Shifting focus from external to in situ organic resources – The redesign of four tropical long-term experiments

Abstract

Long-term experiments (LTEs) are critical for evaluating strategies that can maintain or increase crop yields, soil fertility and soil organic carbon (SOC), and help adapt to climate change. Yet, scientific knowledge is advancing and research questions are evolving. Therefore, it is important to review the objectives of LTEs over time. A change in their design may be necessary to keep the experimental treatments scientifically interesting, innovative, and relevant in the context of evolving agricultural challenges. Here, we describe the process of redesigning four LTEs in Kenya. These LTEs are unique in that they represent four different pedoclimatic conditions but with identical experimental treatments across sites. Initially, they focused on investigating how to maintain or increase SOC and maize yields over time by applying a combination of mineral nitrogen (N) and external organic resources. Specifically, the experimental treatments consisted of maize monoculture with different rates (1.2 and 4 t C ha−1 yr−1) and qualities of organic resources, either with or without mineral N fertilizer input. After about 20 years, it became clear that SOC was lost in most treatments. Therefore, continuing with the current experimental design was not an option. Taking advantage of the fact that the different former treatments led to different levels of soil degradation, we redesigned the LTEs to study the effectiveness of regenerative cropping strategies in rebuilding SOC and increasing crop yields starting from the different levels of soil degradation. The focus shifted from external to in situ organic inputs by increasing the root biomass of the cultivated crops. The newly established cropping system treatments are maize-legume rotation, maize-legume intercropping (double row configuration) and relay intercropping of maize with forage grass. A key finding from the previous phase of the experiments, namely, that external organic inputs with low C:N ratios are most efficient in building SOC, has been incorporated into the redesign. The relative contribution of external versus in situ organic resources is tested by splitting the cropping system treatments into those receiving either farmyard manure or green manure in the form of Tithonia diversifolia prunings and those receiving no external inputs. Split-plot treatments with and without mineral N were retained. The overall objective of studying mechanisms of tropical soil fertility maintenance and, more specifically, SOC formation, remained unchanged. However, the redesign aligned the LTEs with the current state of knowledge and pressing research questions, specifically focusing on the relative effectiveness of in-situ versus external organic inputs in SOC formation.