Stability and instability on Maya lowlands tropical hillslope soils

Substantial lake core and other evidence shows accelerated soil erosion occurred in the Maya Lowlands of Central America over ancient Maya history from 3000 to 1000 years ago. But we have little evidence of the wider network of the sources and sinks of that eroded sediment cascade. This study begins to solve the mystery of missing soil with new research and a synthesis of existing studies of tropical forest soils along slopes in NW Belize. The research aim is to understand soil formation, long-term human impacts on slopes, and slope stability over time, and explore ecological implications. We studied soils on seven slopes in tropical forest areas that have experienced intensive ancient human impacts and those with little ancient impacts. All of our soil catenas, except for one deforested from old growth two years before, contain evidence for about 1000 years of stable, tropical forest cover since Maya abandonment. We characterized the physical, chemical, and taxonomic characteristics of soils at crest-shoulder, backslopes, footslopes, and depression locations, analyzing typical soil parameters, chemical elements, and carbon isotopes (δ13C) in dated and undated sequences. Four footslopes or depressions in areas of high ancient occupation preserved evidence of buried, clay-textured soils covered by coarser sediment dating from the Maya Classic period. Three footslopes from areas with scant evidence of ancient occupation had little discernable deposition. These findings add to a growing corpus of soil toposequences with similar facies changes in footslopes and depressions that date to the Maya period. Using major elemental concentrations across a range of catenas, we derived a measure (Ca + Mg) / (Al + Fe + Mn) of the relative contributions of autochthonous and allochthonous materials and the relative age of soil catenas. We found very low ratios in clearly older, buried soils in footslopes and depressions and on slopes that had not undergone ancient Maya erosion. We found high (Ca + Mg) / (Al + Fe + Mn) values on slopes with several lines of evidence that suggest relative youth, soils possibly formed since Maya abandonment. Carbon isotopes (δ13C) also provide some evidence of past vegetation change on slopes. We found strong evidence for maize or other alien C4 species in an ancient terrace soil and additional evidence in buried footslopes but only evidence for C3 species (like tropical trees) on the backslopes and other crest-shoulders. The fact that steep slopes preserved no evidence of C4 species inputs may mean that the ancient Maya maintained forests here. Alternatively, ancient Maya land uses eroded slopes, with the δ13C signatures detected today being the result of more recent soil development under forest over the last millennium. Additional evidence that these soils are recent in age includes elevated (Ca + Mg) / (Al + Fe + Mn) values, skeletal soil profiles, and low soil magnetic susceptibility. Besides the evidence for truncating backslopes and aggrading footslopes, the ancient Maya built agricultural terraces that accumulated soils and altered drainage. All these ancient Maya slope alterations would have influenced modern tree distributions, because many tree species in the modern forest show strong preferences for different soil types and topographic situations that the ancient Maya changed