4. Discussion
The results show that the SOC content in the whole soil cover of the agriculturally used loess slopes and plateau tops (including microtopography i.e. CDs) of the Nałęczów Plateau is 100-144 Mg·ha-1. If we disregard the specific for loess area microtopography (CDs), this value in the area under study will be slightly lower, i.e. 102.38 Mg·ha-1 of SOC. In eroded sites (loess slopes and plateaus) 51.07 Mg·ha-1 SOC are stored in the topsoil (i.e. the A horizon, mean depth 0.24 m). In depositional sites (i.e. CDs) about ¼ (38.26 Mg SOC·ha-1) of SOC are stored in the topsoil. This confirms the findings of Rumpel and Kögel-Knabner (2011) that high SOC storages may be present below the topsoil and not only within the topsoil. Therefore, more attention should be given to subsoils when assessing SOC stocks. Otherwise, calculated SOC stocks will be strongly underestimated.
For Wallonia, Chartin et al. (2016) determined the mean predicted SOC stock in the topsoil at 59.9 Mg C·ha-1 in cropland (53.5-107 Mg C·ha-1 for different regions). For the same region, Lettens et al. (2005b) determined the mean SOC stock value at 48 Mg C·ha-1 in cropland. These differences results from the different research methods adopted by the authors. In France, mean SOC stock for cropland is 55.7 Mg C·ha-1(Meersmans et al., 2012). At the scale of the whole territory of Belgium, Meersmans et al. (2011) forecasted a SOC stock value of 49.5 Mg C·ha-1 for cropland. In Poland, the present soil organic carbon stock (SOC) in topsoils for the main soil types range from 36 to 42 Mg C·ha-1 (Faber et al., 2012). The results of SOC content in topsoil obtained in this study are comparable with the results for other European regions.
In the agriculturally used loess areas subject to erosion, topography has both a direct and an indirect impact on SOC distribution in the landscape. The direct impact consists of the accumulation of SOC-rich colluvial sediments in depressions; these sediments are products of soil erosion on hillslopes (Sorensen et al., 2006). Erosion causes a loss of SOC in topsoils of sloping land and a reduction of their thickness. In the eroded loess landscapes of the Nałęczów Plateau, CDs are SOC pools where soil erosion-derived colluvial sediments of considerable thickness are accumulated. SOC storage in the CDs originates not only from soil erosion in their catchment but also from fossil soils that have been buried by SOC-rich colluvial sediments. Together with the subsurface fossil soils, closed depressions store more (almost twice) SOC (mean 192.85 Mg·ha-1) than eroded soils on sloping cropland (mean 102.38 Mg·ha-1). These results demonstrate that if one neglects microtopographic features like CDs the true SOC storage will be underestimated.
The study of Grimm et al. (2008) indicated that topographic features explained most of the spatial variability of soil organic carbon content in the topsoil (0–0.1 m) at the regional scale. In the subsoil (0.10–0.50 m), SOC distribution was related to soil texture.
The investigations conducted in the Nałęczów Plateau indicate that the microtopography has also a significant impact on the vertical distribution of SOC in soil profiles. About 1/4 (38.26 Mg·ha-1) of the mean SOC storage in CDs is found in the topsoil (0-0.24 m) of these forms (Table 5). Most SOC stored in the CDs occurs in the subsoil and fossil soil (at a depth ca. 1.0 m). The topsoil at the loess plateau tops and slopes (without CDs) contains about half (51.07 Mg·ha-1) of the mean amount (102.38 Mg·ha-1) of SOC within the soil cover of these areas (Table 7).
The indirect impact of topography on the spatial SOC distribution in the landscape is linked to its influence on the larger soil moisture contents (and even water logging) within the CDs and hence on SOC storage. This has been confirmed in studies by Meersmans et al. (2011) who demonstrated that, at the catchment scale, a higher SOC content is correlated with soils of waterlogged valley bottoms. The bottoms of CDs are areas with a larger moisture content than the surrounding slopes, which can be deduced from the redoximorphic features in colluvial sediments deposited in the CD bottoms. The SOC mass stored in the colluvial sediments of the CDs is similar to that of the underlying fossil soils. On average, 10.23 Mg SOC per CD is stored in the colluvial sediments of the studied CDs, while 10.39 Mg SOC per CD is stored in the underlying fossil soils (Table 2). Furthermore, SOC distribution in fossil soil profiles indicates a vertical leaching of SOC deeper into the soil profiles as a result of surface water infiltration in the CD bottom. This is deduced from the morphology and micromorphology of the soils in the CDs and their geochemical properties (Kołodyńska-Gawrysiak et al., 2017). The indirect impact of topography on SOC storage, expressed by the topographic wetness index (TWI), was emphasized by Wiesmeier et al. (2013). Depressed areas display on average a larger soil moisture content which delays the decomposition of organic matter and is conducive to the accumulation of SOC.
According to Wiesmeier et al. (2019), the influence of topography (topographic position, slope gradient, TWI) on the variation of SOC storage is significant only at a local scale. It is less significant at regional scale where SOC storage is determined by climate, vegetation, texture, parent material, land use and management. Models predicting SOC content at the national or large regional scale usually do not take topography into account either. The model of SOC prediction at the national scale, proposed by Martin et al. (2010) showed that SOC distribution is strongly correlated with land use (forest soils, cropland soils). This study reveals the significant impact of (micro)topography (microrelief) on the SOC storage for the regional scale. For the Nałęczów Plateau area, the inclusion of microtopographic features (i.e. closed depressions), typical for loess areas, enabled a more accurate estimation of SOC storage. 95.86 % of the studied eroded landscape actually holds to 10% more SOC than what models excluding SOC storage in CDs would suggest (Fig. 9). For about 4.13% of these areas, the exclusion of CDs results in an underestimation of SOC storage in the landscape from 10% to 30 % (Fig. 9).
The presence of thick colluvial sediment layers of different ages in the CDs suggests the influence of these microtopographic landforms on SOC storage in the past (Kołodyńska-Gawrysiak, Poesen & Gawrysiak, 2018). Today this carbon storage is often not quantified when conducting organic carbon inventories at local or regional scale. A similar remark has been made for buried peatlands that have accumulated significant C stocks over millennia (Treat et al., 2019).