Carbonate in arid and semiarid soils of the southwestern USA and Mexico interacts with atmospheric CO2 in two long-term fashions. It is a unidirectional sink for CO2 in the lower and drier regions, and carbonate is a cyclic CO2 sink (carbonate formation) and CO2 source (carbonate dissolution) in the higher and wetter regions. The unidirectional sink consists of progressively more carbonate accumulating in progressively older soils, a process that continues until a soil and its geomorphic surface are destroyed by erosion. In southern New Mexico, for example, soil carbonate ranges from filaments in soils of Historical age to calcretes in soils of late Pliocene age. The cyclic sink-source pattern consists of soils that are near the Pedocal-Pedalfer boundary. These soils have less carbonate than their more arid counterparts because Pedalfer areas expanded during glaciopluvial periods dissolving carbonate and contracted during arid intervals giving rise to CaCO3 formation.

Biotic forms of pedogenic carbonate are common in the early stages of accumulation. Such forms include calcified root hairs, fungal hyphae, needle-fiber calcite, and micro-peloids. In the advanced stages of carbonate accumulation, such as calcic horizons and calcretes, micritic carbonate is the dominant form.

Carbon storage in soil carbonate occurs as non-pedogenic and pedogenic carbonate. Non-pedogenic carbonate, also termed primary, geogenic, or lithogenic carbonate, exists in soils as detritus, mechanically inherited from limestone and similar parent materials. Pedogenic carbonate, also termed secondary, authigenic, or in situ carbonate, formed in soil at its present location or was transported from preexisting soils. All would contain sequestered atmospheric CO2, though the time that sequestration took place can differ drastically.