Soil nitrogen dynamics and plant-microbial interactions in climate smart forage cropping systems in Rwanda

Abstract

Rwanda’s nascent dairy industry holds promise for millions of small farmers as both a pathway out of poverty and a remedy for malnutrition. Dairy production in Rwanda is primarily limited by the seasonal availability of forage. A forage is a crop grown for use as animal feed or making silage. However, forage crop production in Rwanda is limited by the intersecting challenges of climate change and declining soil fertility. Climate-smart forages include perennial varieties that offer improvements over local feed sources in their contributions to soil health, nutritional quality, and potential to reduce adverse ecological impacts of farming. Encouraging smallholder livestock farmers to adopt these forage varieties is viewed as a multifunctional intervention that could mitigate greenhouse gas (GHG) emissions through improved livestock digestion, fill dry season feed gaps, and contribute to sustainable soil nutrient cycling. While there is a growing body of research suggesting climate smart forages improve nitrogen (N) use efficiency for reduced GHG emissions and improved soil fertility, evidence of farm scale applicability is lacking. The goal of this thesis work was to address critical research gaps that currently limit the adoption of novel perennial forage cropping systems with multifunctional benefits in Rwanda. We focused on the underexplored realm of plant-soil-microbial interactions that facilitate N supply, N transfer, and N retention in tropical rainfed agroecosystems. To determine whether climate smart forage crops impact microbial N cycling activity under a typical smallholder management regime, we collected plant and soil samplings from three replicated field trials in Rwanda. Forage treatments in each location included Napier grass (Cenchrus purpureus) as the baseline and the climate smart Brachiaria (Urochloa) cv. Mulato II; both of these perennial grasses were grown alone or intercropped with the perennial legume Desmodium sp. As the farmer-preferred annual forage crop, maize was also included. We quantified several N fractions and assayed for nitrification potential (NP) and denitrification enzyme activity (DEA). Additionally, we collected rhizosphere soil samples to characterize forage-associated bacterial and fungal communities. Finally, we collected forage leaf tissue to measure nutritive quality and calculate biological nitrogen fixation (BNF) in intercropped Desmodium. Ultimately, our results support the use of perennial forage crops and innovative perennial legume cropping systems in East Africa and Rwanda in particular. The inclusion of a legume intercrop did not stimulate microbial processes that lead to potential N loss and offered tangible agronomic benefits in terms of forage tissue quality in the companion crop. While maize plots suppressed the growth of nitrifying archaea and bacteria, Brachiaria plots were more resilient to changes in N cycling activity during a dry to wet seasonal transition than maize plots. Using network analysis, we showed that the functional potential of the nitrogen cycling community is seasonally dynamic, with fewer biochemical pathways for N loss in the rainy season. We also found that BNF increased in intercropping arrangements relative to single-cropped Desmodium by 91.6 - 147.1% on average in intercropped stands with Brachiaria and C. purpureus but not maize. Intercropping also induced positive changes in non-legume tissue quality and associated microbial communities in a species- and site-dependent manner.