********************************************* s1-readme.txt Dataset of soil carbon stock in Cambodian forests, Version 1.0. J. Toriyama, 27 February, 2018 ********************************************* This readme file is prepared by J. Toriyama to help users of dataset of soil carbon stock in Cambodian forests. There are two csv files attached, gs2-site.csvh for site data and gs3-profile.csvh for soil profile data. ********************************************* Data Description ********************************************* s2-site.csv (3.5 kB) 1. Plot_N, number of the survey plot. 2. Region, name of the region in Cambodia. KKN, Koh Kong; KMP, Kampot; KPT, Kampong Thom; KPS, Kampong Speu; KRC, Kratie; MDK, Mondulkiri; PRV, Preah Vihear; PST, Pursat; RTK, Ratanakiri; SMR, Siem Reap. 3. Lat and Lon, latitude and longitude, respectively. 4. Forest, forest type at the time of the soil survey. E, evergreen forest; D, deciduous forest; R, rubber plantation; SH, shrubland; SW, inland swamp forest. 5. Category, soil sampling category. 6. Period, period of the soil survey. 7. SoilC_30, soil carbon stock (MgC ha^-1) at 0-30 cm depth. 8. N_horizon, number of horizons in each soil survey plot. 9. Max_depth, maximum depth of the soil profile (cm). 10. Remarks, other remarks and references for each soil survey plot. Ref1. This report. Ref2. Toriyama, J., Ohta, S., Araki, M., Kanzaki, M., Khorn, S., Pith, P., Lim, S. and Pol S. (2007) Acrisols and adjacent soils under four different forest types in central Cambodia. Pedologist 51(1): 35-49. Ref3. Toriyama, J., Ohta, S., Ohnuki, Y., Imaya, A., Ito, E., Kanzaki, M., Hirai, K., Araki, M., Kiyono, Y., and Chann, S. (2013b) Physicochemical properties and carbon storage of forest soils on Cambodian basalt: A preliminary study with a density fractionation approach. JARQ 47, 217-226. Ref4. Toriyama, J., Ohta, S., Ohnuki, Y., Araki, M., Kanzaki, M., Det, S., Lim, S., Pol, S. and Pith, P. (2010) Physicochemical characteristics of plinthic and non-plinthic soils in dry deciduous forests on the east bank of Mekong, Cambodia. Pedologist 54(1): 2-10. Ref5. Toriyama, J., Ohnuki, Y., Ohta, S., Kosugi, K., Kabeya, N., Nobuhiro, T., Shimizu, A., Tamai, K., Araki, M., Keth, N. and Chann, S. (2013a) Soil physicochemical properties and moisture dynamics of a large soil profile in a tropical monsoon forest. Geoderma, 197, 205-211. Ref6. Toriyama, J., Hak, M., Imaya, A., Hirai, K. and Kiyono, Y. (2015) Effects of forest type and environmental factors on the soil organic carbon pool and its density fractions in a seasonally dry tropical forest. For. Ecol. Manag. 335, 147-155. Ref 7. Toriyama, J., Hirai, K., Kiyono, Y., Chann, S., Kanzaki, M., Saito, H. and Takahashi, M. (2011) Carbon storage in rubber plantation of various stand ages in Cambodia. Kanto Shinrin Kenkyu 62: 203-206. *n.d.: no data available. s3-profile.csv (9.5 kB) 1. Serial_N, serial number of soil horizons. 2. Plot_N, serial number of the survey plot, consistent with Plot_N in s2-site.csv. 3. Horizon, type of soil horizon, i.e., A, E, B, and C, which are described in situ; if no horizon is defined, the sampling depth is described. 4. Depth_top and Depth_bot, top and bottom depths of the soil horizon (cm), respectively. 5. Thick, thickness of the soil horizon (cm). 6. C_conc, carbon concentration of fine soil (gC kg^-1). 7. BD_fine, bulk density of fine soil (Mg m^-3). *n.d., no data available. ********************************************* Methodology ********************************************* 1. Study site All of the published soil samples for the dataset were collected at less disturbed forest sites, except for sites in rubber plantations and shrubland. The 66 sites cover evergreen forests (n = 33), deciduous forests (n = 25), rubber plantations (n = 6), a shrubland (n = 1), and an inland swamp forest (n = 1) (see Forest in s2-site.csv). The study sites where deciduous tree species are exclusively distributed are classified as deciduous forest, and those containing evergreen tree species as evergreen forest. The soil surveys were conducted from March 2003 to November 2011 (Period in s2-site.csv). The positions of the study sites were recorded by GPS devices (e.g. 60CSx, Garmin, USA) and are denoted in the dataset using hddd.ddd degree and the WGS84 coordination system (Lat and Lon in s2-site.csv). 2. Soil sampling Soil was sampled using two approaches: Category I and Category II (Category in s2-site.csv). Category I obtained one replicate of a representative soil profile to a soil depth exceeding 30 cm. The location of the representative soil profile was determined after considering the typical topography and composition of tree species in the study area. The data in Category II are based on four replicates in the study plot, and for soil depths up to 30 cm (0-5, 5-15, and 15-30 cm). In both categories, two types of soil samples were collected for measuring the carbon concentration and bulk density, respectively. The soil sample for measuring carbon concentration was collected in a plastic bag, air-dried, and sieved to pass a 2-mm mesh. The soil sample used to determine bulk density was collected using three 100 cm^3 metal cylinders in each soil horizon. For the two deciduous forest sites in Ratanakiri (plots 6 and 7 in s2-site.csv) where gravel larger than the cylinder was observed, the soil bulk density was measured using cuboid-space sampling approach. 3. Chemical analysis The air-dried fine soil (< 2 mm) used for measuring the carbon concentration was oven-dried (105C for 24 h) and weighed to determine the soil moisture content. The carbon concentration of fine soil was determined using the dry combustion method (e.g. Sumigraph NC analyzer NC-22F; Sumika Chemical Analysis Service, Osaka, Japan). The fine soil in the metal cylinders was also dried at 105C for 24 h and weighed to measure the bulk density. The average carbon concentrations and bulk densities for four replicates are shown in the dataset for Category II. 4. Calculating the soil carbon stock The soil carbon stock per cm thickness (SoilC_cm, MgC ha^-1 cm^-1) in each soil horizon was calculated using variables in the soil profile data (s3-profile.csv) as follows, SoilC_cm = C_conc ~ BD_fine ~ 0.1, where C_conc is the carbon concentration of fine soil (gC kg^-1 (= kgC Mg^-1)), BD_fine is the bulk density of fine soil (Mg m^-3), and the coefficient 0.1 is the product of 100 (m^3 ha^-1 cm^-1) and 0.001 (Mg kg^-1). The soil carbon stock at a depth of 0-30 cm (SoilC_30 in (s2-site.csv) is calculated as the sum of SoilC_cm at that same depth. The soil carbon stocks of two sites (plots 4 and 11 in s2-site.csv) were not calculated due to a lack of bulk density data. ********************************************* Citation and Remarks ********************************************* When using the dataset, please cite this report as follows: Jumpei Toriyama, Akihiro Imaya and Keizo Hirai. (2018) Dataset of soil carbon stock in Cambodian forests, version 1.0. Bulletin of FFPRI, 446:175-186. Contact information is as follws: Jumpei Toriyama, Kyushu Research Center, Forestry and Forest Products Research Institute (FFPRI) E-mail: jtori@affrc.go.jp Neither the authors nor FFPRI bear any responsibility for the outcomes created by users of the soil carbon stock dataset.