STOTEN-21787; No of Pages 8

Science of the Total Environment xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Acid drainage from coal mining: Effect on paddy soil and productivity
of rice
Burhan U. Choudhury a,b,⁎, Akbar Malang a, Richard Webster b, Kamal P. Mohapatra a, Bibhash C. Verma a,
Manoj Kumar a, Anup Das a, Mokidul Islam a, Samarendra Hazarika a
a
ICAR Research Complex for North Eastern Hill Region, Umiam, Meghalaya 793 103, India
b
Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, United Kingdom

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Mine drainage acidified paddy soil and

increased exchangeable Al3 + satura-
tion.
• Sulfur and extractable heavy metals in-
creased in excess of critical limits.
• Paddy grain yield declined by 62% in
contaminated than from unaffected
fields.
• Concentration of excess sulfur signifi-
cantly reduced biomass and grain yield.
• Paddy fields recovered some of their
productivity 4 years after mining
ceased.

a r t i c l e i n f o a b s t r a c t

Article history: Overburden and acid drainage from coal mining is transforming productive agricultural lands to unproductive
Received 25 October 2016 wasteland in some parts of Northeast India. We have investigated the adverse effects of acid mine drainage on
Received in revised form 11 January 2017 the soil of rice paddy and productivity by comparing them with non-mined land and abandoned paddy fields
Accepted 12 January 2017
of Jaintia Hills in Northeast India. Pot experiments with a local rice cultivar (Myngoi) as test crop evaluated bio-
Available online xxxx
logical productivity of the contaminated soil. Contamination from overburden and acid mine drainage acidified
Editor: D. Barcelo the soil by 0.5 pH units, increased the exchangeable Al3+ content 2-fold and its saturation on clay complexes
by 53%. Available sulfur and extractable heavy metals, namely Fe, Mn and Cu increased several-fold in excess
Keywords: of critical limits, while the availability of phosphorus, potassium and zinc contents diminished by 32–62%. The
Acid mine drainage grain yield of rice was 62% less from fields contaminated with acid mine drainage than from fields that have
Agriculture not suffered. Similarly, the amounts of vegetation, i.e. shoots and roots, in pots filled with soil from fields that re-
Overburden ceived acid mine drainage were 59–68% less than from uncontaminated land (average shoot weight: 7.9 ±
Paddy soil 2.12 g pot−1; average root weight: 3.40 ± 1.15 g pot−1). Paddy fields recovered some of their productivity
Rice productivity
4 years after mining ceased. Step-wise multiple regression analysis affirmed that shoot weight in the pots and
grain yield in field were significantly (p b 0.01) and positively influenced by the soil's pH and its contents of K,
N and Zn, while concentration of S in excess of threshold limits in contaminated soil significantly (p b 0.01) re-
duced the weight of shoots in the pots and grain yield in the field.
© 2017 Elsevier B.V. All rights reserved.

⁎ Corresponding author at: Division of NRM, ICAR Research Complex for North Eastern Hill Region, Umiam, Meghalaya 793 103, India.
E-mail address: burhan3i@yahoo.com (B.U. Choudhury).

http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
0048-9697/© 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Choudhury, B.U., et al., Acid drainage from coal mining: Effect on paddy soil and productivity of rice, Sci Total Environ
(2017), http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
2 B.U. Choudhury et al. / Science of the Total Environment xxx (2017) xxx–xxx
1. Introduction
The total coal reserve (mostly Tertiary) in the northeast India is esti-
mated to be about 109 t. The northeastern states of Meghalaya have a re-
serve of about 640 Mt. (Indian Minerals Yearbook, 2012), and the Jaintia
Hills district of Meghalaya alone has about 40 Mt of sub-bituminous
coal. Mines from major coal belts in the district, namely Bapung, Sutnga
and Khliehriat, produce nearly 40 million tonnes of coal per year; that is
N80% of the total production of the state. The district has traditionally
depended on agriculture, dominated by the production of paddy rice.
However, with the introduction of opencast coal mining since 1970s
there has been rapid degradation of the natural resources. In this form
of surface mining, huge quantities of spoil, or overburden, in the form
of gravel, rock, sand and soil are dumped on the arable land adjacent
to the mines. These materials typically contain much sulfur (2 to 12%)
in the form of metal sulfides (mostly iron pyrites), haematite and
gibbsite (Dowarah et al., 2009; Nayak, 2013). Acid drainage and over-
burden from coal mines are acidic (pH: 2.0–3.0) and are rich in iron
compounds, which on weathering and oxidation release sulfate, Fe2+
and other metallic ions. Due to heavy rain (N5000 mm year−1), the in-
flux of these highly acidic contaminants to the nearby otherwise fertile
paddy fields are making the land unfit for cultivation (Swer and Singh,
2003; Sahoo et al., 2012). As a result, within the 5 years from 2007 to
2012 the production of rice in mine-affected paddy fields of Jaintia
Hills declined from 1.80 t ha−1 to 1.50 t ha−1 (Anonymous, 2013).
The impacts of the mining, its wastes and acid drainage on the sur-
rounding vegetation, water quality, biodiversity, eco-restoration, and
re-vegetation strategies have been studied across India (Swer and
Singh, 2003; Sarma, 2005; Dowarah et al., 2009; Nayak, 2013; Sahoo
et al., 2012, 2013; Masto et al., 2015). However, the impact of acid
mine drainage on the soil of paddy fields and rice productivity has not
received equal attention. Therefore, we studied the effect of acid mine
drainage from the major coal mining and over burden on soil properties
of the paddy fields and rice productivity in the major coal belts across
Jaintia Hills of Meghalaya, India. We also compared results with unaf-
fected land and paddy soils where mining has ceased and land is recov-
ering naturally. To complement the comparison we also did a pot
experiment involving local rice genotype to evaluate the effect of acid
drainage on paddy soils and its productivity for paddy rice. The findings
should help in devising strategies to restore paddy fields that have suf-
fered as a result of the mining and for sustaining their productivity in
the future.
2. Material and methods
2.1. Location, soil and climate of the study area
Our study area is in the Jaintia Hills district of Meghalaya (between
25° 13′ and 25° 47′ N latitude and 91° 23′ and 91°56′ E longitude)
(Fig. 1), with an elevation ranging from 1150 m to 1300 m above
mean sea level (See Appendix A for details about study area). It is one
of the most extensively mined districts of Northeast India. The district
covers 3819 km2. The topography is characterized by plateau incised
by valleys with steep to gently sloping sides. The soil of the region devel-
oped on rocks of geological ages ranging from Pre-Cambrian to Recent.
The parent material is composed of sandstones and shales with coal
beds, and the Sylhet limestone. The valleys contain transported mate-
rials of shale origin and recent alluvium. Taxonomically, two soil orders,
namely Cambisols and Alisols, dominate the region. The ground water
table in the valley is approximately 3 m deep.
The climate is of the tropical monsoon type with an annual average
rainfall exceeding 5000 mm, 77% is of which falls during Monsoon (May
to September). August is the hottest month, with average minimum and
maximum temperatures of 18.4 °C and 24.5 °C, respectively (see Appen-
dix A). The coldest month is January when the average minimum and
Please cite this article as: Choudhury, B.U., et al., Acid drainage from coal mining: Effect on paddy soil and productivity of rice, Sci Total Environ
(2017), http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
maximum temperatures are 8.7 °C and 16.6 °C, respectively. The aver-
age relative humidity varies from 85% in July to 61% in December.
2.2. Land use
The vegetation and land use in the region are varied. In 1975 forest
occupied about 25% of the land, but by 2007 that cover had shrunk to
only 12.5% (Sarma et al., 2010). Somewhat N 55% of the land is dominat-
ed by scrub. The steeper slopes bear pine forest (Pinus kesiya). In the
low-lying flat land in the valleys farmers grow rain-fed rice. They trans-
plant the rice into puddled soil during the monsoon season (June to Sep-
tember or October) and harvest the crop in October or November. In
winter; most of the land remains fallow. In the recent past, coal mining
has led to major changes in land use: most of the productive arable land
has gradually been affected by the mining. In the Jaintia hills the mined
land increased from 3.26% to 11.2% of the total area between 1975 and
2007, while the land under agriculture decreased from 2.65% to 1.43%
(Sarma et al., 2010). In two years (2005 to 2007) the area under
paddy cultivation has declined from 16,593 ha to 12,195 ha in the dis-
trict (Anonymous, 2013).
Coal mining in Jaintia Hills is done by surface (open-cast or open-pit)
and sub-surface (rat-hole) methods. In the study area, coal is mined by
open-cast methods with extensive surface excavation near both the for-
est and paddy fields. Overburden (mine spoil) from the mines is
dumped in the nearby paddy fields (Fig. 2). During rainy seasons, acid
drainage from the overburden and mines themselves spills on to the
paddy fields, and in extreme cases yellowish precipitates of iron hy-
droxides form in the fields (Fig. 2) (Sarma, 2005). As a result, after 2
to 3 years of continued mining, the productivity in the fields influenced
in this way declines drastically so that farmers are often forced to aban-
don those fields.
2.3. Site selection, soil sampling and analysis
With the help of a landuse map of the Jaintia Hills district, we tra-
versed the region where mining had affected large proportions of the
agricultural land (paddy field), and we chose to sample along a transect
through the centre (Fig. 1). We distinguished three forms of land use, as
follows.
1 Land where mining is current and has continued to pollute the land
with overburden or acid drainage or both for at least 10 years.
2 Land where mining once contaminated the land but where mining
ceased and has been allowed to recover for the last 4 years.
3 Land near to mines but which has not been contaminated by either
overburden or acid drainage.
See Appendix A for detailed lists.
From class 1 we chose 60 fields at random, and from each we took
some topsoil (0–15 cm) with a spade in February and March 2009 be-
fore the onset of the monsoon. From each of classes 2 and 3 we did like-
wise but from only 30 fields in each class. The soil from the
contaminated fields contained some coarse debris. We removed the
large stone sand also coarse plant litter (mostly crop residues from
paddy). A proportion of each soil sample was used, without sieving,
for a pot experiment, described below. Several undisturbed clods of
soil (6–10 cm across) were collected at each site and used for the deter-
mination of aggregation (mean weight diameter) by wet sieving follow-
ing standard procedures (Jalota et al., 1998). The remainder of each
sample was dried in air, ground and passed through a 2-mm sieve
(0.5 mm sieve for determination of organic carbon) for chemical and
particle-size analysis. The last was done by the Bouyoucos hydrometer
method (Jalota et al., 1998)
The air-dried, sieved samples were analysed for pH (1:2 soil: water),
exchange acidity, cations Ca2+, Mg2+, K+ and Na+ extracted in ammo-
nium acetate at pH 7.0, extractable available nitrogen (N) by alkaline
potassium permanganate (Jackson, 1973), available phosphorus (P) by
 B.U. Choudhury et al. / Science of the Total Environment xxx (2017) xxx–xxx 3

Fig. 1. Location of the Jaintia Hills in Northeast India.

Brays-II (Bray and Kurtz, 1966), available potassium (K) by 1 M NH4OAC
and measured by flame photometer, available sulfur (S) by the turbidi-
metric procedure (Jackson, 1973) and organic carbon (size b 0.5 mm)
by the Walkley and Black (1934) method. Exchangeable aluminum
(Al3("") +) content was determined colorimetrically after extraction
with 1 M KCl (Lin and Coleman, 1960). Effective cation exchange capac-
ity (actual field cation exchange capacity) was calculated as the sum of
all exchangeable cations (Ca2("")+, Mg2+, K+ and Na+) and exchange
acidity (Jackson, 1973). Available Fe, Mn, Cu, and Zn were determined
in 0.005 M DTPA extract followed by atomic absorption spectrometry
(Lindsay and Norvell, 1978).
2.4. Pot experiment on rice
A pot experiment was carried out in the shade house of Soil Science
Division, ICAR Research Complex for NEH Region, Umiam, Meghalaya,
during 2010. Pots were filled with un-sieved soil from which coarse
plant residues had been removed and compacted to a uniform bulk den-
sity of 1.40 g cm−3. In all 120 pots were filled with soil collected from
the three classes of land; there were 60 pots of soil currently being con-
taminated and 30 from each of the other two classes. No manures or fer-
tilizers were applied in the pots. The soil was kept saturated by frequent
watering as it would be in the field. The local rice cultivar, Myngoi, was
planted as test crop, and two seedlings were allowed to grow in each
pot for 45 days. The biomass of the rice (root and shoot separately)
was harvested 45 days after sowing. Harvested plants were washed
and dried in air at room temperature. Dry weights of roots and shoots
were determined separately after oven-drying at 60 °C until a constant
weight was reached.
2.5. Field sampling
In the field, 20 fields where farmers where growing rice on land cur-
rently affected by mining, 10 fields on recovering land and 10 fields on
land unaffected by mining were chosen at random for measurements of
rice yield. Again the variety was Myngoi, and it was allowed to mature
and was harvested 95–100 days after planting. Grain was harvested
from an area of 6 m2 in each field, and the grain yield was expressed
on hectare basis at 14% moisture content.
2.6. Statistical analysis
The designs of both the pot experiment and the field sampling lead to
a simple balanced one-way analysis of variance (ANOVA) with three
treatments and residual within-treatment completely randomized struc-
ture. Yields from the fields and biomass from the pots were analysed with
this design in GenStat (Payne, 2015). So too were the chemical properties
of the soil mentioned above. From the residual variances we calculated

Please cite this article as: Choudhury, B.U., et al., Acid drainage from coal mining: Effect on paddy soil and productivity of rice, Sci Total Environ
(2017), http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
4 B.U. Choudhury et al. / Science of the Total Environment xxx (2017) xxx–xxx

Fig. 2. Effects of open-cast mining on paddy fields in the Jaintia Hills: (a) a mine (left) and overburden (right); (b) acid drainage (centre) from overburden (right) spilling into paddy fields;
(c) precipitated iron oxide from acid mine drainage; (d) rice plants 45 days old from unaffected soil (left) and contaminated soil (centre and right).

standard errors of the class means (SEMs) and least significant differences
(LSDs) between them at the 5% level of probability.
To reveal likely causes of the evident differences in yields between
contaminated and uncontaminated land we computed Pearson correla-
tion coefficients between both shoot weight (Y1) and grain yield (Y2)
and the soil chemical properties. Several of the correlations appeared
to be strong, and to rank them in order of importance we did step-
wise multiple regressions of Y1 and Y2 as response variables on the soil
chemical properties as predictors. We judged the predictors significant
if the probability that their contributions were null was b 0.05, i.e.
(p b 0.05 of the null hypothesis).
3. Results
3.1. Effect of mining on soil acidity and other chemical properties
The soil of the region is naturally very acid, and the soil in the paddy
fields unaffected by mining that we sampled was similarly acid with an
average pH of 4.30 (Table 1). Mining, with its discharge of acid drainage
and release of sulfur from minerals in the overburden decreased the
soil's pH by 0.51 units to 3.79 in the soil currently affected. The pH ap-
pears to be increasing somewhat, now 3.96, in soil recovering from
the contamination. Closely related to these changes are those of

Table 1
Effects of acid mine drainage on paddy soil properties and heavy metal concentration in the Jaintia Hills.

Soil variables Paddy fields

Unaffected RM* CF SEM (30s vs. 30s) SEM (30s vs. 60s) LSD(30s vs. 30s) LSD(30s vs. 60s)

pH (1:2 soil:water) 4.30 3.96 3.79 0.021 0.015 0.058 0.050

Exch. Al3+/cmol (P+) kg−1 1.64 2.13 3.11 0.025 0.018 0.07 0.06
Exch. Ca2+ + Mg2+/cmol (P+) kg−1 2.82 2.13 1.82 0.022 0.015 0.061 0.053
Al3+ saturation/% 34.1 41.8 44.9 0.69 0.488 1.932 1.673
Organic carbon/% 1.50 3.01 3.60 0.048 0.034 0.135 0.117
Mean weight diameter/mm 2.82 1.63 1.15 0.055 0.039 0.157 0.112
Available N/kg ha−1 261.3 305.1 306.6 1.8 1.27 5.03 4.36
Available P/kg ha−1 24.3 18.9 16.9 0.302 0.214 0.847 0.733
Available K/kg ha−1 210.2 132.6 77.2 1.02 0.72 2.84 2.46
Available S/kg ha−1 41.1 43.0 170.8 0.81 0.57 2.26 1.96
DTPA Fe/mg kg−1 84.7 200.7 260.0 1.103 0.78 3.091 2.677
DTPA Mn/mg kg−1 7.24 8.94 13.51 0.16 0.11 0.44 0.38
DTPA Cu/mg kg−1 0.40 0.69 0.70 0.013 0.009 0.035 0.031
DTPA Zn/mg kg−1 0.54 0.33 0.19 0.008 0.006 0.024 0.02

* RM: recovering from mining; CF: currently affected by mining; 30 and 60: 30 and 60 numbers of samples collected; LSD at 5% level of significance.

Please cite this article as: Choudhury, B.U., et al., Acid drainage from coal mining: Effect on paddy soil and productivity of rice, Sci Total Environ
(2017), http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
 B.U. Choudhury et al. / Science of the Total Environment xxx (2017) xxx–xxx 5

exchangeable Al3+. Mining appears to have increased the concentration Table 2

of Al3("")+ nearly 2-fold, if we take the uncontaminated soil as reference, One-way analysis of variance (ANOVA) of dry matter (shoot and root), and grain yield of
rice grown across unaffected, recovering from mining and currently mine affected paddy
and increased its saturation of the exchange complex from 34.1% to soils.
44.9%. Four years after mining ceased the saturation by Al3+ has dimin-
ished somewhat to 41.8%. The concentrations of the exchangeable Source of variation d.f. s.s. m. s. F ratio P

Ca2("") + and Mg2 + have been affected in a contrary sense. Mining Shoot dry matter
seems to have caused a significant (p b 0.05) 35% decrease in the con- Treatment 2 500.56 250.28 738.6 b0.001
Residual 117 39.64 0.34
centrations of both cations. However, 4 years after mining ceased, the
Total 119 540.21
concentrations of Ca2+ and Mg2+ increased by 17%. Root dry matter
The soil affected by mining now has a larger proportion of sand and Treatment 2 98.85 49.43 636.4 b0.001
correspondingly less clay than the original soil (see Appendix A). This Residual 117 9.08 0.08
extra sand is presumably from the overburden. Soil aggregation was sig- Total 119 107.94
Grain yield
nificantly (p b 0.05) weaker in the soil affected by mining; the mean Treatment 2 27.00 13.50 600.9 b0.001
weight diameter was less than half that of the reference soil (Table 1). Residual 37 0.83 0.022
Four years after mining ceased, however, aggregation had strength- Total 39 27.83
ened; the mean weight diameter had recovered by about 40% (Table 1).

3.2. Effect of mining on organic carbon, macro-nutrients and heavy metals
The mine-affected soil contained much more organic carbon (3.6%)
than the reference (1.5%) (Table 1). We attribute this extra carbon to
its source in the overburden. It was accompanied by a greater availabil-
ity of nitrogen (N). The mine-affected paddy soil contained 31% less
available P and 63% less available K than the unaffected soil. Contamina-
tion from mining increased the available sulfur in soils by 7–8 times in
excess of critical limits for Indian soils (22–25 kg ha− 1, Goswami et
al., 2009) and was 4 times greater than in the unaffected soil (41 kg
ha−1). Recoveries in these nutrients after mining ceased were mixed.
The amounts of available N and P had scarcely changed. That of K in-
creased substantially, and the amount of S had decreased almost to
that of the reference (Table 1) in just the 4 years.
The concentrations of the heavy metals Fe, Mn, and Cu extracted by
DTPA were substantially greater in the mine-affected soil than in the
reference (Table 1) and exceeded by far critical limits for the metals ex-
tracted by DTPA, namely 2.5–4.5 mg kg−1 soil for Fe, 2.0 mg kg−1 soil
for and 0.2 mg kg−1 soil for Cu in Indian soils (Katyal and Randhawa,
1983). However, only Fe content exceeded the toxic thresholds
(N100 mg kg− 1 soil) (Bitondo et al., 2013). We presume that these
metals derive from the mine drainage and the overburden. In contrast,
much less zinc was available in the mine-affected soil than in the refer-
ence. In contrast to heavy metals Fe, Mn, and Cu, contamination from
mining resulted in a marked diminution in the concentration of Zn:
64% less than in the unaffected soil. Four years after mining ceased the
concentrations of Fe and Mn were less by 20–30% than in the soil still
being affected by mining; they were nevertheless well in excess of the
threshold for critical (Mn:2.0 and Fe:2.5–4.5 mg kg−1 soil) and toxic
limits of Fe (N100 mg kg−1). The concentration of Zn increased in that
time but was still much less than the critical threshold of 0.6 mg kg−1
soil (Katyal and Randhawa, 1983) (Table 1).
3.3. Effect of mining on rice productivity
Parameter estimation reflected steep and positive slopes of pH and
available K with shoot weight, suggesting that with increase in pH and
available K, shoot dry matter also increased significantly (Table 4). Ex-
cess of available S (N170 kg ha−1) from acid drainage exceeded the crit-
ical threshold limits (22–25 kg ha− 1) (Goswami et al., 2009), by
several-fold, and this effect is almost certainly the cause of the negative
influence of S (Table 4).
3.3.2. Field survey
Table 2 summarizes the analysis of variance of grain yields from the
farmers' fields. As in the pot experiment, the effect of soil is hugely sig-
nificant (p b 0.001). Table 3 lists the mean yield from the three kinds of
soil. The mean yield from the currently mine-affected soil (1.21 t ha−1)
was only about one third that of the reference (3.20 t ha−1) (Table 3).
The fields recovering from mining produced 2.14 t ha−1. Our multiple
regression model revealed that pH followed by Zn and K all significantly
(p b 0.001) and positively influenced grain yield while S negatively in-
fluenced the grain yield (Table 4). Fig. 3 summarizes these effects as re-
gressions of yield on the four soil variables (the black lines) and by the
scatter of the individual measurements. The figure also shows how
strongly clustered the points for three classes of land and why the anal-
ysis of variance of yield, and shoot growth in the pots, produces such
large F ratios. Notice that for pH, K and Zn the clusters for the land recov-
ering from mining lie between those for the land currently being affect-
ed and the unaffected reference soil, whereas the cluster for S in the
recovering land lies directly above that of the reference soil.
4. Discussion
4.1. Effects of coal mining on soil acidity and heavy metals
Seepage and influx of acids from mining in paddy fields further acid-
ified the already acidic soils (pH b 4.4) of Jaintia Hills. Mining resulted in

Table 3
3.3.1. Pot experiment
Effect of mine drainage on shoot and root dry matter production in pot and grain yield of
Table 2 summarizes the analysis of variance for the weights of both rice in paddy fields.
shoots and roots in the pots. The soil effect is hugely significant for
both (p b 0.001). Table 3 lists the mean dry weights of shoots and Treatment Dry matter, Treatment Grain yield
g pot−1
roots grown in the three kinds of soil, their standard errors and the
least significant differences at the 5% level. The mean weight of shoots Shoot Root (t ha−1)

(3.20 g per pot) in the soil currently affected by mining was less than Reference (30s) 7.94 3.40 Reference (10s) 3.20
half that of the unaffected soil (7.94 g pot−1), and the mean weight of RM("")(30s) 6.30 2.32 RM("")(10s) 2.13
CA("")(60s) 3.20 1.21 CA("")(20s) 1.21
roots (1.21 g pot−1) less than one third (3.40 g pot−1). Root growth
SEM (30s vs. 30s) 0.107 0.051 SEM (10s vs. 10s) 0.047
in the soil recovering from mining (2.32 g pot−1) was still substantially SEM (30s vs. 60s) 0.076 0.036 SEM (10s vs. 20s) 0.034
less than in the reference soil, but the mean weight of shoots (6.30 g LSD(30s vs. 30s) 0.30 0.143 LSD(10s vs. 10s) 0.136
pot−1) was only a little less than in the reference soil. Step-wise multi- LSD(30s vs. 60s) 0.26 0.124 LSD(10s vs. 20s) 0.118
ple regression analysis revealed that five variables, namely pH, N, K, S, LSD at 5% level of significance; 10, 20, 30 and 60: 10, 20, 30 and 60 numbers of samples
and Zn significantly (p b 0.001) influenced the weight of shoots. collected.

Please cite this article as: Choudhury, B.U., et al., Acid drainage from coal mining: Effect on paddy soil and productivity of rice, Sci Total Environ
(2017), http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
6 B.U. Choudhury et al. / Science of the Total Environment xxx (2017) xxx–xxx

Table 4 2007; Sahoo et al., 2013). This might have resulted in the large (toxic)
Parameters of multiple regression of shoot dry matter (Y1) and grain yield (Y2) of rice on concentration of Al3 + and its substantial saturation of the exchange
variates: pH, potassium (K), zinc (Zn), sulfur (S), nitrogen (N) of paddy soils
(Y1 = a0 + a1 × pH + a2 × K + a3 × Zn + a4 × S + a5 × N;
complex, and the accumulation of Fe, Mn and S in the soil. Our reported
Y2 = a0 + a1 × pH + a2 × K + a3 × Zn + a4 × S). For grain yield, the effect of N was mar- values for Fe content (200–260 mg kg−1) in the currently affected soil
ginal (non-significant at P b 0.1), so it was not included in the model. The residual degrees and the soil recovering from mining still exceed the toxic threshold
of freedom for shoot dry matter and grain yield are 114 and 36, respectively and the ex- for soil (N 100 mg kg−1) (Bitondo et al., 2013). Unlike Fe and Mn, the
planatory variables were significant at P b 0.001.
soil's content of available Zn increased with increase in pH (r =
Parameter Shoot dry matter Grain yield +0.58 at p b 0.05). The availability of Zn depends strongly on pH and
F-ratio 874.5(df-5,114) 906.9(df-3,36) is greater in acid soils (Katyal and Randhawa, 1983). Despite the low
a0 −11.30 −2.755 pH of the soil, lack of sufficient Zn-containing minerals (sphalerite) in
a1 3.076 0.728 the coal deposits of Jaintia Hills, which are rich in Fe dominant pyritic
a2 1.546 0.011 ores, might have resulted in the small Zn content (Sahrawat, 2005;
a3 0.009 0.626
a4 −0.008 −0.001
Carmona et al., 2009).
a5 0.015 *
R2 adj 97.4 98.9 4.2. Effect of coal mining on soil organic carbon and availability of macro-
d.f: degrees of freedom; R2 adj: Adjusted coefficient of determination. nutrients

The significantly larger (N57%) content organic C in mine-affected

toxic concentration of exchangeable aluminum (Al3 +) and enhanced
saturation of the exchange complex by Al3+. It also led to large, toxic
concentrations of S and heavy metal Fe, and large concentration of
other heavy metals, Mn and Cu (Masto et al., 2015). These too can be at-
tributed to the continuous influx of mineral contaminants (iron sulfides,
oxy-hydroxide and oxyhydroxy-sulfates, and trace elements) into the
fields. The Tertiary coal deposits of Jaintia Hills are rich in sulfur (2 to
12%) and the dominant mineral phases are biogenic pyrites (rich in Fe
and S), marcasite, jarosite, siderite, etc. (Sahoo et al., 2012; Nayak,
2013). As a result, drainage from the mineral waste is strongly acidic
(pH: 2.0–3.0), and also very rich in S, Fe, Al, Mn, and SO2− 4 (Dowarah
et al., 2009; Sahoo et al., 2013). Oxidation of pyrites (at pH b 4.0) re-
leases oxidized Fe (Fe3+) and sulfate (SO2−4 ) compounds (Nikhil et al.,
soil may be attributed to the presence of coal dust and carbonaceous
shell in the overburden materials and their deposition through influx
in contaminated paddy field (Maiti, 2007). In like manner, Masto et al.
(2015) reported 46% more organic C in open-cast than underground
mine-affected non-agricultural soils in the Raniganj coalfield of India
and they thought that deposition of fine coal from spoil on the soil sur-
face contributed to the effect. The mine-affected soil contained more
available N than did the unaffected soil. This contrast with the reported
mean value of only 84.2 kg ha−1 in the mine spoils soil of Gevra colliery
of Madhya Pradesh, India (Banerjee et al., 2000). In the mine-affected
paddy fields of Jaintia Hills, where rice yields are poor (1.1 tha− 1),
farmers often apply large dressings of nitrogen fertilizer in the form of
urea, and this might account for the large amount (N260 kg ha−1) of

Fig. 3. Scatter diagrams of grain yield against four soil properties: (a) pH, (b) K, (c) S and (d) Zn. The oblique lines are of the regressions of yield as the dependent variable on those four
properties, and the points are the individual observations.

Please cite this article as: Choudhury, B.U., et al., Acid drainage from coal mining: Effect on paddy soil and productivity of rice, Sci Total Environ
(2017), http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
 B.U. Choudhury et al. / Science of the Total Environment xxx (2017) xxx–xxx
available N that we found. Nevertheless, it falls well within the range
(260 to 490 kg ha−1) as reported by Sarma (2005) for the mine-affected
soils of Meghalaya. In contrast, the mine-affected fields were severely
deficient in available P (b 25 kg ha−1), for which the critical threshold
lies in the range 34–68 kg ha−1 in acid soils (Goswami et al., 2009). Sev-
eral factors could be responsible: e.g. low pH (3.80) and large amounts
of Fe-, Mn- and Al-oxides, which might have adsorbed much of soluble P
and fixed it in unavailable forms (Banerjee et al., 2000; Maiti, 2007).
The substantial decrease in available K in the mine-affected soil may
be attributed to the significantly larger saturation of the exchange com-
plex by Al3+, Fe2+ and Mn2+ cations, which might have competed for
retention sites and forced K+ into solution and led to its leaching. The
mean value for available K (77.2 kg ha−1) that we found in the mine-af-
fected paddy soil falls within the previously reported range (67.8 to
289.4 kg ha−1) for the mined soils of Jaintia Hills (Nikhil et al., 2007).
4.3. Effect of coal mining on biological and economic (grain) yield of rice
crop
The growth of shoots and roots was significantly (p b 0.05) less in
mine-affected soils than in the reference soil. We attribute this to the
greater acidity, toxic concentration of Fe and S and to the smaller con-
centrations of K and Zn. A similar trend was also observed on significant
(p b 0.05) reduction in grain yield of rice in the mine-affected fields. This
finding corroborates the results of pot experiment as well as the under-
lying causes for declining biological (biomass) and economic (grain)
yield. Dowarah et al. (2009), Nayak (2013) and Sahoo et al. (2013)
have reported the strong acidity and small concentrations of K and Zn
due to toxic of Fe- and S-containing pyrite minerals in mine-contami-
nated paddy soil. Along with poor soil physical condition, this might
be the reason for the small grain yields that we found. Low pH results
in excess Fe (toxic) that causes deficiencies of several nutrient elements,
especially P, K, Ca, Mg, and Zn in lowland acidic rice soil and results in
yield reduction of rice from 12% to complete failure (Sahrawat, 2005).
In our mine-affected soils, abiotic stresses (toxicity in Al3+, Fe and S
plus P deficiency) induced by acidity reduced both the dry matter in
the pots and grain yield in the field compared with the non-mined ref-
erence fields. We also observed significantly (p b 0.05) greater concen-
trations of extractable heavy metals in shoots of rice grown in pots with
mine affected soils: Fe (62 mg kg−1) and Mn (130 mg kg−1) compared
with the rice grown in pots filled with non-mined soil (Fe:53 mg kg−1,
Mn: 73 mg kg−1). The concentration of Zn in shoots were significantly
(p b 0.05) greater in non-mined soils (53 mg kg−1) than the mine af-
fected soils (42 mg kg−1). Cessation of mining for 4 years, however, re-
duced the concentration of Fe in the shoots drastically (41 mg kg−1),
while the concentrations of Mn (126 mg kg−1) and Zn (44 mg kg−1)
showed no sign of change (improvement) and were comparable to
those in the mine-affected soil.
Our results are similar to those of Martinez et al. (2013) who
experimented on coal-mine-contaminated paddy soil in QuangNinh
province of Northeastern Vietnam; they reported reductions of 10–
60% rice root weight of both native and introduced cultivars. They also
reported absorption of toxic heavy metals (cadmium, copper and
lead) by rice plants (shoot and grain) exceeded permitted concentra-
tions in foods. Several other investigators (Abreu and Lavorante, 2003;
Dowarah et al., 2009; Martinez et al., 2013) have discovered that exces-
sive aluminum, very low pH and concentrations of S, Fe and Mn exceed-
ing thresholds in mine-contaminated soils resulted in phyto-toxicity
and accumulation of toxic metals (Fe, Cu, Zn, Mn, Pb, Cd, Hg, etc.) in
plants, including rice; root growth was inhibited, uptake of phosphorus
was diminished, and as a result crop performance severely declined.
Other investigators (Stephan et al., 1994; Conesa et al., 2006;
Nagajyoti et al., 2010; Manara, 2012) have found that these conditions
often lead to accumulation of metals in human organs through the
food chain and in turn to severe disease.
Please cite this article as: Choudhury, B.U., et al., Acid drainage from coal mining: Effect on paddy soil and productivity of rice, Sci Total Environ
(2017), http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
7
4.4. Soil properties and rice productivity―response to abandoned coal min-
ing activities
Our results show that when mining ceases the soil begins to recover
its productivity. After 4 years the pH had risen somewhat, the concen-
tration of Al3+ had diminished, and the availability of the macro-nutri-
ents, in particular K, and the micro-nutrient Zn had increased. The heavy
metals Fe, Mn and Cu had declined in concentration. Only S had
returned to its pre-mining status as judged by the measurements on
the unaffected soil, which we treat as the reference.
Cultivation of paddy involving intense physical manipulation (pud-
dling), and occasional draining of water might have acted naturally in
the reclamation and partial improvement in the soil resulting from acid-
ification. Absence of any fresh influx of acid drainage into the paddy
fields might have arrested further acidification of the soil, and prevented
further accumulation of sulfur and heavy metals (Monterroso et al.,
1999). In the paddy fields, cultivation of a single rice crop in year, and
leaving the fields fallow for large parts of the year (N 7 months) under
heavy rainfall allowed luxurious growth of weeds, and this might have
helped in partial recovery of soil properties through internal catalytic
action (Choudhury et al., 2015). Maiti (2007) also reported a recovery
of soil pH by 0.5 units in reclaimed mine spoils (only through planta-
tion) in the coal fields of Jharkhand, India.
This recovery of the soil's fertility was reflected in significantly great-
er production of shoots and roots in the pot experiment (p b 0.05) and
larger yields of grain in the fields. However, in absence of any positive
measures for restoration of land when mining ceased recovery was
slow, and after 4 years the yields of rice (2.13 t ha−1) were still consid-
erably less than from fields unaffected by mining (3.20 t ha−1). It seems
that natural regeneration on its own is likely to take much longer than
4 years for full recovery. If production is to be restored more quickly
then positive reclamation measures are likely to be needed.
5. Conclusions
We conclude that overburden from the coal mines in the Jaintia Hills
dumped on to paddy land and the acid drainage from it have acidified
the soil, contaminated it with heavy metals and left the soil depleted
in potassium and zinc. As a result, the soil thus affected is much less pro-
ductive for paddy rice than uncontaminated land. Rice yields from the
affected soil measured in our survey (average 1.21 t ha−1) were little
more one third than of those from unaffected land (average
3.20 t ha−1). The land does appear to recover its fertility after mining
ceases, even without special remedial treatment. Four years after min-
ing ceased, the pH had risen somewhat, the concentrations of exchange-
able Al3+ and the heavy metals Fe, Mn and Cu had diminished, and the
concentrations of potassium and zinc had increased. The rice yield had
increased on average to 2.13 t ha−1. However, the lost production of
rice as a result of mining and the slow recovery when it ceases is serious
in the regions of northeast India where too little food is produced to feed
their populations.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.scitotenv.2017.01.074.
References
Abreu Jr., M.T., Lavorante, A.F., 2003. Relationship between acidity and chemical proper-
ties of Brazilian soils. Sci. Agric. 60 (2), 337–343.
Anonymous, 2013. Department of Agriculture, Government of Meghalaya. http://www.
megagriculture.nic.in/public/download/CropStatistics.aspx.
Banerjee, S.K., Das, P.K., Mishra, T.K., 2000. Microbial and nutritional characteristics of coal
mine overburden spoils in relation to vegetation development. J. Indian Soc. Soil Sci.
48, 63–66.
Bitondo, D., Tabi, F.O., Kengmegne, S.S.A., Ngoucheme, M., MvondoZe, A.D., 2013. Micro-
nutrient concentrations and environmental concerns in an intensively cultivated
typic dystrandept in Mount Bambouto, Cameroon. J. Soil Sci. 3:283–288. http://dx.
doi.org/10.4236/ojss.2013.36033.
Bray, R., Kurtz, L.T., 1966. Determination of total, organic and available forms of phospho-
rus in soil. Soil Sci. 59, 39–45.
8 B.U. Choudhury et al. / Science of the Total Environment xxx (2017) xxx–xxx
Carmona, D.M., Cano, A.F., Arocena, J.M., 2009. Cadmium, copper, lead and zinc in second-
ary sulfate minerals in soils of mined areas in Southeast Spain. Geoderma 150,
50–177.
Choudhury, B.U., Fiyaz, A.R., Mohapatra, K.P., Ngachan, S.V., 2015. Impact of land use sys-
tems, agro-physical parameters and altitudinal variation in soil organic carbon con-
tent of Northeastern Himalayan Region of India. Land Degrad. Dev. 27:1163–1174.
http://dx.doi.org/10.1002/ldr.2338.
Conesa, H.M., Faz, A., Arnaldos, R., 2006. Heavy metal accumulation and tolerance in
plants from mine tailings of the semiarid Cartagena–La Unión mining district (SE
Spain). Sci. Total Environ. 366, 1–11.
Dowarah, J., Boruah, H.D.P., Gogoi, J., Pathak, N., Saikia, N., et al., 2009. Eco-restoration of a
high-sulfur coal mine overburden dumping site in northeast India: a case study.
J. Earth Syst. Sci. 118, 597–608.
Goswami, N.N., Rattan, R.K., Dev, G.N., Das, D.K., Sanyal, S.K., et al., 2009. Fundamentals of
Soil Science. Indian Society of Soil Science. Cambridge Printing Works, New Delhi.
Indian Minerals Yearbook, 2012. Ministry of Mines, Indian Bureau of Mines, Government
of India, New Delhi. Accessed at: Vol-III 4734 - http://ismenvis.nic.in/Database/
IndianMineralsYearbook.
Jackson, M.L., 1973. Soil Chemical Analysis. Prentice−Hall, New Delhi.
Jalota, S.K., Romesh, K., Ghuman, B.S., 1998. Method in Soil Physics. Narosa Publishing
House, New Delhi.
Katyal, J.C., Randhawa, N.S., 1983. Micronutrients. FAO Fertilizer and Plant Nutrition Bul-
letin 7: 1–82. Food and Agriculture Organization of the United Nations (ISBN:
9251014450).
Lin, C., Coleman, N.T., 1960. The measurement of exchangeable aluminum in soils and
clays. Soil Sci. Soc. Am. Proc. 24, 444–446.
Lindsay, W.L., Norvell, W.A., 1978. Development of DTPA soil test for Zn, Fe, Mn and Cu.
Soil Sci. Soc. Am. J. 42, 421–428.
Maiti, S.K., 2007. Bioremediation of coalmine overburden dumps with special emphasis in
micronutrients and heavy metals accumulation in tree species. Environ. Monit. As-
sess. 125, 111–122.
Manara, A., 2012. Plant responses to heavy metal toxicity. In: Furini, A. (Ed.), Plants and
Heavy Metals, Springer Briefs in Biometals. Springer, Dordrecht, pp. 27–44.
Martinez, R.E., Marquez, J.E., Hoa, H.T.B., Giere, R., 2013. Open-pit coal-mining effects on
rice paddy soil composition and metal bioavailability to Oryza sativa L. plants in
Cam Pha, northeastern Vietnam. Environ. Sci. Pollut. Res. 20, 7686–7698.
Please cite this article as: Choudhury, B.U., et al., Acid drainage from coal mining: Effect on paddy soil and productivity of rice, Sci Total Environ
(2017), http://dx.doi.org/10.1016/j.scitotenv.2017.01.074
Masto, R.E., Sheik, S., Nehru, G., Selvi, V.A., George, J., et al., 2015. Assessment of environ-
mental soil quality around Sonepur Bazari mine of Raniganj coalfield, India. Solid
Earth 6:433–811. http://dx.doi.org/10.5194/se-6-811-2015.
Monterroso, C., Alvarez, E., Fernandez, M.L., Macias, F., 1999. Geochemistry of aluminum
and iron mine soils from As Pontes, Galicia (N.W. Spain). Water Air Soil Pollut. 110,
81–102.
Nagajyoti, P.C., Lee, K.D., Sreekanth, T.V.M., 2010. Heavy metals, occurrence and toxicity
for plants: a review. Environ. Chem. Lett. 8, 199–216.
Nayak, B., 2013. Mineral matter and the nature of pyrite in some high-sulfur tertiary coals
of Meghalaya, Northeast India. J. Geol. Soc. India 81, 203–214.
Nikhil, K., Singh, A.K., Soni, A.K., Prasad, V.V.R., 2007. Effect of coal mining on acid soils: a
case study in Jaintia Hills of Meghalaya. In: Singh, A.K., Patra, S.C. (Eds.), Characteriza-
tion of Land Resources and Agro-Eco-Zones in India, pp. 112–114.
Payne, R.W., 2015. The Guide to GenStat Release 17 – Part 2: Statistics. VSN International,
Hemel Hempstead, UK.
Sahoo, P.K., Tripathy, S., Panigrahi, M.K., Equeenuddin, S.M., 2012. Mineralogy of Fe-pre-
cipitates and their role in metal retention from an acid mine drainage site in India.
Mine Water Environ. 31, 344–352.
Sahoo, P.K., Kim, K., Equeenuddin, S.M., Powell, M.A., 2013. Current approaches for miti-
gating acid mine drainage. Rev. Environ. Contam. Toxicol. 226, 1–32.
Sahrawat, K.L., 2005. Iron toxicity in wetland rice and the role of other nutrients. J. Plant
Nutr. 27, 1471–1504.
Sarma, K., 2005. Impact of Coal Mining on Vegetation: A Case Study in Jaintia Hills District
of Meghalaya, India. Dissertation. International Institute for Geo-information Science
and Earth Observation, Enschede, The Netherlands.
Sarma, K., Kushwaha, S.P.S., Singh, K.J., 2010. Impact of coal mining on plant diversity and
tree population structure in Jaintia Hills district of Meghalaya, North East India. New
York Sci. J. 3, 79–85.
Stephan, U.W., Schmidke, I., Pich, A., 1994. Phloem translocation of Fe, Cu, Mn, and Zn in
Ricinus seedlings in relation to the concentrations of nicotianamine, an endogenous
chelator of divalent metal ions, indifferent seedling parts. Plant Soil 165, 181–188.
Swer, S., Singh, O.P., 2003. Coal mining impacting water quality and aquatic biodiversity
in Jaintia Hills District of Meghalaya. ENVIS Bull. Himalaya Ecol. 11, 25–33.
Walkley, A., Black, I.A., 1934. An examination of the Degtjareff method for determining
soil organic matter and a proposed modification of the chromic acid titration method.
Soil Sci. 34, 29–38.