Two and a Bud 58: 160-169,2011

RESEARCH PAPER

Assessment of co-compost quality produced from easily available biowaste by physico-chemical and exploratory data analysis T. Karak1, Chimpi Sarmah, I. Haque, R. M. Bhagat and K. Z. Ahmed Department of Soils, Tocklai Experimental Station, Tea Research Association, Jorhat -785008, Assam, India

ABSTRACT The dynamic physico-chemical parameters during transformation of six modes of co-compost with six easily available bio-wastes in and around tea garden are reported. Physico-chemical properties of all the co-composts revealed that all the prepared composts attained the Indian compost standard within 45 days. Hierarchical agglomerative cluster analysis during composting formed a cluster that included temperature, Cd, Cr, Cu, Pb, Ni, Zn, municipal solid waste (MSW) and phosphate (P) while the other included cation exchange capacity (CEC), cow dung, germination index (GI), Hg, nitrogen, total organic carbon (TOC), pH, pruning litters, potassium, pond sediments (PS), tea waste (TW) and water hyacinth. Principal component analysis, applied to the data sets indicate that the parameters responsible for heavy metals and P in prepared composts were MSW whereas nitrogen was IE inly related to PS and TW. These findings would be helpful in future planning with regard to eco-friendly composting of organic waste material.

there are varieties of organic materials which can be composted to obtain fertilizer value. However, traditional sources of soil organic raw materials such as animal waste, crop residues and green manure are rapidly declining due to their use as fuel and fodder (Lal, 2005).Under this circumstance, to switch over from conventional farming to organic farming there is a huge requirement of mammoth quantities of organic raw materials. To meet this requirement, an amalgamation of biowaste through composting is essential as application of undecomposed wastes or non-stabilized compost to land may lead to damage of crops by competing for oxygen, immobilization of plant nutrients and phytotoxicity (Epstein, 1997).

INTRODUCTION There is no denial of the fact that non-judicious uses of chemical fertilizers have caused a significant decline in soil health including soil organic matter and crop productivity over the last decades. To mInimize the adverse effects of conventional farming, organic farming is gaining familiarity (Stratton et aI., 1995). Organic farming is one key objective for sustainable soil management due to several reasons, including the improvement of soil protection and carbon sequestration in the context of climate change, soil aggregate (Lal, 2005), reduce runoff and soil erosion from typical hilly or terrain region (Edwards et ai., 2000), microbial biomass (Metzger et ai.,1987), the production of binding carbohydrates and humic substances (Albiach et ai., 2001) and soil fertility (Epstein, 1997) and at the same time it attracts higher financial premium. On this backdrop of information it can be inferred that composted organic amendment through organic farming should be one of the strategies towards environmentally favorable and safe agricultural management options (Kasim et ai., 2009). However, the progress of organic agriculture in India is very slow and only 41,000 ha of the total cultivated land is under organic farming, which is a mere 0.03% of the total cultivated area in India (Ramesh et ai., 2010). In nature, J

The composts prepared from different organic wastes differ in their quality and stability, which further depends upon the composition of raw material used for the compost production (Goyal et ai., 2005). Indian cities generate about 14 million tons of municipal solid waste (MSW) annually (Karak et ai., 2011) and in the face of a global decline of organic resource, MSW is gaining familiarity as an organic raw material for compost (Stratton et ai., 1995). However, there is justified concern regarding potential accumulation of heavy metals in surface soils from com posted MSW, including the possible migration of toxic metals into the food chain (Smith, 2009). Therefore, to reduce the risk of

Email address:[email protected]

160

excessive heavy metal contamination in soils and its uptake through plants, it is not advisable to utilize MSW as the sole raw materials for compost preparation. On the other hand, co-composting with different organic matters can increase the amount of raw materials and nutrient status in

composting of easily available biowastes in and around tea garden. Agglomerative hierarchical cluster analysis and PCA were also been used to investigate the suitability of raw materials used and to assess the influence of raw materials compost.

fmal product. Apart from MSW, other organic raw materials like pond sediments, water hyacinth, cow dung, tea pruning litter and tea waste are commonly available in and around tea (Camellia sinensis L.) gardens. Recent studies showed that pond sediments could be used effectively for rice cultivation as it contained several plant nutrients (Karak, 20 I 0). Water hyacinth (WH) and cow dung (CON) can be used as composting materials (Malik, 2007). Ample availability of tea pruning litter (PL) (litter produced from agricultural practices for tea garden) and tea waste (TW) (waste produced from tea factory) can also be used for cocomposting.

MATERIALS

AND METHODS

All the composting raw materials were collected from in and around the tea garden. Fresh cow dung (CON), municipal solid waste (MSW), pond sediment (PS), tea pruning litter (PL), tea waste (TW) and water hyacinth (WH) were used as raw materials for the present investigation. CON, PL and WH were collected from Tocklai experimental firm at Cinnamara, Jorhat, Assam, India. Non segregated MSW was collected from Jorhat Municipality. Thereafter MSW was sorted out to get biodegradable fractions by removing non-biodegradable materials such as plastics, polyethylene, glass and metallic cans. Bottom PS and TW were collected from main campus and model tea factory of Tocklai respectively. MSW, PL and WH were crushed to 1-1.5 em particle size prior to use. The major plant nutrients (N, P and K) of each waste used for composting are given in Table 1. Six modes of composting mixture were used for present investigation. Wet weight ratios ofthe composting mixtures were prepared in the following proportions: MIM2M3M4M5M6-

CON:MSW:PL:PS:TW:WH:: CON:MSW:PL:PS:TW:WH:: CON:MSW:PL:PS:TW:WH:: CON:MSW:PL:PS:TW:WH:: CDN:MSW:PL:PS:TW:WH:: CON:MSW:PL:PS:TW:WH::

2:2:2:2:1:1 1:1.5:1.5:2.5:2.5:1 2:2:1:3:0.5:1,5 1:4:2:1.5:1:0.5 3:2.5:1:1:1 :1.5 1.5:2.5: 1.5: 1.5: 1.5: 1.5

Fifteen kilograms (15 kg) of compostable material on wet weight basis for each treatment was thoroughly mixed and moisture was adjusted to 60% of water holding capacity (WHC) using pond water. The material was put in 20 L earthen pot. The top of the pot was covered with polyethylene sheet followed by mud layer (0.5 em) on it and allowed to decompose. All the composting modes were replicated thrice. The composting materials were turned

Therefore, the present work was conducted to monitor the physico-chemical changes that take place during co-

l. N, P and K (%) in the composting substrates

Parameters

of the prepared

Composting substrates and modes

Besides composition and physico-chemical changes during composting, compost quality is dependent on its stability and maturity (Goyal et al., 2005). A large number of literatures related to the physico-chemical changes in organic matters during compo sting have been reported. pH, CEC, total organic carbon (TOC), total nitrogen, total phosphorus, total potassium, heavy metals and germination index (Gl) have been considered as reliable variables of compost maturity (Goyal et al., 2005). However, in some maturity studies only few parameters have been discussed, while routine monitoring of composting progress needs to measure several variables. Furthermore, from the bibliographic survey, it is evident that there is paucity of interrelationship among the generated data of compost maturity. Therefore, the problem of data reduction and interpretation of multiconstituent physico-chemical measurements may be approached through the application of multivariate statistics and exploratory data analysis for compost maturity. The usefulness of multivariate statistical approach such as hierarchical agglomerative cluster analysis and principal components analysis (PCA) are widely used as unbiased methods to explore structure and relationships in multivariate data, and in some cases to predict responses (Hair et al., 1995). Till now no available work has dealt with the standardization for the present amalgamation and compost maturity on the basis of exploratory data analysis.

Table

on the quality and maturity

1.79 1.23 1.92 MSW PL WH PS 0.59 CDN (0.20 (0.700) I) 0.36' 0.29 0.62 (0.030) 0.19 (0.106) (0.090) (0.203) 1.48TW (0.209) 0.70 0.61 (0.265) 2.24(0.311) 2.20 3.68 1.61 2.65 (0.101) (0.200) (1.227) 0.21 (0.654) (0.103) 1.96 (0.225)

161

over one to three times per week upto 30 days and moisture was maintained 60% throughout the experiment. After every turn, the composting mass was covered with mud as stated earlier. Temperature was recorded in each pot twice daily, one in the morning (9:30 AM) and other in the afternoon (2:30PM) at a depth of 15cm.

tea soil was taken in 250mL plastic beaker. 0.2g (equivalent to 5 ton compost ha·l) of produced compost was thoroughly mixed with soil. Five seeds were placed in each beaker. Beakers were incubated at ambient temperature and natural humidity. Soil without compost was used as a control. Seed germination and root length in each beaker were measured after 96 h. In both the germination tests, the percentages of relative seed germination (RSG), relative root growth (RRG) and germination index (GI) were calculated by the protocol described by Hoekstra et al., (2002) as follows:

Physico-chemical analyses Homogenized samples of all feed mixtures were drawn on 15th, 30th and 45th day respectively. Based on the criteria for maturity of compost as revealed by Epstein (1997), 45th day was considered as the end of compo sting as all samples smelled like forest soil, appeared granular, became much darker in color and homogeneous than initial feed. For physico-chemical analysis, the samples were air dried in shade at room temperature, ground in an electric blender and stored in plastic vials. All the physico-chemical analysis were done on dry weight basis. For all the analysis the required chemicals were analytical reagent (AR) grade supplied by E. Merck, Mumbai, India. Double distilled water was used for analytical work. All the samples were analyzed in triplicate and results were averaged. The results were reproducible with in ±4% error limits.

RSG(UAi = IJumkrq/ s¥.:ds KcnninaJl!tI in 'Wil amcm/cd 1t',,11I.:wnpo,fit ,feed!> gC""lIIwle,/ msoil HJllhulIl rompo.ff numherof

RRG(%)=

mean

&1 (%)= R,~

1'001 h:ng'!I ill menn ruuf leI/gill

x

soil amended ,!'i111 compost ill .wil Wit/10ll' Lnll1l)(M"

x

x

100

JOO

RRG

100

Statistical analysis Row data table of weight ratio of composting substrates in different modes, physical and chemical parameters of the composting experiments were produced where rows of K data table refer to the same set of individuals, but their columns are associated to different sets of variables indexed by k

pH was measured in the slurry of the dried sample: water ratio of 1:5 following FAI (2007). Oxidizable organic C was determined by dichromate wet oxidation method described by Walkley and Black (1934). For determination of total N, samples were digested using concentrated H2S04 and catalyst mixture in Kjeldahl flask and subsequently N content in the digest was determined following steam distillation and titration method (Bremner and Mulvaney, 1982). Total phosphorus (P) and potassium (K) were determined by total digestion of the compost in di-acid mixture (HCI04: HNO] at 9:4 ratio) with subsequent analysis by spectrophotometer (VARIAN, Cary 50Bio, Australia) after color development using vanadomolybdate yellow color method for P (Olsen and Sommers, 1982) and flame photometric method for K (Systronics, Model-130, India) (Black, 1965). Heavy metals were measured only in the fmal product. For analysis of heavy metals Zn, Cu, Cd, Ni, Hg and Cr samples were burnt (dry ashing) in silica crucible at 550°C followed by digestion in Teflon Beaker using concentrated HF, H2S04, HNO] and HCI04 (Baker and Amacher, 1982). For analysis of Pb, samples were digested with concentrated HCI04 and HNO] acids as described by Black (1965). Determination of heavy metals was carried out through flame AAS (Varian Model No. AA 240, Australia). The accuracy and precision of the analyses for heavy metals were checked against the NIES, CRM 10-a, and the NIST SRMs 1515 apple leaves (USA) and BCR 62 Olea europaea (Brussels). The results found in all certified and reference elements were within 5-15% of

.

The brief notations for statistical analysis are as follows: Let,

Xk

(k = 1,....., K)

be the

nx

mk matrix oftheth set of

variables on the individuals and be an matrix, where . Each variable (the th column of) has zero mean and unitvariance. The rank of is assumed to be . These latter assumptions are made mainly in order to keep the notation simple. is the subspace of spanned by the columns of . and are, respectively, the correlation coefficient and the covariance between two variables and. is the variance of the variable. The data were also log-transformed to closely correspond to normal distribution. Further, all the 22 variables were standardized by calculating their standard scores (-scores) as follows: =

z ,

where

Xi

-x

S

= standard score of the sample; = value of sample = mean and = standard deviation.

Standardization scales the log-transformed data to a range of approximately ±3 standard deviations, centered about a mean of zero. In this way, each variable has equal weight in the statistical analyses. Besides normalizing and reducing outliers, these transformations also tend to homogenize the variance of the distribution (Rummel, 1970).

expected values (85-105% recovery). For the germination tests, green gram (Vigna radiate L.) and chickpea (Cicer arietinum L.) seeds were used. Two hundred grams (200g) 162

Standardization also tends to increase the influence of variables whose variance is small and reduce the influence

sanitized the composting process and killed many of the pathogens that might have been initially present in the mixture. This pattern of temperature change is consistent with that of other composting studies (Goyal et aI., 2005; Tognetti et al., 2007). The rapid progress from initial mesophilic phase to thermophilic phase indicates a high proportion of readily degradable substances and selfinsulating capacity of the waste irrespective of the composting mode used.

of variables whose variance is large. Furthermore, the standardization procedure eliminates the influence of different units of measurement and renders the data dimensionless. The raw data were also used for graphical representation (Box-Whisker diagram) whereas, the transformed (logtransformed and standardized) ones were used for hierarchical agglomerative cluster analysis (HCA) described by Sing et al. (2004) and factor analysis (PCA) were performed by the procedure described by Kano et al. (2001). A Pearson-correlation analysis was carried out to check the significances of the linear relations between 22 physical and chemical variables. All these statistical analyses were performed using SPSS version 13.

pH The minimum pH recorded was 6.98 at the start of composting (Fig. ID). After composting for 15-30 days there was a rapid and significant (p < 0.05) increase in pH to 8.92 (Fig. IE). An increase in pH during composting of different compo sting substrates were reported in other studies (Sundberg et al., 2004; Tognetti et al., 2007). After 30 days of composting pH declined. Thereafter, composting progressed and availed the Indian compo sting standard with respect to pH (Le. 6.5-7.5), regardless of composting modes (Fig. IF). Composting is affected by pH through selective microbial activity of acidophiles and alkalophiles and their successions act as an indicator of advancement in composting process (Sundberg et al., 2004).

RESULTS AND DISCUSSIONS Box-and-whisker plots provide a visual impression of the different modes of composting and shape of the underlying distributions. In the box-and-whisker plots, line across the box represents the median, whereas the bottom and top of the box show the locations of the first and third quartiles. The whiskers are the lines that extend from the bottom and

(8' "'"1

top of the box to the lowest and highest observations inside the region. Temperature

-

(..:>]

--

'l~

•

~:j

It

i~~

profiles

Y

I

~

d

~

~

~l._..J-!I, .._).C,_:),,0.... _,--_ .. U4 ),15

Figs. I (A-C) show box plots for temperature change related to the progress of composting. By inspecting these plots it was possible to perceive differences among the different modes of compo sting. In the first duration of composting (Fig. I A), the maximum temperature was recorded at 48.9°C in M2 and minimum was recorded at 29.3°C in M5. In the

I

T"" •••--,---,

Ml.,,"'¢!~ .•.•.• ,~

MI

M6

M:- ).II h.:"~).I~

)"C

ModtMorJlupo.«-«

•••

1'Io.)·r

"1 ,.."'" r!.

i

I~J ;....j t t!

H!

T

second phase of composting (Fig. I B) a sharp variation of temperatures was recorded. In this phase M2 method maintained temperatures greater than 50°C and MI with the least, 38 to 41.6°C. However, in the final stage (Fig. IC), the minimum temperature for all the composting modes showed ambient temperature. This drop seems sufficient to denote the completion of the compo sting process or one could suggest, could have been near to the end of the composting process, since after the last turning the maximum values were not regained (Stickelberger, 1975). In general composting is a biological process driven by active saprophytic microorganisms. Temperature changes selectively affect the presence, succession and activity of these saprophytes and can thus be used to track the progress ofthe compo sting process. In this study, there was an initial mesophilic phase which was succeeded by rapid change to a thermophilic phase before the temperature again dropped to a second mesophilic phase and continuec into the compost maturation phase irrespective of the modes used. Thermophilic stage in the compo sting is important as it

I

9

IT'] R

R

MlI'''II\.Il~llalull

:;60 cmol kg' I) (Fig 2F). Thus, it was found that CEC increased progressively from the start to the final phase at about e"60 after 45 days of composting. Similar result was reported by Harada and Inoko (1980).

>..

Carbon

(8)

;Q--~

~ "'~ .:!

.k1-!

~ 20:

]

1'-

i

fl-

Ml

r

r

"r--

M"2 \1"1 M4

M(-

(t)

~ ~

ratio

and 1.93 folds in final product than the initial. Decline of C:N ratio to less than 20 indicates an advanced degree of organic matter stabilization and reflects a satisfactory degree of maturity of organic wastes (Senesi, 1989). Furthermore, in the present study, the fmal values ofthe CIN ratio suggest that all composts had reached an acceptable degree of maturation (FAI, 2007) except in M3 (C:N=20.12).

·'-·-~·f M~

nitrogen

The initial C:N ratio has long been used to predict decomposition rate. Very wide (>35) and narrow «12) CIN ratios are not suitable for composting (Epstein, 1997). However, in the present study, the initial C:N values for composting substrates were within the recommended range. The loss of carbon as CO2 and simultaneously the loss of nitrogen lower the C:N ratio of the substrate. Since C:N ratio decreased over time, C losses were presumably greater than TN losses (Figs. 2A-C and 3A-C, respectively). It is evident that C:N ratios (data not shown) decreased with time in all modes of composting. The C:N ratios in the first fifteen days of composting were in the range between 24.32 and 38.52. C:N ratios in fmished product were in the range between 14.52 and 20.12. Decline of C:N ratios were between 1.31

6(1 ... !

(.

of total nitrogen

(I))

.

l-fWu",,,n lrldMl bnt

$

;

Q

~

F.

~

M'l

~ri

M)

M~

U:S U6

Total phosphorus

M-.>dIl ••.• fO-