Soil genesis and heterogeneity of phosphorus forms and carbon below mounds inhabited by primary and secondary termites

Termite activities are known to significantly influence small-scale soil properties in tropical savannas. The lateral and vertical extent of the alterations to the nest's surrounding, and particularly resulting impacts on diagnostic soil horizons remain largely unresolved until today. We examined the effects of mound-building termites on soil genesis and constitutive chemical soil properties in and below their nests. Two transects to a soil depth of 100 cm were dug below three younger mounds of Cornitermes silvestrii (the primary nest builder), three older mounds in which C. silvestrii had died out and which were secondarily colonized mainly by Nasutitermes kemneri, and three reference sites in the Brazilian Cerrado. The samples were characterized by standard procedures for soil classification; in addition, phosphorus extractions were conducted on selected samples using NaHCO₃ for labile P forms, and concentrated HCl for stable P forms. This data set was then used to build calibration models for the prediction of labile and stable inorganic (P_i) and organic (P_o) P forms, as well as for contents of organic carbon (OC), for the remaining samples applying mid-infrared spectroscopy in combination with partial least squares regression (MIRS-PLSR). We can show that the termite influence on the soil was sufficiently large to change diagnostic characteristics of the soils under termite mounds. The MIRS-PLSR predictions were suitable for quantifying organic carbon and most of the labile and stable phosphorus fractions. They showed an enrichment of OC, NaHCO₃-P_o and NaHCO₃-P_i contents in nests inhabited by primary and secondary termites by factors of 1.6–2.0 and 1.4–1.5, respectively. The soils surrounding the nests had higher contents of OC and NaHCO₃-P under both nest types vertically down to 30 cm below the lower nest border, and OC and NaHCO₃-P_i contents were elevated at minimum to a lateral distance of 60 cm away from the nest border. As the pattern of HCl_conc-P_i, which comprised 95% of total P, showed no variations, we conclude that the higher NaHCO₃-P_i amount was formed in termite nests by changing the availability of the more stable HCl_conc-P_i. In contrast to the contents, the OC and NaHCO₃-P stocks below the mounds inhabited by primary termites were comparable to those inhabited by secondary ones, because the bulk density of the secondarily inhabited nests was elevated. This was due to a transport of clay-rich material from the subsurface argic horizons into the nests. Here, the secondary termites even reverted the lessivation observed in the reference soils and under mounds inhabited by primary termites, thus causing the soil types to change from Alisols and Acrisols to the properties of Umbrisols.