Sawahiko Shimada

The carbon content characteristics of tropical peats in Central Kalimantan, Indonesia: Estimating their spatial variability in density

Clarification of carbon content characteristics, on their spatial variability in density, of tropical peatlands is needed for more accurate estimates of the C pools and more detailed C cycle understandings. In this study, the C density characteristics of different peatland types and at various depths within tropical peats in Central Kalimantan were analyzed. The peatland types and the land cover types were classified by land system map and remotely sensed data of multi-temporal AVHRR composites (1-km pixel size), respectively. Differences in the mean values of volumetric C density (CDV) were found among peatland types owing to the variability in physical consolidation from peat decomposition or nutrient inputs, although no vertical trends of CDV were found. Using a step-wise regression technique, geographic variables and the categories of peatland type and land cover type were found to explain 54% of the variability of CDV within tropical peatlands in some conditions.

Carbon Stock Estimate

Tropical peatlands have changed their role from carbon sinks to carbon sources mainly by recent anthropogenic disturbances. It is an urgent issue to evaluate the importance of tropical peatlands as carbon stocks and to preserve the ecosystems including their carbon dynamics. Spatial distribution of carbon mass at a regional level needs to be delineated in order to utilize it in the simulation of carbon release impact of peat fires or in preservation planning strategy from a carbon dynamic perspective. In this chapter, a simple method to predict peat thickness was introduced. This pridiction method focuses on the differences in phenological characteristics due to the differences in hydroperiod and thickness of peat layer. Since the hydroperiod is a seasonal characteristic of peatlands in Southeast Asia, the phenology of the peat swamp forest was hypothesized to be a predictor of underlying peat thickness. Monthly NOAA-AVHRR data (Sep. 1992–Aug. 1993) were used to trace the fluctuation of vegetation activities among three seasonal periods. The peat swamp forests of Kalimantan was discovered to be classified into eight major phenology types and the classified map was found out be a good indicator to estimate the accumulated peat volume in peat swamp forests. According to our further estimation analysis, the carbon mass below the peat swamp forests (2.04 Mha) and the non-forest area (0.36 Mha) of Central Kalimantan peatlands were estimated to be 1.69 and 0.55 Gt C Mha−1, respectively. Extrapolating these values, we estimate that ca. 27 Gt C is stored within Indonesian peat (16.90 Mha) and 29.9–67.6 Gt C within Southeast Asian peat (19.7–41.5 Mha).

Tropical Peat Formation

A summary of the peat formation process, and the classification and characterization of the peatlands in Southeast Asia, particularly those in Kalimantan in Indonesia was undertaken through a review of published studies. Based upon the location, mode of formation, and age of the peat deposits, ombrotrophic and eutrophic peatlands, or topogenous and ombrogenous peatlands are developed by the accumulation of plant debris in coastal and sub-coastal areas, inland areas and high altitude areas. In the areas along the coastline, the youngest peat formation started to occur between 3500 and 6000 years BP in response to the wet conditions generated by rising sea levels at the end of the last glacial period. In comparison, peat in inland peatland areas began to form much earlier, more than 20,000 years BP during the late Pleistocene era. Some tropical peatlands are likely to have been involved in the global carbon cycle before the initiation of boreal and temperate peatlands. One of the characteristics of the peatlands in Southeast Asia is the formation of a convex-shaped dome that formed beyond the extent of river floodwater and under rain-dependent conditions. This is known as ombrogenous peat.

EVALUATION OF IMPACT OF EARTHQUAKE ON AGRICULTURE IN NEPAL BASED ON REMOTE SENSING

The big earthquake happening in April 2015 killed over 9000 people in Nepal. The effect of earthquake affected not only safety of local people but also agricultural field. Agricultural economy dominates in income of local people. Therefore, restoration of agricultural areas is required for improving life of locals. However, lack of information about agricultural areas is the main problem for local government to assess and restore damaged agricultural areas. Remote sensing was applied to access damaged agricultural field due to its advantages in observing responds of environment without temporal and spatial restriction. Accordingly, the objective of the study is to evaluate disaster risks based on data from questionnaire survey, remote sensing and geographic information system (GIS) in agricultural areas of Nepal. Firstly, we conducted questionnaire survey about thirty indicators of agriculture-related issues. Moreover, based on USLE (Universal Soil Loss Equation), soil erosion risk was compared between before and after the earthquake. To clarify the relation between soil erosion risk and land-use, land-use map was created based on Worldview-3. Finally, statistical analysis was conducted based on the collected data. From the results of field survey and analysis, it turned out that there was little damage on agricultural areas but huge damage on houses and barns in the villages in the research site. It is attributed to the vulnerable house materials. Soil erosion risk, that has been little observed in agricultural area, decreased in forest area and increased in residential area compared to the pre-earthquake time. From the statistical analysis, multi regression analysis was applied and age of house and elevation was computed as dominant factors of building damage in the research site. It is suggested that it is important to improve house materials in the villages and increase vegetation cover to prevent from further soil erosion in the research site.

A method to estimate suitable irrigation timing for afforestation in arid areas using changes in stem diameter and spectral reflectance

In general, shrinking and swelling of the stem take place in extensible tissues found mainly in a narrow ring outside the dead xylem vessels in woody stems. We propose that by monitoring the extension and contraction of these tissues we can design an appropriate water application system to minimize the amount of irrigation water required. Accordingly, we expect not only to prevent application of excess water but also control salinization. Design of plant irrigation scheduling by determining water requirements using soil and meteorological properties can be a limitation in arid environments where plant stress can develop very quickly due to high evaporative demands. We have observed hysteresis between shrinking and swelling of plant stem diameter with relative humidity, air temperature and sunshine radiation under dry soil moisture conditions and also recognized the retrogression under high dry moisture stress, (the stem has begun to swell with a decrease of solar radiation and vice versa). We introduced a method of Relative Stem Diameter (RSD), which was the ratio of the tree stem diameter at any particular time to the tree diameter at the beginning of the day, to evaluate when irrigation is required. RSD’s values began to decrease at pF 3 and it could be possible to delay irrigation up to pF 3.5 under drier conditions. However this RSD method is quite complicated and not easy for practitioners to use for irrigation scheduling. In this paper, we introduce a simple method to evaluate the appropriate timing for watering of trees in arid and semi-arid areas using spectral reflectance of leaf.