Elucidating relationships between the soil food web, soil processes, and agroecosystem function is a critical step toward a more sustainable agriculture. Soil and crop management practices can alter these relationships, and their effects can persist even after imposing new management practices. In 2005, the Cornell Organic Grain Cropping Systems Experiment was established in central New York. Four cropping systems that varied in fertilizer inputs, tillage practices, and weed control were compared: High Fertility, Low Fertility, Enhanced Weed Management, Reduced Tillage. Two crop rotation entry points were included in the experiment. In June 2017, the entire experimental site (plots and alleyways) was plowed and seeded with sorghum sudangrass [Sorghum bicolor (L.) Moench x S. sudanense (Piper) Stapf] as part of a uniformity trial to assess legacy effects of past management practices. Prior to initiating the uniformity trial, soil samples were collected and analyzed for soil health indicators. Soil samples were also collected to assess soil invertebrate abundance and community structure 34 and 70 days after planting. Sorghum sudangrass and weed biomass were sampled at the end of the uniformity trial in September 2017. Legacy effects of past management that were observed during the uniformity trial were associated with differences in nutrient inputs and soil disturbance, as well as the preceding crop. The High Fertility system had greater soil phosphorus than the Low Fertility system, and in one of the two crop rotation entry points, soil aggregate stability and soil respiration were greater in the Reduced Tillage system compared to the Enhanced Weed Management system. The Enhanced Weed Management cropping system also had a soil invertebrate community characterized by more disturbance tolerant taxa. Weed biomass varied by crop rotation entry point, but not cropping system. Sorghum sudangrass biomass was greater in the Reduced Tillage system than the Low Fertility system, and the entry point that had greater weed biomass also had greater sorghum sudangrass biomass. Piecewise structural equation modelling (SEM) was used to test relationships between response variables and showed that soil phosphorus, soil aggregate stability, and soil respiration explained variation in abundance of some invertebrates, and that aggregate stability, soil respiration, soil moisture, weed biomass, and a select group of invertebrates affected sorghum sudangrass biomass production. Overall our findings show that soil invertebrates can mediate the relationship between soil health indicators and crop productivity, and provide support for including direct measurements of soil invertebrates in soil health assessments.
Agricultural Systems, Volume 177, January 2020, 102719,