Effects of Converting Secondary Forest to Oil Palm Plantation on Peat Soil Carbon and Nitrogen and other Soil Chemical Properties

Problem statement: Peatlands are natural sequesters of carbon and nitr ogen. Once they are disturbed the tendency to lose carbon and nitrogen to the environment is very high. This study investigated the effect of converting peat land for est into oil palm plantation on soil chemical properties with particular emphasis on carbon and n itrogen storage. Approach: Soil samples were collected randomly at depths 0-25 and 25-50 cm from a secondary forest and from four different ages of oil palm plantations at woodman oil palm plantat ion located in Sarawak, Malaysia. Soil pH in water and KCl, Organic Matter (OM), Organic Carbon (OC), Total Nitrogen (TN), Organic Nitrogen (ON), ammonium, nitrate, available phosphorous, carbon to itr gen ratio, carbon to phosphorous ratio and bulk density were determined using standard procedu res. The bulk density method was used to quantify Carbon (C), Nitrogen (N), ammonium, nitrat e and available phosphorous storage on per hectare basis. Results: Statistical analysis showed that the OC content wa s st tistically similar for all soil depths and vegetation types (forest or plantation). The TN content was statistically higher for second ary forest. Conclusion: Regardless of depth, C sequestration was not alter ed due to land use change but the secondary forest had higher stores of soil N.


INTRODUCTION
Peat is defined as the accumulation of purely one hundred percent organic matter and the distinction between soil and vegetative accumulation is not clear [1] . The soil division of Sarawak, Malaysia adopted a more recent definition which is based on soil partition, i.e. soils that have 50 cm or more Soil Organic Matter (SOM) within 100 cm or more than twice that of mineral soil materials overlying bedrock within 50 cm [2] . Peat soils form generally in large basin swamps which are dome-shaped. Organic soil deposits occupying the central portion of the dome, generally known as 'ombrogenous peat', comprises mainly of disaggregated tree trunks, branches, leaves, roots and fruits.
Occurrence of peat soil around the globe is about 423.7 million ha. In Asia, there are about 24.9 million ha of peat soil and in Malaysia about 2.5 million ha of peat soil where the estimated percentage of SOM in tropical forest is 100 [11] . The occurrence of peat soil in Sarawak, Malaysia is estimated to be about 1.8 million ha which is the largest hectarage for any of the Malaysian states [3] .
Organic carbon and TN contents in peat soil are very high with the OC contents ranging from 12-60% [1] . The surface of peat soils contain more C compared to the subsoil with values such as 58% at the surface and 25% in the subsoil [4] and typical of deep peat soils. Organic nitrogen in shallow peat soils can range from 0.5-2.1% and in the deep peat soils the values can range from 1.1-1.7% [5] with N levels in the deep peat being higher compared to shallow peat soils [1] .
High contents of C and N as well as other nutrients in peat soil have led to large hectares of peat land being developed for agricultural uses. Water holding capacity, cation exchange capability (CEC) and high moisture content in peat soils make them ideal for agriculture. Therefore, it is no surprise that in Sarawak, Malaysia around 554.8 thousand ha of peat land has been utilized out of the total 1.8 million ha of peat land [3] . The objective of this study was to compare the TN, OC storage and other soil chemical properties of a secondary forest and young oil palm plantations that were previously secondary forest.

MATERIALS AND METHODS
The soil samples used in this study were taken from a secondary forest and Woodman oil palm plantation Kuala Tatau, Sarawak, Malaysia. The size of each experimental plot for the two locations was 30×40 m. The ages of the oil palm plantations available sampled were one, three, four and five years old. Ten soil samples were taken using a peat soil auger at 0-25 and 25-50 cm depths. The samples were air dried and then sieved to pass 2 mm sieve.
The coring method was used to determine the bulk density at these depths. The determination of soil pH was done by using a ratio of 1:2.5 soil to distilled water suspension (pH in water) and 1 M of KCl (pH in 1 M potassium chloride) using a glass electrode [7] . Ten grams of soil samples were weighed and placed in plastic vials to which 25 mL of distilled water was added, shaken and allowed to settle for 24 h. The samples were again gently swirled and the pH determined using a glass electrode. The same was done using 25 mL of 1 N KCl but the samples were allowed to equilibrate for 10 min before the pH measurement with a glass electrode.
SOM was determined by the wet oxidation using sulfuric acid and digestion catalyst (K 2 SO 4 , CuSO 4 and SeO: ratio 100:10:1). Soil TN was determined by the kjeldahl method [7] .
To determine the inorganic nitrogen content (ammonium and nitrate N) in the peat, 0.5 g of soil sample was weighed and placed into a conical flask. A 50 mL of 2 M of KCl solution was added in the conical flask and shaken for 1 h using a mechanical shaker. The mixture was filtered and the filtrate used for subsequent ammonium measurement through distillation and titration. Twenty mL of the filtrate was pippeted into a distillation flask and 0.2 g of MgO was added, followed by distillation. The gas produced was collected in a conical flask containing 10 mL of Boric acid + indicator solution. The resulting solution containing the gas was titrated using 0.01 M HCl solution. Nitrate content was determined using the same sample for ammonium determination. Exactly 0.2 g of Devarda alloy was added to the same distillation flask and the distillation process carried out. The gas produced was collected in a conical flask with 10 mL of Boric acid + indicator solution. The resulting solution containing the gas was titrated was titrated using 0.01 M HCl solution.
TOC and TON per hectare basis was then calculated using the bulk density method [9] . The ratio of percentage of the TOC and TON was used to determine the C/N ratio of the soil.
Available P in the peat soil was measured by the colorimetric determination of P [11,12] .
Data was analyzed using the Statistical Analysis System (SAS) version 9.1 by means of ANOVA and means separation was done using the Tukey test. T-test was used to compare results at 0-25 and 25-50 cm soil depths.

RESULTS
The soil pH of secondary forest and four year old oil palm plantation showed no significant differences regardless of depth ( Table 1). The soil pH of the secondary forest and 1, 3 and 5 year old oil palm plantation at 0-25 cm were not significantly different but were so for 25-50 cm. The soil pH at 0-25 cm for the 5 year old oil palm plantation was significantly higher than those of 1, 3, 4 and secondary forest ( Table 2) which was not the case for 25-50 cm.
Vegetation type (oil palm or forest), oil palm age and soil depth had no significant effect on soil bulk density (Table 3 and 4). However, the bulk density values were typical of tropical peat bulk density.
Organic matter content decreased significantly with soil depth between secondary forest and oil palms aged 4 but no significant differences between oil palm ages 1, 3 and 5 ( Table 5). Depth of sampling did not show any significant effect on soil organic matter content (Table 6).    There were significant differences in soil carbon contents for the two depths for secondary forest, 4 and 5 year old oil palm plantation but not for the 1 and 3 year old plantations. However, the amount of carbon per hectare was neither affected by depth, vegetation type and age of oil palm plantation (Table 7 and 8).   In general, soil depth did not affect the amount of N and the quantity of N per hectare (Table 9). However, these variables were significantly higher for the secondary forest than the oil palm ages of 1, 3, 4 and 5 (Table 10).    (Table 11) for 1, 3 and 4 year old plantations but there was significant increase with increasing depth for secondary forest and 5 year old oil palm plantation. However,  Comparing the ammonium N content and the amount per hectare for secondary forest for the four different ages of oil palm plantation indicate that 1, 3, 4 and 5 year plantations had significantly higher quantities regardless of depth (Table 12). A similar observation was made for nitrate N (Table 13).
Available P decreased with increasing depth (Table 14) and the amount per hectare for the four oil palm plantations had significantly higher amount compared to the secondary forest at 25-50c cm depth (Table 15).    The C/N ratio was not statistically different regardless of depth (Table 16) but it was generally lower for the secondary forest compared with the different oil palm ages especially for 0-25 cm soil depth (Table 17).

DISCUSSION
The pH values of this study were typical of tropical peat soils [3] . This may be attributed to water table management which is common in utilization of peat land soil. Lower water table causes microbes to decompose peat under aerobic conditions compared to anaerobic conditions in natural and undisturbed peat.
Previous studies by Andriesse [1] , Driessen and Suhardjo [12] and Tie and Kueh [13] reported bulk density values of 0.2 g cm −3 and lower. In this study, the bulk density was 0.3 g cm −3 which may be the result of mechanization and management practices done by the plantations. Compaction of the peat before planting is a common practice by plantations. For instance, heavy duty machinery is used during land clearing and drainage construction. Frequent machine traffic during fertilization and harvesting also compact the peat.
Percentage OM content in peat showed no significant differences within the sampling depth and among the ages of oil palm plantations. Tie [13] stated that organic soil must at least contain 65% of organic matter. The observations in this study showed that organic matter content is as high as 80%.
There was no significant difference in percentage organic carbon content between the secondary forest and the oil palm plantations for both sampling depths. However, for between depth comparisons, the secondary forest, four and five year old oil palm plantations showed significant differences. The values ranged from 40-50%, this is consequent with the results obtained by Andriesse [1] and Kanapathy [4] . There were no significant differences within depths and among locations of sampling for secondary forest and ages of oil palm plantations for organic carbon content in peat soil by weight. This is comparable to the values obtained for percentage organic carbon content.
The total N contents were consistent with those reported by Tie [13] and higher total nitrogen content in secondary forest peat soil could be related to the genesis or the origin of the parent materials and the water management. For the oil palm plantation was much drained compared to the secondary forest. One, three, four and five year old oil palm plantations were significantly lower than the secondary forest because of initial land clearance, water table management, fertilizer application and liming. These practices tend to increase peat soil pH and boost the microbial activity which would degrade or mineralize the nitrogen in peat.
Percentage ammonium content and weight per hectare was lowest in secondary forest for both 0-25 and 25-50 cm depths. This shows that inorganic form of nitrogen content in secondary forest is very low and does not affect the total nitrogen content much. The higher content of ammonium for the different ages of oil palm plantations could be due to N fertilizer inputs.
The insignificant differences in nitrate regardless of soil depth, forest type and oil palm plantation age could be due to leaching which is known to be high in tropical peat soils.
Available P content in both topsoil and subsoil was significantly higher compared to different ages of oil palm plantation. This is because of artificial low water table (managed) in oil palm plantation, which causes leaching during high rainfall that is prevalent in the tropics such as Malaysia.
The secondary forest showed lower mean C/N ratio due to higher water table level which causes a condition wherein microbes need longer time to decompose the carbon and nitrogen. The different ages of oil palm plantation had higher C/N ratio due to breakdown of nitrogen by microbes.

CONCLUSION
Regardless of depth, carbon storage was statistically similar to forest type and different ages of oil palm plantation, but that of nitrogen was significantly higher in the secondary forest. Studies on plantations much older than five years (for example 15-20 years and over) is worth doing as this could result in more clear differences in results emanating from the prolonged land use change and management.

ACKNOWLEDGEMENT
The researchers acknowledge the financial support and other services of University Putra Malaysia.