Variation in soil carbon stocks with depth along a toposequence in a sub-humid climate in North Africa (Skikda, Algeria) Zohra Bounouara a , Tiphaine Chevallier b, * ,J  er ^ ome Balesdent c , Joele Toucet b , Mahtali Sbih d , Martial Bernoux b , Noureddine Belaissaoui a , Ouarda Bouneb a , Rabah Bensaid a a University of Skikda, Algeria b IRD UMR Eco & Sols, Montpellier, France c Aix Marseille Univ, CNRS, IRD, Coll France, INRA, CEREGE, Aix-en-Provence, France d University of Batna, Algeria article info Article history: Received 4 April 2016 Received in revised form 9 January 2017 Accepted 6 February 2017 Keywords: Deep carbon Toposequence Mediterranean soils Sub-humid climate Size fractionation of organic matter Organic matter mineralization abstract Deep soil carbon (>30 cm) is the primary pool of soil organic carbon (SOC). However, deep SOC is often not included in carbon inventories or management strategies, especially in North-Africa. The objective of the study was to determine the distribution and the main factors governing SOC storage along a representative toposequence in North-Eastern Algeria. The soil organic matter (SOM) was characterized by SOC content, carbon stocks, particle size distribution and mineralization potential. The amount and the dynamics of the deep SOC depended on the topographical position. In the mountain soils, most of the SOC stock (100 t C ha 1 ) was in the topsoil (0e30 cm) and associated with the >50 mm fractions (plant materials) whereas in the piedmont and alluvial plain soils, most of the SOC stock (168 t C ha 1 ) was accumulated in subsoil (30e100 cm) and associated with the <20 mm fractions (mineral fractions). The SOC in the subsoil contributed to SOC potential of mineralization and should not be underestimated, especially in the alluvial cultivated plains. This carbon should be considered as an important component in agro-ecological agriculture but it is also a potential source of CO 2 . © 2017 Published by Elsevier Ltd. 1. Introduction Soil organic matter (SOM) has many properties that contribute to ecosystem services. It is a major component of soil fertility and productivity and a potential sink for atmospheric CO 2 . SOM is a key factor in the global carbon (C) cycle (Paustian et al., 1997) and the importance of SOM in the Mediterranean area has been clearly established (Annabi et al., 2009; Ben Hassine et al., 2008; Boix- Fayos et al., 2009; Brahim et al., 2011; Martinez-Mena et al., 2008). In the slightly acidic, low-activity clay soils of northern Algeria, which cover about 413,760 ha, SOM is a key component of soil fertility. Here, in the hilly relief which is exposed to frequent rainfall events, the SOM content is the major factor that protects against erosion (Roose et al., 2005). The SOM content and behavior need to be determined for these vulnerable soils. However there are few data published from these regions and there are still gaps in our knowledge of the mechanisms that govern SOM dynamics. The few studies of the quantity and quality of SOM and its dynamics carried out in Algeria have often been limited to preliminary characterization (Bensid et al., 2015; Dellal and Halitim, 1992). Different types of land use and soil management along a top- osequence affect the stocks, dynamics and forms of SOM in surface soils (Bernoux et al., 2006; Brahim et al., 2009) but their effect on deep C is still unclear (Don et al., 2011; Poeplau et al., 2011; Rumpel and Kogel-Knabner, 2011; Shi et al., 2013). Deep soil carbon (>30 cm) is the primary pool of soil organic carbon (SOC) and, in recent years, has received increasing attention (Mathieu et al., 2015; Rasse et al., 2006; Rovira and Vallejo,1997; Shi et al., 2013). This “deep carbon” can contribute to overall soil fertility and may be a sink or a source for atmospheric CO 2 , depending on its dy- namics (Rumpel and Kogel-Knabner, 2011). In general, a major part of the SOM in subsoil horizons comes from root biomass, dissolved OM and bioturbation transfers. It is characterized by a high pro- portion of microbial derived OM and has, therefore, a high * Corresponding author. E-mail address: tiphaine.chevallier@ird.fr (T. Chevallier). Contents lists available at ScienceDirect Journal of Arid Environments journal homepage: www.elsevier.com/locate/jaridenv http://dx.doi.org/10.1016/j.jaridenv.2017.02.001 0140-1963/© 2017 Published by Elsevier Ltd. Journal of Arid Environments 141 (2017) 25e33