Variation in soil organic carbon stock in different land uses and altitudes in Bagrot Valley, Northern Karakoram

ABSTRACT

Soil organic carbon tends to respond more sensitively to climate change and land use intensification in ecologically fragile and economically marginal regions of mountainous areas. This study aims to evaluate the soil organic carbon stock dynamic across various land uses at different altitudes in the Bagrot valley, Northern Karakoram, Gilgit-Baltistan, Pakistan. Soil samples from 0–20, 20–40 and 40–60 cm depth were collected from three land uses: pasture, forest, and adjacently located arable land at different altitude (ranging from 2100–4163 m). The variables investigated were soil bulk density (BD), soil organic carbon concentration (SOC), soil organic carbon stock (SOCS) and pH. A significant variation in all tested variables were found across the land uses and altitudes. Likewise, soil under forest had significantly higher values of SOCS (59.35 Mg ha⁻¹) than pasture (42.48 Mg ha⁻¹) and arable land (23.63 Mg ha⁻¹). Similarly, SOCS increased with increasing altitude and decreased with soil depth in all land uses. In addition, SOCS had a negative relationship with BD and pH. Overall results indicated that the land use intensification and climate change (increase in temperature and decrease in precipitation) were associated with declining SOCS. These results suggest restoration of degraded agricultural land to the forest, especially at higher altitude, and decrease in intensity of land use could increase SOCS in the study area as well as other similar mountainous regions.

KEYWORDS:

• Bagrot valley
• climate change
• Gilgit-Baltistan
• mixed mountain agriculture
• vegetation density

Introduction

Soil organic carbon (SOC) plays an important role in food security and climate change mitigation (Lal 2004). The soil is the largest and a significant pool of carbon in the terrestrial ecosystem (Schlesinger 1997) and a significant constituent of global C cycle. Two-thirds of SOC is stored in soil (Schlesinger 1997), which is three times greater than that of plants and two times that in the atmosphere (Wang et al. 2011). SOC improve soil properties (Mulongoy & Merckx 1993) by holding a great amount of important soil nutrients for plant growth (Leu 2007) and providing energy to the microbial community (Mulongoy & Merckx 1993).

SOC can be influenced by various factors such as altitude and climate (Djukic et al. 2010), soil parent material and texture (Hoffmann et al. 2009), slope- aspect (Schwanghart & Jarmer 2011; Lozano-García et al. 2016), soil pH (Falloon & Smith 2009), vegetation (Luyssaert et al. 2008), and land uses (Morgan et al. 2010). Several studies have shown significant variation in SOC with relation to land use (Lemenih & Itanna 2004; Shufang et al. 2012; Iqbal et al. 2014; Mauryaa et al. 2014; Meng et al. 2014) and it decreases with the increase in soil depth (Yang et al. 2010). The type of land use is a vital element controlling SOC stock in soil; the land use affects the amount of input and the rate of decomposition of litter in soil (Lehmann et al. 2000). Likewise, SOC stock was found to be decreased by cultivation (Chen et al. 2010; Vikas et al. 2014).

Knowledge of vertical patterns of SOC can improve our understanding of the dynamics of SOC along with a profile and the potential response of SOC to climate change; a small variation in climate at higher altitude will drastically change the ecology and function of the mountain ecosystem (Jodha 1992). Similarly, altitude has a strong effect on the micro-climate and landscape (Bendix et al. 2008; Gerold et al. 2008). With increasing altitude, normally temperature becomes cooler and precipitation increases, resulting in vegetation distribution change, which impacts SOC stock (Tanner et al. 1998; Proctor et al. 2007). Various researchers have reported a positive relationship between SOC and altitude (Woldeamlak & Stroosnijder 2003; Genxu et al. 2004; Emiru & Gebrekidan 2013) although not always (e.g. Sheikh et al. 2009; Jahangeer et al. 2013). Similarly, SOC stock has been observed to increase with an increase in precipitation and decrease with an increase in temperature (Schlesinger 1997; Kidanemariam et al. 2012). However, SOC stock depends on C inputs from plant biomass production and outputs through decomposition (Schlesinger 1997).

Considerable studies were undertaken on effects of land use and altitude (climate change) on SOC stock in stable core areas. However, the researchers have given less attention to ecologically highly vulnerable and economically extremely marginal regions: these regions tend to respond more sensitively to changes in their framework conditions such as a change in land cover, land uses and climate change as compared to the stable core areas of human activities (Ehlers 2005). The natural environment of Karakoram in Gilgit-Baltistan is that of high-altitude limits of human land use and heavily glaciated milieu with an extremely steep and diversified topography; it is also a distinct vegetation zone, with cold and snow-rich winters (Ehlers 1995, 2005). Human occupation confined to the narrow river terraces and valley bottom allowing irrigated agriculture and mountain slopes with humid forest and high altitude pastures permitting seasonal grazing. Heavy rains, earthquakes, avalanches and other natural phenomena are continuous, which causes natural disasters such as landslides, mudflow, rock fall etc., sometimes causing severe damage to land and life (Hewitt 1988; Kreutzmann 1994). In general, due to the topographical setting and harsh climatic situation of the mountain areas, there are limited natural resources in this region (Ehlers 1995).

However, increasing population growth and construction of the Karakoram Highway (KKH) in 1970 has drastically increased the interference of man in these ecologically highly vulnerable and economically extremely marginal areas of the Karakoram; it involves over-exploitation of natural resources such as expansion and intensification of agriculture activities, overgrazing and severe deforestation (Ehlers 1995, 2005). These activities enhance the processes of degradation, soil erosion, slope destabilization, rockfalls, and rocks slides, which already common in the Karakoram mountains (Kreutzmann 1994; Ehlers 1995). Furthermore, these processes causing environmental degradation and local deterioration of the natural environment and soil health (Sitaula et al. 2004; Ehlers 2005). The possible changes that may occur increase the emission of greenhouse gases (Post & Kwon 2000), decrease SOC stock (Fenton et al. 1999) and other soil parameters (Kay 1995; Sahani & Behera 2001; Abbasi et al. 2010; Abbasi et al. 2010).

The objective of this study is to determine the effects of different land uses and altitudes (in terms of climate change) on SOC dynamics in the ecological and economically marginal land in the Bagrot Valley of Karakoram. Understanding the effect of land use and altitude on SOC stock is essential for effective land use planning and mitigation of climate change. This study may also provide information on what might happen to SOC pool under a warming global climate scenario in ecologically highly vulnerable and economically extremely marginal mountain areas.

Materials and methods

Study area

The Bagrot Valley is situated in the southern part of the Karakoram Range, about 17 km to the northeast of Gilgit, Pakistan. The Karakoram Range lies in the center of the high mountain node: Hindukush in the west, Pamir and Tien Shah to the northwest, Kulun Shah in the northeast, and Himalaya in the southeast. The total watershed area is 440 km²: 16.14% comprised of pasture, 16.43% forest, 3.35% arable land, 64.08% ice, glacier, and rock (WWF Pakistan 2009). With rising altitude, summer becomes shorter and cooler whereas winters are longer. The relationship between altitude and climate (temperature and precipitation) variation in the study area (Bagrot Valley) has been calculated by Cramer (2000): There is an average temperature decline of 0.63°C /100 m and a statistical precipitate increases from + 17.6 mm per 100 m. The annual mean air temperature decreases from 11.6°C at 2210 m (Sinker) to – 2.2°C in 4030 m (Diran). Similarly, the annual precipitation increases from 142 mm (2210 m) to 721 mm (4030 m). Geologically, the area consists of the Chalt Volcanic Group, which has a high content of SiO₂, MgO, Fe₂O₃ and CaO (Petterson et al. 1991). In the Bagrot Valley, people have practiced ‘Mixed Mountain Agriculture’ for at least 1000 years. This includes agriculture, pastoralism, and forestry. The forest site in this study represents an undisturbed ecosystem extending from 2700 m to 3600 m altitude, where Juniperus, Pinus wallichiana (pine/chi), Picea smithiana (spruce/kachul) and Betula utilis are the dominant plants. Similarly, the native grassland sites in the study started from 3000 m altitude to 4200 m, and mostly consist of alpine grass including Bistorta affinis, Leontopodium, Gentiana tianschanica, Pedicularis cheilanthifolia, Poa alpina, Carex stenocarpa, Festuca alatavica and Swertia petiolate. The agriculture land are located from 2000 m to 3010 m altitude, where wheat (Triticum aestivum), maize (Zea mays), and potatoes (Solamum tuberosum) are grown (Figure 1).

Figure 1. Bagrot valley, Central Karakoram National Park, Gilgit-Baltistan, Pakistan.

Experimental design and sampling

In the present study, variation in SOCS was determined across the different land uses and along the altitudinal gradients in the Bagrot Valley, Northern Karakoram, Pakistan. To determine the variation in SOCS across the land use, soil samples were collected from adjacent land uses such as pastures, forests and arable land located at the similar altitudes. Five sample plots of 20 × 20 m² were randomly selected in each land use. Similarly, for the investigation of variation in SOCS along the altitude gradient, 10 sampling sites at different altitudes were selected in each of the land uses including pasture (3317–4163 m), forest (2787–3600 m), and arable land (2100–2930 m). Furthermore, five sample plots of 20 × 20 m² were located in the sites.

In each plot, five randomly points were chosen for sampling. After removing the vegetation, litter, roots, stones, and debris from the surface of the sampling point, then soil samples were taken from three depths (0–20, 20–40 and 40–60 cm) using soil agar in a ‘X’ type pattern and making a composite sample for each layer. In this way, five composite replicates samples were taken from each plot. One undisturbed core sample was also collected from each plot for bulk density by using a 100 cm³ core ring and which was packed in the sealed bag. A total of 720 soil samples were collected for the current investigation. This includes 270 samples for land use study (three land uses × five plots × three depths × five replicates), and 420 samples (three land use × ten sites × three depths × five replicates) for altitudinal investigation. The exact location of sampling points and altitude were recorded by a GPS (Garmin nuvi 3490 LMT 4.3-Inch). The vegetation and morphological characteristics were also recorded. The samples were packed in separate polyethylene bags and labeled. These samples were brought to the laboratory and then air dried. The dried samples were lightly ground, sieved through a (< 2 mm), and used for analysis SOC and pH.

Laboratory analysis

Bulk density was determined by the core method described by Blake and Hartge (1986). The samples were weighed in their sealed bags, clumps broken by hand and then oven dried at 40°C to constant weight. Thereafter, each sample was oven dried at 105°C for 4 h, which allowed for the calculation of soil bulk density. The samples were then be dry sieved (< 2 mm) and the rock fragment > 2 mm was weighed. The method described by Rowel (1994) was used for determination of pH in 1: 2.5 (soil: water). The soil organic carbon was determined by Walkley and Black (Nelson & Sommers 1982). The conversion of oxidizable organic C to total C was done by multiplying with the factor 1.30. The total soil organic carbon stock (Mg ha⁻¹) was calculated by multiplying the C concentration (g 100 g) with the calculated bulk density (g cm⁻³) and depth of soil layer (cm) by using the following formula described by Batjes (1996): Where, F is the volume of the fraction of fragments > 2 mm.

Data analysis

An analysis of variance (ANOVA) was used to determine the effect of land use and altitude on BD, SOC, and SOCS at the different depths. A Least Significant Difference (LSD) comparison test was carried out to separate statistically different means (p < 0.05). Linear regression analysis was carried out to determine the relationship of SOC with altitude and other variables. Statistical Package for Social Sciences (SPSS) was used for all analyses (Einstein & Abernethy 2000).

Results and discussion

Effect of land uses on soil organic carbon stock

Analysis of variance (Table 1) showed that land use and soil depth had significant effects on bulk density (g cm⁻³), SOC concentration (%) and SOC stock (Mg ha⁻¹). The overall mean SOC and SOCS in the forest (3.74% and 59.35 Mg ha⁻¹) was significantly higher than pasture (2.51% and 42.48 Mg ha⁻¹) and was the lowest in arable land (0.93% and 23.63 Mg ha⁻¹) (Figure 2). Similar trends were observed for each depth. The highest SOC and SOCS were observed more often in the upper layer (0–20 cm) than in the middle layer (20–40 cm) and was lowest in the bottom layer (40–60 cm). SOC plays a significant role in food security and climate change mitigation (Lal 2004). It also helps to control soil erosion, increase infiltration and porosity of clay soil, improve soil structure and stability; which contributes to the cycling of plant nutrients and provides energy to microbes and fauna, increasing their activity (Schlesinger 1997). In this study, SOC was found to be very sensitive to land use intensification especially in ecologically highly vulnerable and economically extremely marginal land. Due to intensive agricultural activities, about 60.18% less SOC was found in arable land as compared to the forest and 44.37% less than pasture. The intensive land uses such as continuous plowing and cultivation accelerated the rate of organic matter decomposition and mineralization. Further, the removal of biomass during harvesting and periodic tillage breaks up macro aggregates, resulting in increased soil erosion and runoff. Other researchers also reported that cultivation can reduce SOC (Billings 2006; Wang et al. 2008), and can change the balance between humification and mineralization processes (Saviozzi et al. 2001). Likewise, tillage changes the soil moisture, aeration, and nutrient concentrations, resulting in increased oxidation of soil organic matter (Lal 1995; Pandey et al. 2010) and the loss via soil microorganisms (Reicosky & Forcella 1998); this can in turn reduce the inputs of organic matter from above- and below-ground vegetation (Zhao et al. 2005; Yimer et al. 2006). In contrast, the higher SOC in the forest could be due to the greater biomass input by vegetative material, which is the main source of soil organic carbon. Organic matter generally has a positive relationship with plant density (Rees et al. 2005; Thomas et al. 2007). Forests store more C than any other terrestrial ecosystems and represent a significant carbon pool for the global carbon budget (Houghton 2007). It is estimated that almost 60% of the world’s terrestrial carbon is stored in forest vegetation and soil (Winjum et al. 1992); it comprised of 80% of aboveground terrestrial C and 40% of soil C (Dixon et al. 1994; Batjes 1996; Waring 1998; Goodale et al. 2002).

Figure 2. Variation Soil bulk density, SOC (%) and SOCS (Mg/Ha) across different land uses in various depths. Values having the same letter do not differ significantly from each other with respect to land uses and depth at p ≤ 0.05. Letters x, y, z indicate the variation within land uses in different depths, while, letters a, b, c indicate the variation among the land uses in each depth. Abbreviation BD (bulk density); SOC (soil organic carbon) and SOCS (soil organic carbon stock).

Table 1. Effect of land uses on soil BD (g cm⁻³) and SOC (%) and stock in different depth.

Variables	BD (g cm^−3)	SOC (%)	SOCS (Mg ha^−1)
Land use	1107.00***	3977.00***	1126.00***
Depth	244.30***	3701.00***	1020.00***
Land use * depth	3.58*	314.76***	32.48**

Notes: Values in column and rows represents F-values. BD, Bulk density; SOC, soil organic carbon; SOCS, soil organic carbon stock.

*^, **^, ***Significant at 0.1, 0.05 and 0.01 level, respectively.

In this study, about 50–60% of SOC was found in the top layer, while 25–30% was in the middle and 10–15% in the bottom layer. The highest SOC and SOCS in the upper layer could be due to continuous addition of undecayed and partially decomposed plant and animal remains in the top soils. Other researchers have also reported that surface layers have accumulated more organic materials than deeper layers (Wang et al. 2008; Emiru & Gebrekidan 2013).

Effect of altitude on soil organic carbon stock

Analysis of variance (Table 2) showed that both altitude and soil depth had significant effect on BD (g cm⁻³), SOC (%) and SOCS (Mg ha⁻¹) in all land uses (p < 0.01). In arable land the overall mean SOC was significantly increased from 0.52–1.0% and SOCS from 15.48 to 23.87 Mg ha⁻¹, with increasing altitude from 2100–2930 m. A similar trend was observed for each depth. Furthermore, SOC and SOCS decreased significantly with increasing depth (Figure 3). In contrast, in the forest the SOC increased (from 2.43 to 12.49%), as well as SOCS (from 48.42–165.26 Mg ha⁻¹), with an increase in altitude from 2797 to 3460 m (Figure 4). Above this altitude, a slight decrease in SOC was found up to 3600 m. The same pattern was also observed in the 0–20, 20–40 and 40–60 cm soil layers (Figure 5). Similarly, in pasture, the SOC concentration significantly increased from 1.21–8.57% as did SOCS (from 33.82–102.0 Mg ha⁻¹), with increasing altitude from 3017–3767 m, and then slightly decreased to 4163 m (Figure 5). Similar results were observed for the top, middle and bottom layers. Furthermore, SOC (%) and SOCS (Mg ha⁻¹) had a significantly negative relationship with bulk density, soil depth and soil pH (Figures 3–5).

Figure 3. Relationship of soil organic carbon stock with different variable in arable land.

Note: * indicates correlation is significant at p ≤ 0.05; abbreviations BD (bulk density; SOC (soil organic carbon) and SOCS (soil organic carbon stock).

Figure 4. Relationship of soil organic carbon stock with different variable in forest.

Note: * indicates correlation is significant at p ≤ 0.05; abbreviations BD (Bulk density; SOC (soil organic carbon) and SOCS (soil organic carbon stock).

Figure 5. Relationship of soil organic carbon stock with different variable in Pasture.

Note: * indicates correlation is significant at p ≤ 0.05; abbreviations BD (bulk density; SOC (soil organic carbon) and SOCS (soil organic carbon stock).

Table 2. Effect of altitude on soil organic carbon stock in each land uses in various depth.

Variables	SOCS (Forest)	SOCS (Pasture)	SOCS (Arable land)
Altitude	5920.01***	1986.01***	1109.00***
Depth	50470.00***	1599.00***	8000.00***
Altitude * depth	633.27***	109.08***	377.37***

Notes: Values in column and rows represents F-values. SOCS, soil organic carbon stock.

***Significant at 0.01 level.

Knowledge of vertical patterns of soil organic C can improve our understanding of the dynamics of SOC along a profile and the potential response of SOC to climate change; a small variation in climate at higher altitudes will drastically change the ecology and function of the mountain ecosystem (Jodha 1992). Similarly, altitude has a strong effect on the micro-climate and landscape (Bendix et al. 2008; Gerold et al. 2008). With increasing altitude, temperature normally becomes cooler and precipitation increases, resulting in vegetation distribution change, which impacts SOC stock (Tanner et al. 1998; Proctor et al. 2007). In addition, climate shifts in temperature and precipitation have a major influence on the decomposition and amount of SOC stored within an ecosystem and released into the atmosphere.

In the study area have the same parent material and vegetation along with the altitudes, however increased in SOC and SOCS with increasing altitude could be due to variation in climatic parameter along with an elevation gradient. In the area, the average temperature decreased by 0.63°C per 100 m and precipitation increased + 17.6 mm per 100 m Cramer (2000). Cooler temperatures and higher precipitation with altitude could retard the decomposition of litter. In turn, the accumulation of organic matter may have increased with this altitude. Other researchers have reported a positive relation between SOC with altitude (Sims & Nielsen 1986; Ineson et al. 1998; Charan et al. 2012). Furthermore, the lightly decreased amount of SOCS at higher altitudes in forest and pasture could be due to harsh climatic conditions, which provided less time for soil formation, restrict decomposition of the organic matter and produced low net primary production. At the upper timberline (Dame 3780 m), there were 102 frosting and 135 ice days, and vegetation period were short (112 days). Likewise, in the alpine meadow (4150 m), there were seventy-three frost and 188 ice days indicating the rough climate conditions. The thermal vegetation period lasted 95 days. As a result, vegetation density decreased in forest and pasture at higher elevations due to short growing period. The vegetative material, which is the main source of soil organic carbon and SOC, generally has a positive relationship with plant density (Rees et al. 2005; Thomas et al. 2007). Other researchers have also found a decrease in SOC at higher altitudes (Panthi 2010; Saeed et al. 2014). The low SOC at lower altitudes could be due to higher temperatures and less precipitation. In addition, mineralization of soil organic matter increased with rising temperature (Garcia-Pausas et al. 2007; Djukic et al. 2010) and the rate of reaction doubled with each 10°C increase in mean temperature (Bot & Benites 2005).

The highest SOC (21.61%) in this analysis was observed in the top layer in Betula utilis forest at an altitude of 3460 m. Similar SOC (21.9%) have been reported by Schickhoff (2005) in Batura II (3500 m). Bhat (2010) reported the highest SOC in Betula utilis forest (3550 m) in high mountain temperate Himalayan forests of India and further suggested that Betula utilis was a good source of biomass production and carbon balance. In this study, about 50 -60% SOCS was observed in the top layer, while 25–30% was in the middle and 10–15% in the bottom layer. This could be due to the continuous accumulation of undecayed and partially decomposed plant and animal residue in the surface soils. Other researchers have also reported the highest SOCS in the top soil layer (Woldeamlak & Stroosnijder 2003; Emiru & Gebrekidan 2013).

SOC had a negative relationship with bulk density and pH (Figure 3–5) in this current study. Bulk density value is dependent on the amount of organic matter. The soil with high organic matter has a low bulk density (White 1997; Gupta 2004). Furthermore, high organic matter, by oxidation, produced organic acids in the soil solution causing a decrease in soil pH. Fey (2001) also has reported that organic matter can have a negative relationship with pH.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Shamsher Ali, Ph.D., is a Lecturer and Environmental Chemist at the Karakoram International University, Pakistan, working on the response of soil organic carbon to land use intensification and environmental change.

Farida Begum, Ph.D., is an Assistant professor and Environmental Scientist at the Karakoram International University, Pakistan, working on Carbon Dynamics and Sequestration potential under different land use systems for sustainable management of agro ecosystem and mitigation of climate change.

Rifat Hayat, Ph.D., is an Associate Professor and Soil Scientist at the Pir Maher Ali Shah Arid Agriculture University, Pakistan, working on Soil microbiology/Systematic/Soil Metagenomics under different cropping systems.

Brendan J. M. Bohannan, Ph.D., is a Professor of Biology and Environmental Studies at the University of Oregon, Eugene, USA, working on causes and consequences of microbial biodiversity and the response of bacterial and archaeal communities to environmental change. Prof. Bohannan is especially interested in promoting the integration of microbial ecology into the general science of ecology.