Index Terms—Forest dieback, soil organic matter, lead,
In tropical mountains, it is common to encounter forests of
Manuscript received March 29, 2012; revised June 12, 2012.
University of Peradeniya, Sri Lanka (e-mail: firstname.lastname@example.org).
P. I. Yapa is with the Department of Export Agriculture, Faculty of
Agricultural Sciences, Sabaragamuwa University of Sri Lanka, Sri Lanka
S. P. Nissanka is with the Department of Crop Science, Faculty of
S. P. Perera is with the Department of Soils and Plant Nutrition, Rubber
Research Institute, Agalawatta, Sri Lanka.
to lowland forests. One of the best examples for a typical
montane cloud forest in the world is “Horton Plains”, Sri
Lanka, as it was in 1947, was described as a low, dense,
slow-growing forest with a healthy and vigorous appearance
. It is located on the highest plateau of Sri Lanka, which lies
between 1,500 and 2,524m average sea level  and the
geographical location is in the Central Highlands of the
Central Province, 6’47 – 6’50’N, 80’ 46’- 80’50’E. Annual
rainfall in the region is about 2540 mm. Temperatures are low,
with an annual mean of 13°C, and ground frost is common in
February . The landscape characteristically consists of
gently undulating highland plateau at the southern end of the
central mountain massif of Sri Lanka. Soil order Ultisol is
characterized by a thick, black, organic layer at the surface.
Horton Plains is considered to be the most important
catchment area of the country as it is the originating point of
the tributaries of three major rivers, the Mahaweli river
flowing to the north, the Keleni river to the west and Walawe
to the south of Sri Lanka. Belihul Oya, a small stream feeding
the Walawe, tumbles over a cliff as a large and spectacular
waterfall within the reserve itself. These forests remained
largely untouched by the 3000-year-old history of human
agricultural activity on the island and the hydraulic
civilizations that shaped the landscapes of the lowlands left a
comprehensive record that attests to this fact. Horton Plains is
rich in biodiversity and most of the fauna and flora within the
park are endemic while some of them are confined to
highlands of the island.
The land area covered by this montane rain forest is
approximately 3,160 ha. There are 54 woody species, of
which 27 (50%) are endemic to Sri Lanka. The area covered
grasslands are locally known as “patana”. The canopy of
commonly found cloud forest is dominated by the endemic
keena (Calophyllum walkeri) in association with varieties of
Myrtacea (Syzygium rotundifolium and S. sclerophyllum) and
relatives of domesticated plants, such as pepper, guava,
tobacco and cardamom.
Belonging to different size and age classes of these forest,
have been dying due to a yet unknown factor. This
phenomenon was first observed in the Horton Plains National
Park and the earliest reports of a significant level of dieback in
the forest were by . Estimations using recent satellite
images combined with ground surveys revealed that about
654 ha, equivalent to 24.5% of the forest in the park has been
subjected to dieback . One of the worst affected trees was
Soil Pollution and Forest Dieback: Will the Compost and
Mycorrhizal Treatments be Effective in Mitigating Forest
Gunadasa H. K. S. G., Yapa. P. I., Nissanka S. P., and Perera S. P.
International Journal of Chemical Engineering and Applications, Vol. 3, No. 3, June 2012
ovalifolium, Neolitsea fuscata, Syzygium revolutum and
regeneration in the area is slow . Healthy forest in the park
amounts to about 2012 ha. The extent of the damage to the
forest from dieback appears to be so severe that the stand
structure in affected areas shows dramatic changes. If this
dieback continues with the current rate, the majority of the
large trees will disappear from the forest soon. The vital
functions offered by this precious forest will then be subjected
to significant changes most probably towards the negative
side. Work done by many researchers so far has ended up with
no significant clues about the causal agents and remedial
measures for the dieback though work done by  has
indicated the contamination of soils in the Horton plains by Pb
and Cd and possible links of the soil pollution to forest
dieback. Therefore, the main objective of the study was to
assess the influence of SOM in remediating Pb and Cd
pollution in the affected soils. The specific objectives will
include how the different concentrations of Pb and Cd`in the
soil affect on the mortality of Syzygium rotundifolium
Fig. 1. A map of sri lanka locating the horton plain.
The location of the experiment was in Horton Plains, the
2,524m. Twenty-four permanent experimental plots of 20 m
20 m were demarcated using GPS (Global Positioning System)
the Horton Plain National Park. Randomized Complete Block
Design (RCBD) was used with six blocks to replicate each
and every treatment six times. Plot locations were selected to
cover a 61 – 80 % dieback of trees and to maintain soil and
topography as constant as possible. Canopy health was
assessed using a map published by . Four soil amendments
the study while taking Syzygium rotundifolium as the
indicator plant. The most important reason for the selection of
the tree species Syzygium rotundifolium was due to the fact
that of all species that have been affected, this specie was the
worst affected. The second reason is that it is one of the
dominant canopy tree types in the forest .
An Investigation of harmful elements such as Pb and Cd in
the soil samples were measured by wet ash method  and the
extractants were analyzed for the above elements by Atomic
Absorption Spectrophotometry . In addition, the soil
organic matter content was determined using the method of
total organic C by Walkley and Black described by . The
soil samples were collected from 0.20m depth and 0.3m-0.5m
away from each sapling representing three different time
periods. Furthermore, Death rates of the saplings were
calculated by keeping records of the selected saplings
throughout the experimental period and counting the deaths at
the end of the trial. Standard GENSTAT statistical software
was used for analysis of variance (ANOVA), t-test and
regression analysis of the results.
A comprehensive research done for two years within a
(HPNP), Sri Lanka was the base the following outcomes. Soil
organic matter content and heavy metals such as Pb and Cd
were compared first among the treatments under three stages
of sampling. In addition, the data collected were compared
with the death rate of the saplings as well.
The soil organic matter content in a soil expresses the
relationship between the sources of organic materials and the
decomposing factors (soil biota). Soil organic matter (SOM)
level in the study area of Horton Plains has not reached upper
levels in the range, up to 12%, as expected in tropical moist
evergreen forests . In ordinary tropical moist evergreen
forests, SOM content varies around 6% . Relatively low
plant nutrient levels in montane forests are not unusual
according to past studies (e.g., (.). For each 1000m rise in
altitude, there is a 7
C drop in temperature . This has a
ecosystem. With the elevation of about 2524m, Horton Plains
is cold (mean annual temperature 15
C) and contains a very
changes in the environment than normal tropical forests .
Under the prevailing conditions in the montane environment
–low sunlight, low temperature, shallow soil depth and so on,
production of SOM is weaker in the Horton Plains than in an
ordinary tropical forest . As far as the SOM content is
concerned, there are significant differences among the
treatments at soil sampling stage 1 (p = <.001), stage 2 (p =
soils treated with compost and compost + mycorrhizae
mixture showed the higher values of soil organic matter
though soils treated with mycorrhizae only and the control
showed the lowest at all three stages. Fluctuation of SOM
levels in the area may be linked with temperature, rainfall, soil
depth and addition of organic debris from the aggressively
growing undercover vegetation such as Strobilanthus spp.
The function of SOM springs from its effects on soil structural
stability (its action as a bonding agent between primary and
secondary mineral particles leads to enhanced amount, size
and stability of aggregates) and soil water retention (as a
water adsorbing agent, it enhances water acceptance and
same time, SOM controls soil nutrients that affect biomass.
 emphasized that soil structural stability is influenced by
the type of organic matter, as well as its amount. Therefore, in
some cases, high SOM content is not accompanied by high
structural stability.  pointed out that some fungi exude
oxalic acid, which enhances dispersion and breakdown of
aggregates. Humic substances are the components of SOM
which play the key role in detoxifying the soil from pollutants
such as Pb and Cd residues of Agro-chemicals from
surrounding areas . Unsatisfactory levels of SOM exhibit
the poor activity of humic substances and resultant soil
pollution. It should also be noted that even a milder form of
soil contamination in the Horton Plains cannot be afforded
since the montane vegetation is highly sensitive to the changes
in the environment.
Fig. 2. Status of SOM% among the treatments.
The level of soil Pb and Cd has gone up to 106 and 7.29
ppm respectively. The maximum allowable limit of Pb is 100
ppm while it is 3ppm for Cd . Even the smallest amount of
both Pb and Cd may impose severe damages on plant’s
metabolism leading to dieback . Results from soil
analysis clearly indicated contamination of soil from these
two trace elements in Horton Plains. Treatments used for the
study have significantly influenced the soil Pb at sampling
stages 1 (p=0.01) and 2 (p=0.004) but there is no significant
influence detected at stage-3 (p=0.79) (Fig 3) and the highest
Pb content was observed in the control. Cadmium content in
the soils of the study area is not significantly different with the
treatments at stage-1 (p=0.18), -2 (p=0.35), and -3 (p=0.51)
though the highest is observed in the control (Fig 4). A
fraction of those elements may leach out from the top soil
while another fraction may be absorbed by the vegetation.
Results from foliar analysis indicate the entry of Pb and Cd
into the plant bodies (see table 1).
Kandy, a major city, has been identified as the worst
polluted city in Sri Lanka with heavy motor traffic and
resultant vehicle emissions . Burning diesel, gasoline and
lubricants releases Pb and Cd to the atmosphere. Additionally,
the friction by brake pads, clutch liners and tires releases these
elements to the atmosphere. Strong monsoon winds seem to
be the most possible transportation source of Pb and Cd from
the polluted south western part of the country and following
pioneer studies, Pb and Cd are subjected to long-range
atmospheric transportation to a greater extent  and 
where, Pb can be transported for a distance greater than
120km  . Past studies reported that forest soils exceeding
1800m elevation were contaminated with higher levels of Pb
and the atmospheric origin of the excess soil Pb was
confirmed by high Pb levels in precipitation  and  .
Moreover, with increasing visitors to the Horton Plains,
motor traffic within the Horton Plains itself has increased.
Therefore, contamination of atmosphere may have been
increased to an alarming level so that it is very unlikely the
rain falling onto the area is free from Pb and Cd.
Mycorrhizae significantly increase the absorption of
various elements from the soil including heavy metals such as
Pb and Cd . Therefore, it could be assumed that
mycorrhizae are responsible for the reduction of Pb and Cd in
the soil treated with mycorrhizae. Soil microorganisms play a
vital role in maintaining overall soil quality. They have been
proved to be effective in detoxifying pollutants in the soil that
include heavy metals such as Pb and Cd. Soil microbes (e.g,
mycorrhizae) on the other hand, maintain extremely useful
symbiotic associations with the forest vegetations which
provide additional advantage for the plants to mine nutrients
and water .
However, high levels of heavy metals in soils have been
shown to decrease populations of soil microorganisms  .
Contribution of the microbes in humification process during
organic material decomposition should also be noted because
humic substances formed during the process play a very
special role in controlling the effects of organic and inorganic
pollutants in the soil . So, the deterioration of the
activities of soil microorganisms as a result of the acidity
conditions in the soils of Horton Plains may have placed the
forest vegetation in a vulnerable state for soil contaminants
like Pb and Cd. Acidic pH conditions also increase the
availability of micronutrients in the soil unnecessarily and this
situation results in the development of toxic conditions from
micronutrients on plants .
Fig. 3. Status of Cd among the treatments at three different stages of
Fig. 4. Status of Pb among treatments at three different stages of sampling.
It was clearly evident that the addition of standard compost
and mycorrhizae has significantly controlled the death of
Syzygium rotundifolium saplings (Fig 6). Treatment effect on
the death of saplings is significant (p=< 0.001) whilst the
control clearly shows the highest death rate (Fig 5). The
standard compost consists of humic and fulvic acids that are
formed during the microbial decomposition of organic
materials. These specific molecules, known as humic
substances, possess extraordinary capability of immobilizing
soil contaminants such as Pb and Cd. Additionally, dozens of
fractions in compost help the plants to withstand stressful
conditions such as drought, nutrient imbalances, acidity and
so on . In addition, standard compost is a good reservoir
of all forms of essential plant nutrients and growth factors of
plants . Mycorrhizae, on the other hand, act as a
remarkable symbiotic mechanism for the plants to survive
under stressful conditions such as droughts, nutrient
deficiency, soil contaminants such as Pb and Cd . Thus, it
could be argued that treating the Syzygium rotundifolium
samplings with standard compost and mycorrhizae until they
become grownup trees might help to fill the gaps caused by
the dieback in the forest.
Fig. 5. Death rate of the saplings after 2 years with four different treatments.
The relationship between Pb concentration and the death
rate of Syzygium rotundifolium saplings was significant (p =
<0.001) while the correlation showed the death rate of the
saplings has been largely affected by the Pb concentration in
the soil (Fig 7). Findings clearly exhibits that the death rate of
saplings used for the experiment increases with the
increasing availability of Pb in the soil. Results further
revealed that the critical level of Pb in relation to the survival
of Syzygium rotundifolium saplings was around 60ppm in the
Horton Plains soil and above this level, an abrupt increment of
death rate of the saplings could be observed. It means that
even a slightest increase of available Pb in the soil above the
threshold of 60ppm appears to impose severe damages on
plant’s metabolism leading to dieback .
Fig. 6. A dying sapling of syzygium rotundifolium.
Fig. 7. Pb concentrations in the soil Vs Death rate of saplings.
Death rate of the saplings (Syzygium rotundifolium) used
for the experiment appears to be increased with increasing Cd
availability in the soil. Some linear relationship between
increasing death rate of the saplings and the increment of
available soil Cd was observed though the correlation was not
significant under the
level 0.05 (p=0.08) (Fig 8). It has been
crucial metabolic functions of plants leading to death.
However, the nature of the Cd toxicity on Syzygium
rotundifolium is different from the nature of Pb toxicity on the
same plant. A threshold level of soil Cd in relation to the death
rate of Syzygium rotundifolium saplings cannot be observed
within the range of soil Cd found in the study area.
Fig. 8. Cd concentrations in the soil Vs Death rate of saplings.
Y= 24.14 - 0.001/ (1-0.02X)
P = < 0.001
An inversely proportional relationship between soil Pb and
the SOM content was observed (Fig 9) and the relation was
also statistically significant (p = <0.001). The findings clearly
indicate that the availability of Pb in the soil for the vegetation
in the study area could be reduced by increasing SOM level.
The nature of the decline of soil Pb with the increasing SOM
level seems to be a linear-by-linear type. It means that the
concentration of soil Pb decreases with the increasing SOM
level but the rate of decline of Pb gradually decreases.
Immobilization of soluble Pb in the soil by the humic and
fulvic acid molecules present in SOM has been documented
by several researchers (e.g.,  and ).
Fig. 9. Soil organic matter Vs Pb in soil.
Fig. 10. Soil organic matter Vs Cd in soil.
Soil Organic Matter Vs Cd in the Soil
The relationship between the availability of Cd in the soil
and SOM content was significant ( p = 0.01). The nature of
the relationship was a linear-by-linear as shown in fig. 10.
According to the graph, soil Cd levels gradually decreased
with the increasing SOM level. The results clearly show that
the effect of available soil Cd on montane vegetation could be
reduced by improving SOM level. Humic and fulvic acids are
also proven to be effective in immobilizing Cd as well .
The results in general indicate that the maintenance of SOM
will help to mitigate the Cd toxicity on forest vegetation.
Soil Organic Matter Content in the Soil and Dieback of
Results clearly show that the increase of SOM level helps
to reduce the death of saplings. The relationship between
SOM level and the death rate of the saplings (Syzygium
relationship seems to be linear-by-linear and it further
indicates that by maintaining SOM level somewhere above
4%, the death rate of the saplings could significantly be
reduced (see fig 11). Humic and fulvic acid molecules in
SOM effectively immobilize toxic metals such as Pb and Cd
in the soil .
Fig. 11. Soil organic matter content in the soil vs death rate of saplings.
Soil pollution in the montane forest with Pb and Cd as
affected by increasing vehicle emissions and consequential
polluted rain appears to be one of the key causes for the forest
dieback and resultant disappearance of the most precious
natural forests in Sri Lanka. Extra sensitivity of the montane
forest vegetation to the changes in the soil may have triggered
the impact of soil pollution. Enrichment of the polluted forest
soils with standard compost and montane mycorrhizae
appears to be effective in saving the saplings of Syzygium
death but, the compost treatment may be applicable under
emergence situations and could therefore be considered as a
As a long-term solution, reestablishment of the deteriorated
areas of forest with native flora with a creation of identical
multilayered tropical moist evergreen forest architecture and
a climax of biodiversity will be suggested. The suggested
design will enrich the soil with good quality organic matter
which is recognized as the basis of the success of tropical
moist evergreen forests.
Maintenance of good quality SOM at satisfactory levels in
the soil appears to be effective in reducing the toxicity levels
of both Pb and Cd in the soil. Therefore, enriching the status
of SOM will maintain the continuity of natural vegetation of
the forest flora by minimizing the death rates of the plant
species. Threshold levels of Pb should be taken into a specific
consideration during forest management. Every possible
measures should be taken to not to allow the soil Pb level to
exceed the threshold level of 60ppm. It would also be
important to maintain the SOM level in the forest soil above
4% so that severity of the effect of soil pollution on forest
flora could be avoided.
This study was conducted with the financial support of
Sabaragamuwa University of Sri Lanka and the Department
of Wildlife Conservation. We are also grateful to the Park
Warden and the rest of the staff at Horton Plains National
Park for their support given throughout the study. Our very
special appreciation should go to the Rubber Research
sophisticated laboratory analysis related to the research.
T. W. Hoffmann, “The Horton Plains, Good and Bad news,” Loris, vol.
18, no. 1, pp. 4-5, 1988.
T. C. Whitmore, Tropical Rain Forests of the Far East, Claredon Press,
R. A. Wijewansa, “Horton Plains: a plea for preservation,” Loris, vol.
16, pp. 188-191, 1983.
W. L. Werner, “The Upper Montane forests of Sri Lanka,” The Sri
N. K. B. Adikaram, K. B. Ranawana, and A. Weerasuriya, “Forest
dieback in the Horton Plains National Park, Sri Lanka Protected Areas
Management and Wildlife Conservation Project, Department of Wild
Life Conservation,” Ministry of Environment and Natural Resourses,
H. K. S. G. Gunadasa, P. I. Yapa, S. P. Nissanka, and S. P. Perera,
example from Sri Lanka,” International conference on environmental
USEPA, Method 3050B, acid digestion of sediments, sludges and soils,
E. Dale and H. Norman, “Atomic absorption and flame emission
spectrometry,” in Methods of Soil Analysis, 2
ed. vol. 9, A.L. Page, R.
D. W. Nelson and L. E. Sommers, “Total carbon, organic carbon and
organic matter,” in method of soil analysis, 2
ed, USA, American
Society of Agronomy, vol. 9, 1982.
P. L. Weaver, E. Medina, D. Pool, K. Dugger, J. Gonzales, and E.
Cuevas, “Ecological observations in the dwarf cloud forest of the
Luquillo Mountains in Puerto Rico,” Biotropica, vol. 10, pp. 278-291,
D. M. Dombois, P. M. Vitousek, and K. W. Bridges, “Canopy dieback
and ecosystem processes in Pacific forests: a progress report and
research proposal,” Hawaii Bot Sci, vol. 44, pp. 100, 1984.
A. Kaplan, M. A. Cane, Y. Kushnir, A. C. Clement, and M. B.
temperature,” J Geophys Res-Oceans, vol. 103-C9, pp. 18567-18589,
L. L. Loope and T. W. Giambelluca, “Vulnerability of Island Tropical
Montane Cloud Forests to Climate Change, with Special Reference to
East Maui, Hawaii,” Climatic Change, vol. 39, pp. 503-517, 1998.
P. J. Edwards and P. J. Grubb, “Studies of mineral cycling in a
matter in the vegetationand soil,” J Ecol, vol. 65, pp. 943-969, 1977.
P. Dutartre, F. Bartoli, F. Andreux, J. M. Portal, and A. Ange,
some sandy soils from West Africa,” Geoderma, vol. 56, pp. 459-478,
R. P. Voroney, J. A. V. Veen, and E. A. Paul, “Organic carbon
dynamics in grassland soils. II. Model validation and simulation of the
long-term effects of cultivation and rainfall erosion,” Can J Soil Sci,
vol. 61, pp. 211-224, 1981.
A. Kloke, “Orientierungsdaten für tolerierbare gesamtgehalte einiger
elemente in kulturboden mitt,” VDLUFA, vol. H.1-3, pp. 9-11, 1980.
A. B. Pahlsson, “Toxicity of heavy metals (Zn, Cu, Cd, Pb) to Vascular
Plants,” Water Air Soil Poll, vol. 47, pp. 287-319, 1989.
O. A. Illeperuma, “Kandy air most polluted,” Da i ly News, Tu esd a y,
2 0 1 0 .
E. Steinnes, J. P. Rambaek, and J. E. Hanssen, “Large scale
multi-element survey of atmospheric deposition using naturally
growing moss as biomonitor,” Chemosphere, vol. 35, pp.735-752,
T. Berg, O. Røyset, E. Steinnes, and M. Vadset, “Atmospheric trace
element deposition: Principal component analysis of ICP-MS data
from moss samples,” Environment Pollution, vol. 88, pp. 67-77, 1995.
M. F. Billett, E. A. Fitzpatrick, and M. S. Cresser, “Long term changes
– East Scotland,” Water, Air and Soil Pollution, vol. 59, pp. 179-191.
T. G. Siccama and W. H. Smith, “Lead accumulation in a northern
hard wood forest,” Environmental science and Technology, vol. 14, pp
T. G. Siccama, W. H. Smith, and D. L. Mader, “Changes in lead, zinc
white pine stands in central Massachusetts over 16 years,”
M. J. Harrison, “The arbuscular mycorrhizal symbiosis: An
underground association,” Trends in Plant Sci, vol. 2, pp. 54-59, 1997.
 A.J. Friedland, R. Gregory, K. Kilrenlampi, and A. H. Johnson,
“Zinc, Cu, Ni and Cd in the forest floor in the northeastern United
States,” Water, Air and Soil Pollution, vol. 29, pp. 233-243, 1986.
W. R. Jackson, “Humic, Fulvic, and Microbial Balance: Organic Soil
F. G. V. Jr, “Micronutrient Availability, Chemistry and Availability of
Micronutrients in Soils,” Journal of Agricultural and Food Chemistry,
vol.10-3, pp. 174-178, 1962.
I. Weissenhorn, C. Leyval, and J. Berthelin, “Bioavailability of heavy
atmospheric deposition from a smelter,” Biol Fert Soils, vol. 19, pp.
W. W. Wenzel, D. C. Adriano, D. E. Salt and R. Smith,
“Phytoremediation: a plant-micro based system,” in: Remediation of
contaminated soils, D.C. Adriano, J.M. Bollag, W.T. Frankenberger Jr,
R. C. Sims, Eds. SSSA Spec Monogr 37,1999, pp. 457-510.