Soil organic carbon variability under montane ecosystems: assessing the influence of landscape attributes

Abstract

Soil organic carbon (SOC) is integral to soil functioning and serves as a common indicator for soil security, water security, food security, energy security, climate change abatement, biodiversity protection, and ecosystem service. Since SOC varies in space and time, its quantification is essential to better understand the carbon (C) dynamics at various scales. As a gamut of montane ecosystems, the Himalayan region serves as one of the major C storage pools in the world due to its rich biodiversity and good forest cover. However, much of the Himalayan region is environmentally vulnerable because of rapid socio-economic development taking place, unsustainable land management, and land degradation. As a result, C emissions are steadily increasing in the region and, thus, contributing to global climate change. Despite this, there are still substantial gaps in terms of data and knowledge on SOC in the Himalayas, as in many other part of the globe. Therefore, this thesis was aimed to: (i) investigate the spatial distribution of SOC stock and its variation under different land use and land cover (LULC) types; (ii) examine the vertical distribution of SOC density in relations to LULC types, altitudinal zones, and aspect directions; (iii) determine the complex interrelationships of SOC fractions with LULC types and landform attributes; and (iv) assess the impacts of LULC type on aggregate stability, and hence aggregate size distribution and aggregate-associated organic carbon (AAOC). This study focused on a sub-catchment of Bhutan in the eastern Himalayas. First the SOC stock and its spatial distribution under different LULC types was determined using digital soil mapping technique of regression kriging (RK). Among the several environmental covariates used in RK are altitude, LULC type, slope gradient, and aspect directions. The results show that the mean top one meter SOC stock (mapped to 90 × 90 m grid resolution) decreased in the order of fir > mixed conifer > shrubland > grassland > broadleaf > blue pine > dry land (rain-fed agriculture) > orchard > paddy land. The overall SOC stock for the study area was estimated at 27.1 Mt (average 24.9 kg m-2), which is very high given the relatively small spatial extent of the study area. This indicates that the Himalayan region is characterized by high SOC storage capacity, which needs to be protected to remain as one of the major C sinks in the world. Second, in order to gauge the profile homogeneity of SOC density, its vertical profiles were each fitted with a smooth spline function followed by the estimation of the proportions of SOC density in top 0.20 m relative to the total density in the top one meter. The SOC density homogeneity values under different LULC types, altitudinal zones, and aspect directions were then compared. It revealed that the homogeneity under agricultural land was estimated at 34% compared to 38% for forest, 43% for shrubland, and 59% under grassland. In the case of influence of altitudinal zones on the SOC density, the latter was more uniformly distributed in the 3500-4000 m zone with homogeneity of about 35%, significantly less than 41% estimated for each of the 1769-2500 m and 2500-3000 m zones, and even lesser than 43% for the 3000-3500 m zone. Conversely, the homogeneity of SOC density estimated for different aspect directions indicated that the northern aspect exhibited higher vertical homogeneity compared to other aspect directions whose homogeneity values were indeterminate. These results suggest that vegetation and soil moisture, as influenced by LULC type, altitude, and aspect direction, are the main factors controlling the vertical distribution of SOC density. These outcomes have contributed significantly to our understanding of the conditions favorable to most vigorous vegetative growth and litter production leading to deeper incorporation of organic matter into the soil. The third component of this study was focused on explaining the complex interrelationships of SOC pools with environmental controlling factors of climate, landscape, and anthropogenic. It was hypothesized that the various SOC fractions/pools would respond differently to the environmental factors controlling the physical, chemical, and biological processes. Therefore, the multivariate statistical techniques of canonical correspondence analysis (CCA) was applied to reveal the subtle interrelationships of both the particulate organic carbon (POC) and humic organic carbon (HOC) with the different LULC types and landform attributes as the environmental covariates. It revealed that POC exhibited higher variation than HOC under different LULC types and landform attributes probably due to faster turnover rates of POC compared with that of HOC. Unlike HOC, the POC was largely concentrated in the upper depths of the soil profiles and hence more susceptible to soil erosion and rapid mineralization. The POC and HOC ratio became larger with depth suggesting that SOC becomes more stable with depth. This study also revealed that the influence of the natural LULC types on POC and HOC was higher than was the impacts due to agriculture. The findings have contributed to better understanding of POC and HOC in relation to various environmental factors in maintaining an optimum level of POC and HOC for enhancing sustainable agriculture, environmental quality, and ecosystem services. Lastly, the thesis explored how the vulnerability of the Himalayan region to land degradation can potentially be addressed by studying the impacts of LULC types on aggregate size distribution and stability as influenced by aggregate-associated organic carbon (AAOC). This is aimed to inform appropriate future soil and water conservation measures to combat land degradation. The results of aggregate size distribution show that the large macroaggregates (> 2 mm) accounted for 86-93% of the total aggregates under all LULC types except under dry land (64%) and paddy land (35%). The aggregate stability under different LULC types decreased in the order of fir > shrubland > grassland > orchard > blue pine > broadleaf > mixed conifer > dry land > paddy land. As expected, the AAOC in all aggregate fractions was much higher in the non-agricultural soils than under different agricultural lands. The AAOC in the large macroaggregates constituted more than 76% of the total AAOC across all LULC types except under dry land (65%) and paddy land (38%). Furthermore, the AAOC of the large macroaggregates depicted a similar trend with aggregate stability under different LULC types indicating its significant role in aggregate stability compared to AAOC of other aggregate fractions. The quadratic correlation between aggregate stability and AAOC of the large macroaggregates suggests an upper threshold for SOC to further improve soil aggregation and aggregate stability. In conclusion, this thesis has revealed the spatial/vertical distribution of SOC stocks as impacted by the different controlling factors. The thesis examined the complex interrelations of SOC fractions with LULC and landform attributes to help maintain an optimum level of all functional SOC fractions to improve soil quality. Furthermore, it investigated the microscopic/macroscopic aspects of soil aggregate stability in relation to the AAOC and how this can impact on potential soil degradation in the study region. This is the first systematic study on SOC undertaken in Bhutan which is envisaged to provide the springboard for further investigations across the Himalaya region. The findings of this research are anticipated to contribute in formulating appropriate sustainable land management and C sequestration strategies (e.g. conservation agriculture, agro-forestry, change in land use policy to safeguard and conserve the current forest cover) to combat C emission, soil erosion, and other forms of land degradation, specifically in Bhutan and generally in the Himalayan region. In addition, it is expected to contribute towards land use policy reform agenda to conserve the natural vegetation to ensure continued ecosystem service delivery in the region.