114 and 56 kg N·ha−1 in the C and C + M treatments, respectively. No replicates for this estimation could be performed (Table 3). Organic fertilizers can be supplied in accordance with the legal limit for N application, which is 170 kg N ha−1 in the UE (Council of the European Union, 1991).
Production and harvest quality
Ripe tomatoes (red stage) were harvested at three different times, which were determined according to when there were a significant number of ripe tomatoes. For production measurements, only central plants were considered, discarding the plants situated at the edges of the blocks. Crop yield was assessed by the total weight of the harvested fruit (total crop yield) and the weight of the marketable fruit (commercial crop yield). The individual weight, diameter and °Brix (Digital refractometer PR-32α, Atago, Tokyo, Japan) of 10 fruits per repetition were measured as quality parameters.
Extraction of phenolic compounds
One fruit per block was collected for the analysis of phenolic compounds, giving nine samples per treatment. The extraction procedure was performed according to the method described by Ribas-Agustí et al. (Reference Ribas-Agustí, Cáceres, Gratacós-Cubarsí, Sárraga and Castellari2012).
Total polyphenol content
Total polyphenols were assessed by spectrophotometry according to the method published by Singleton and Rossi (Reference Singleton and Rossi1965) with minor modifications. The methanolic extract (1 mL) was mixed with 3 mL ultrapure water (Millipore, Madrid, Spain) and 0.25 mL of 2 m Folin-Ciocalteu's reagent (Sigma-Aldrich, Madrid, Spain). After 1 min incubation, the reaction was stopped by adding 2.5 mL of 2.08 M Na2CO3 solution and 3.25 mL ultrapure water. After 2 h at room temperature, the absorbance of the sample was measured at 725 nm (UV-240 Graphicord, Shimadzu, Kyoto, Japan). Total polyphenols, expressed as caffeic acid equivalents, were calculated according to an external calibration curve of caffeic acid standard (Sigma-Aldrich, Madrid, Spain).
Analysis of phenolic compounds
Phenolic compounds were analyzed by UHPLC-DAD-MS/MS according to the method of Ribas-Agustí et al. (Reference Ribas-Agustí, Cáceres, Gratacós-Cubarsí, Sárraga and Castellari2012). Phenolic compounds were identified on the basis of their chromatographic retention time, UV-visible spectra and MS/MS spectra, matching with the commercial standards (5-caffeoylquinic acid and quercetin-3-O-rutinoside from Sigma-Aldrich, Madrid, Spain; kaempferol-3-O-rutinoside from Extrasynthèse, Genay, France). Other phenolic compounds were tentatively identified by comparing chromatographic retention time, UV-Visible spectra and MS/MS spectra with literature information (Gómez-Romero et al., Reference Gómez-Romero, Segura-Carretero and Fernández-Gutiérrez2010; Vallverdú-Queralt et al., Reference Vallverdú-Queralt, Jáuregui, Di Lecce, Andrés-Lacueva and Lamuela-Raventós2011; Ribas-Agustí et al., Reference Ribas-Agustí, Cáceres, Gratacós-Cubarsí, Sárraga and Castellari2012). Quantification was performed using diode array detection at λ = 323 nm for hydroxycinnamic acid derivatives and λ = 350 nm for flavonoids (quercetin and kaempferol derivatives). Compounds with no commercial standard available were quantified as equivalents of the standards with the most similar UV-VIS absorption properties (5-caffeoylquinic acid, quercetin-3-O-rutinoside, kaempferol-3-O-rutinoside).
Statistics
Tomato quality and crop yield, as well as phenolic compounds and total polyphenol contents were analyzed by one-way analysis of variance (ANOVA). Post-hoc differences between groups were further determined using the Tukey (HSD) test. In all instances, a significance level (α) was set at 0.05. SPSS software was used for all analyses (IBM, Armonk, NY).
Results and Discussion
Effect of MSW compost fertilization on tomato yield
The use of compost alone (C) had a detrimental effect on crop yield as compared with the use of mineral fertilizer (M) or the combination of compost and mineral fertilizers (C + M) (Table 4). The C treatment had 43% lower total and commercial crop yields than the M treatment. Similarly, C had 45 and 48% lower total and commercial crop yields, respectively, than C + M treatment. However, no significant differences in crop yield parameters were observed between the C + M and M treatments (Table 4).
Quality parameters and crop yield measured in tomato (varietal type ‘Penjar’) following compost (C), compost and mineral (C + M) and mineral (M) fertilization treatments.
Different small letters: significant difference between treatments; n.s., no significant difference (P<0.05).
Previous studies testing compost from different origins with and without mineral fertilization have reported contradictory results, with some studies demonstrating similar production rates (Clark et al., Reference Clark, Horwath, Shennan, Scow, Lantni and Ferris1999; Martínez-Blanco et al., Reference Martínez-Blanco, Muñoz, Antón and Rieradevall2009; Martínez-Blanco et al., Reference Martínez-Blanco, Muñoz, Antón and Rieradevall2011), and others demonstrating significant decreases under compost conditions (Ghorbani et al., Reference Ghorbani, Koocheki, Jahan and Asadi2008; Riahi et al., Reference Riahi, Hdider, Sanaa, Tarchoun, Ben Kheder and Guezal2009). It is generally assumed that N assimilation under mineral fertilization is direct and rapid, allowing regular production yields, while the dynamics of N mineralization from compost or other organic sources are more irregular and unpredictable (Clark et al., Reference Clark, Horwath, Shennan, Scow, Lantni and Ferris1999). Due to the high complexity of factors affecting quality and crop yield, care is called for when comparing studies conducted under different agronomic conditions, and the results must be interpreted within the context of the study. In the case of the present study, we need to consider that the fertilizer doses in all three treatments followed the regular recommendations for intensive tomato cultivation and for compost fertilization in this region (Martínez-Blanco et al., Reference Martínez-Blanco, Lazcano, Christensen, Muñoz, Rieradevall, Möller, Antón and Boldrin2013), which probably contributed to the differences estimated among treatments for the input of N (Table 3). Such differences may explain some of the production differences obtained between the C treatment and the other test groups. On the other hand, fertilizer responses were evaluated for a coarse-textured soil with low organic matter content, which would have seriously limited soil N supplying power and thereby enhanced the need for fertilizer N. The yield limitation observed with the C treatment may not have occurred with a soil better able to supply N through mineralization. In addition, it has been demonstrated that an increased yield can be expected in crops after 4–5 years of organic management due to physical, chemical and biological soil improvement (Creamer et al., Reference Creamer, Bennett, Stinner and Cardina1996; Colla et al., Reference Colla, Mitchell, Poudel and Temple2002). The historical compost use of the C treatment parcel originated three years before the current study. Hereto, higher production yields might be expected in the future if compost fertilization is an ongoing practice.
Effect of MSW compost fertilization on tomato fruit quality
The use of compost, alone (C) or combined with mineral fertilizer (C + M), did not affect quality parameters (fruit weight, diameter and °Brix) of ‘Penjar’ tomato in comparison with what was obtained with mineral fertilization (M) (Table 4). These results are in accordance with those of Riahi et al. (Reference Riahi, Hdider, Sanaa, Tarchoun, Ben Kheder and Guezal2009) and Martínez-Blanco et al. (Reference Martínez-Blanco, Muñoz, Antón and Rieradevall2011), who found no quality differences in other tomato varieties grown under compost and conventional systems.
Phenolic content in ‘Penjar’ tomato as affected by MSW compost fertilization
A chromatogram of the main phenolic compounds of ‘Penjar’ tomato is shown in Figure 1. Quercetin-3-O-rutinoside (rutin) and 5-caffeoylquinic acid (chlorogenic acid) were identified on the basis of their matching with commercial standards. Eight other major phenolic compounds were tentatively identified in ‘Penjar’ tomato (as described in the section ‘Methods’) according to data from Gómez-Romero et al. (Reference Gómez-Romero, Segura-Carretero and Fernández-Gutiérrez2010), Vallverdú-Queralt et al. (Reference Vallverdú-Queralt, Jáuregui, Di Lecce, Andrés-Lacueva and Lamuela-Raventós2011) and Ribas-Agustí et al. (Reference Ribas-Agustí, Cáceres, Gratacós-Cubarsí, Sárraga and Castellari2012). From these, five hydroxycinnamic acid derivatives were quantified as 5-caffeoylquinic acid equivalents: two isomeric forms of caffeoyl-hexose (I and II), coumaroyl-hexose, 4-caffeoylquinic acid and tricaffeoylquinic acid. In addition, two quercetin derivatives were quantified as equivalents of quercetin-3-O-rutinoside: quercetin-rutinoside-hexoside and quercetin-rutinoside-pentoside. A kaempferol derivative was quantified as equivalents of kaempferol-3-O-rutinoside: kaempferol-rutinoside-pentoside. Compounds within the group of caffeoylquinic acid derivatives constituted the majority of the phenolic content in ‘Penjar’ tomatoes, although significant levels of flavonoids were also present (Fig. 2).
HPLC-DAD chromatogram showing the identified phenolic compounds in Solanum lycopersicum L., varietal type ‘Penjar’, λ = 350 nm. Identification of peaks: 1, coumaroyl-hexose; 2, caffeoyl-hexose I; 3, caffeoyl-hexose II; 4, 4-caffeoylquinic acid; 5, 5-caffeoylquinic acid; 6, quercetin-rutinoside-hexoside; 7, quercetin-rutinoside-pentoside; 8, quercetin-3-O-rutinoside; 9, kaempferol-rutinoside-pentoside; 10, tricaffeoylquinic acid.
Total hydroxycinnamoyl derivatives and flavonoids (as measured by HPLC), total polyphenols (as measured by Folin Ciocalteu's method) and effect of fertilization treatment in tomato var. ‘Penjar’. Bars represent the average of n = 9 and error bars show standard error of the mean. 1 mg·kg−1 caffeic acid equivalents, fresh weight.
It has been reported that ‘Penjar’ tomatoes accumulate high levels of total polyphenols (Cortés-Olmos et al., Reference Cortés-Olmos, Leiva-Brondo, Roselló, Raigón and Cebolla-Cornejo2014; Figàs et al., Reference Figàs, Prohens, Raigón, Fita, García-Martínez, Casanova, Borràs, Plazas, Andújar and Soler2015). However, this work profiles for the first time the individual phenolic compounds of this varietal type. The phenolic profile in tomato fruit depends strongly on the variety (Slimestad and Verheul, Reference Slimestad and Verheul2009; Ribas-Agustí et al., Reference Ribas-Agustí, Cáceres, Gratacós-Cubarsí, Sárraga and Castellari2012). In this sense, a remarkable trait of the ‘Penjar’ tomato investigated in this study was the absence of the phenolic intermediate naringenin chalcone, which is on the contrary accumulated in significant amounts in other tomato varieties (Ribas-Agustí et al., Reference Ribas-Agustí, Cáceres, Gratacós-Cubarsí, Sárraga and Castellari2012, Reference Ribas-Agustí, Bouchagier, Skotti, Erba, Casiraghi, Sárraga and Castellari2013; Erba et al., Reference Erba, Casiraghi, Ribas-Agustí, Cáceres, Marfà and Castellari2013).
MSW compost fertilization resulted in subtle differences in the phenolic content of ‘Penjar’ tomato, namely in the case of the flavonoid kaempferol-rutinoside-pentoside, whose content was significantly higher in compost-fertilized ‘Penjar’ tomatoes (Table 5). The rest of the hydroxycinnamic acid derivatives and flavonoids did not change significantly, regardless of the fertilization treatment. No significant differences were observed in the sum of hydroxycinnamic acids, flavonoids or in total polyphenols between the three fertilization strategies (Fig. 2).
Phenolic content (mg·kg−1 fresh weight±standard error of the mean) in tomato (varietal type ‘Penjar’) after different fertilization treatments: compost (C), compost and mineral (C + M) and mineral (M).
1 Quantified as 5-caffeoylquinic equivalents.
2 Quantified as quercetin-3-O-rutinoside equivalents.
3 Quantified as kaempferol-3-O-rutinoside equivalents.
The effect of the treatment is expressed with the one-way ANOVA P-value. Different letters within a row indicate significant differences after post-hoc testing (α = 0.05).
As in the evaluation of the crop yield, special attention needs to be paid when comparing different studies, due to the potential confounding effect of uncontrolled factors. It has been suggested that lower N availability, which would be expected with compost fertilization, is likely to produce such a plant stress as to induce phenolic compound synthesis in order to better protect against disease (Brandt and Mølgaard, Reference Brandt and Mølgaard2001). For example, the significant increase in the accumulation of kaempferol-rutinoside-pentoside in ‘Penjar’ tomato under MSW compost fertilization is likely due to reduced N availability. However, if the whole phenolic content is taken into account (Table 5), the fertilization regime had no effect on the total phenolic content under the assayed conditions. The results presented herein are consistent with those of other authors who also tested compost fertilization for affecting tomato phenolic content (Riahi et al., Reference Riahi, Hdider, Sanaa, Tarchoun, Ben Kheder and Guezal2009). Some groups have reported clearer positive effects of organic management, including compost fertilization, on tomato phenolic content (Caris-Veyrat et al., Reference Caris-Veyrat, Amiot, Tyssandier, Grasselly, Buret, Mikolajczak, Guilland, Bouteloup-Demange and Borel2004; Mitchell et al., Reference Mitchell, Hong, Koh, Barrett, Bryant, Denison and Kaffka2007). In these cases, we need to consider that agronomic practices other than fertilization may induce additional plant stress and have an additive effect on tomato phenolic content (Rosales et al., Reference Rosales, Cervilla, Sanchez-Rodriguez, Rubio-Wilhelmi, Blasco, Rios, Soriano, Castilla, Romero and Ruiz2011).
Conclusions
In major urban areas such as Barcelona, which borders an important agricultural zone, the generation and use of composted MSW as fertilizer provides a particularly attractive opportunity. However, the use of composted MSW needs to be supported with studies regarding the possible effect on crop yield and quality. The aim of this study was to find a viable way to minimize urban waste while maximizing the reutilization potential of this nutrient-rich source. This work demonstrated that similar tomato production and quality values (i.e., fruit weight, diameter, °Brix and phenolic content) can be achieved by partially replacing mineral N fertilizer with composted MSW, as compared with mineral fertilization only.
The present study was focused on the agronomic effects of fertilizing with MSW compost. It would be of special relevance to perform future studies on its environmental implications, in order to have a comprehensive evaluation of using MSW compost in those fertilization programs aiming to be sustainable.
Acknowledgements
The authors gratefully thank the EU Integrated Project TRUEFOOD (FOOD-CT-2006-016264) for its financial support. Albert Ribas-Agustí was a holder of an AGAUR-FI grant (IUE/2658/2007), with the support of Generalitat de Catalunya and the European Social Fund. The authors also thank the technical assistance of Mariano Bellelli and Elena Viñas, and the linguistic assistance from Rachel Kopec.
Conflict of interest
The authors have no conflicts of interest to disclose.
