Biological indicators of soil quality and impacts of land use management in agricultural landscapes of northern Nicaragua

Background/Question/Methods: The tropical dry forest region along the western slope of Central America is a highly biodiverse and fragile area that is under increasing pressure from agricultural production, thus threatening the provision of ecosystem services, biodiversity, and ultimately the integrity of these landscapes and the communities who depend on them.  To address this issue we evaluated the impact of common agricultural management practices (cropping and livestock systems) vs. the Quesungual slash-and-mulch agroforestry system (QSMAS), a promising alternative land use based on the inclusion of native trees with active pruning and residue management.  In February of 2011 soil sampling was conducted on a series of hillside farms near the town of Somotillo in northern Nicaragua to assess soil physical properties, chemical fertility and the abundance and diversity of soil macrofauna within four management systems: 1) QSMAS based on maize production; 2) traditional maize cropping system (TC); 3) pasture system (PS); and 4) secondary forest (SF), used as a reference.  In addition to examining the influence management, we sought to develop indicators of soil quality based on the presence or absence of particular soil invertebrate species as well as basic soil physical and chemical properties.

Results/Conclusions: Macrofauna excavated from the soil (0-30 cm) were more abundant under QSMAS (657 individuals m^-2) than in TC (177 individuals m^-2; P = 0.011), with PS and SF demonstrating intermediate populations.  At the same time, SF presented significantly higher richness of taxonomic groups than either TC or SP (P < 0.001), and demonstrated the greatest populations of Arachnida, important top-predators and potential indicators of food web structure.  While we found no significant influence of management system on soil chemical variables, soil compaction (penetration resistance and shear strength) was significantly higher under SP than either SF or QSMAS (P < 0.05), with TC intermediate in value.  Multivariate co-inertia analyses suggested significant covariation between data sets (chemical fertility, physical status, aggregate morphology and macrofauna).  Notably, a strong association between macrofauna and soil physical status (P = 0.013), suggested that compaction by livestock strongly limits the activity and diversity of soil macrofauna.  Finally, we used the Indicator Value index, which ranks species according to their specificity and fidelity across sites, along with farmer consultation to identify key indicator species of soil quality that could greatly facilitate future evaluation land management impacts by farmers and technicians in the region.