Pedogenic Controls of Deep Soil Organic Carbon Storage Along a Tropical Highland Toposequence

Technical Summary: Pedogenic Controls of Deep Soil Organic Carbon Storage Along a Tropical Highland Toposequence

Problem Statement
Deep soil organic carbon (SOC; >30 cm) constitutes a major component of terrestrial carbon stocks, particularly in tropical highland environments, yet it remains poorly constrained in global inventories. Existing studies, including those in the Ethiopian highlands, have largely focused on surface soils (typically <100 cm), leading to systematic underestimation of total carbon stocks. While previous work in the Kersa sub-watershed documented soil variability and quantified SOC to 60 cm, the distribution of deep SOC and the specific pedogenic controls regulating its redistribution along toposequences remain insufficiently understood.

Methodology
This exploratory, pedon-scale study was conducted in the Kersa sub-watershed of the eastern Ethiopian highlands (sub-humid climate, 1968–2127 m a.s.l.). The research design involved:

• Sampling Strategy: Eight representative pedons were described and sampled across three slope positions (upper, middle, and lower) to capture a full toposequence.
• Soil Characterization: Soils were morphologically described and classified according to the World Reference Base (WRB) system. Sampling was conducted by genetic horizons down to 200 cm depth.
• Laboratory Analysis: Standard procedures were employed for particle size analysis (Bouyoucos hydrometer), bulk density (core method), and SOC concentration (Walkley–Black wet oxidation). Note that SOC values were not corrected for incomplete oxidation.
• Calculations: SOC stocks (Mg ha⁻¹) were calculated for each horizon using bulk density, SOC concentration, horizon thickness, and coarse fragment content.
• Statistical Analysis: A General Linear Model tested slope differences, and Pearson correlations assessed SOC–clay relationships. Due to limited replication (n=8 pedons), analyses are explicitly framed as exploratory.

Key Results

• Soil Development and Depth: Soil depth increased significantly downslope, ranging from 25 cm on upper slopes to 200 cm on lower slopes. Five Reference Soil Groups were identified: Leptosols and Regosols (upper slope), Cambisols (middle slope), and Luvisols and Vertisols (lower slope).
• SOC Stock Distribution: Total SOC stocks (0–200 cm) ranged from 43 to 347 Mg ha⁻¹. Lower slope positions stored 4.4 to 8.1 times more SOC than upper slopes.
• Vertical Distribution: In a representative Luvisol, 75% of the total SOC was stored below 30 cm depth. The 100–200 cm layer alone contained 33% of the total stock.
• Pedogenic Controls: SOC concentration showed a strong positive correlation with clay content, with correlation coefficients increasing with depth (r = 0.60 at 0–20 cm to r = 0.89 at 40–60 cm). This indicates a shift toward organo-mineral stabilization in subsurface horizons.

Key Contributions
The study quantifies deep SOC stocks to 200 cm in a tropical highland setting and identifies specific pedogenic mechanisms driving their distribution:

1. Topographic Control: Demonstrates that colluvial deposition and reduced erosion on lower slopes facilitate the accumulation of deeper, SOC-rich profiles.
2. Pedogenic Mechanisms: Highlights the role of clay translocation (forming argic horizons in Luvisols) and vertic structure development (shrink-swell dynamics in Vertisols) in concentrating and stabilizing deep carbon.
3. Sampling Depth Implications: Reveals that standard 0–30 cm sampling captures only 25% of total SOC in deep Luvisols, suggesting that current inventories in similar landscapes significantly underestimate carbon stocks.

Significance and Limitations
The authors position these findings as hypothesis-generating rather than definitive regional estimates. The study emphasizes that pedogenic processes—specifically clay translocation, argic horizon formation, and colluvial deposition—jointly regulate soil depth and SOC stabilization along slope gradients.

The paper explicitly acknowledges limitations, including the small sample size (eight pedons), the single-watershed design, and the lack of SOC fractionation or isotopic analyses (such as radiocarbon dating) to confirm mechanistic stabilization pathways. Furthermore, the use of the Walkley–Black method may underestimate true SOC values by 10–30%. Consequently, the authors conclude that these results should support future regional-scale studies that incorporate SOC fractionation and isotopic analyses to better understand deep carbon dynamics in tropical highlands.