Biochemical proxies indicate differences in soil C cycling induced by long-term tillage and residue management in a tropical agroecosystem

Abstract

Background & aim A potential benefit of conservation agriculture (CA) is soil organic carbon (SOC) accrual, yet recent studies indicate limited or no i m p a c t o f C A o n t o t a l S O C i n tr o p i c a l agroecosystems. We evaluated biochemical indicators of soil C cycling after 9 years (18 seasons) of contrasting tillage with and without maize residue retention in western Kenya. Methods Potential activities of C-cycling enzymes (β-glucosidase, GLU; β-galactosidase, GAL; glucosaminidase, GLM; cellobiohydrolase, CEL), permanganate-oxidizable C (POXC), and soil organic matter (SOM) composition (by infrared spectroscopy) were measured.Results POXC tended to be greater under reduced tillage and residue retention, but did not significantly differ among treatments (≤ 2% of SOC). Despite no significant differences in SOC concentrations or stocks, activities of all 4 C-cycling enzymes responded strongly to tillage, and to a lesser extent to residue management. Activities of GLU, GAL, and GLM were greatest under the combination of reduced tillage and residue retention relative to other treatments. Reduced tillage produced an enrichment in carboxyl C = O (+6%) and decreased polysaccharide C-O (−3.5%) relative to conventional tillage irrespective of residue management. Conclusions Though enzyme activities and POXC are typically associated with SOC accrual, changes in soil C cycling at this site have not translated into significant differences in SOC after 9 years. Elevated enzyme activities may have offset potential SOC accumulation under CA. However, the ratio of C-cycling enzyme activities to SOC was higher under reduced tillage and residue retention relative to other treatments, indicating that stoichiometric scaling of SOC and enzyme activities does not explain absence of significant differences in SOC among tillage and residue managements. Potential factors that may explain the low SOC accrual rates in this tropical agroecosystem included the low, albeit realistic, levels of residue retention, nutrient limitations, and high temperatures favoring decomposition.

Responses of β‐glucosidase, permanganate oxidizable carbon, and fluorescein diacetate hydrolysis to conservation practices

“…However, the randomization tests showed POXC's ES were significantly higher in fine-textured than medium-or coarse-textured soils (Figure 3b) and increased with study length (Figure 3d). Both of these trends are consistent with the idea that POXC can serve as a measure of newly stabilized SOC (Margenot et al, 2017). Instances where POXC did not accrue, or where differences were not statistically significant, could result from positive priming that results in net decay of organic C or SOC saturation.…”

Soil nitrogen cycling under contrasting management systems in Amazon Coffea canephora agroecosystems

“…Though no relationships were found between POXC and Ccycling enzyme activities, associations among soil labile C and corresponding cellulolytic enzymes have been identified in tropical agroecosystems. For example, in subhumid tropical Kenya, POXC was strongly and positively associated with Ccycling enzyme activities, despite POXC being similar among tillage and residue treatments (Margenot et al, 2017). Total and labile C pools including POXC have been found to be insensitive to N fertilization rates in annual cropping systems in temperate midwestern United States (Grandy et al, 2013;Mahal et al, 2019), consistent with evidence that N fertilization impacts on soil C are indirectly mediated by biomass inputs and not priming (Ladha et al, 2011;Poffenbarger et al, 2017).…”

Improved soil biological health increases corn grain yield in N fertilized systems across the Corn Belt

“…While this may be due to increases in SOC (Table S6; r = 0.18, p = 0.002), N fertilization can also influence other more temporally-variable soil biological characteristics, such as microbial biomass 56 , plant residue composition 57 , and ligninolytic activity 58 . These parameters are not explicitly represented in our soil health factor, but they are indicative of alterations in C and N cycling that are highly related to our indicators of soil biological health 31,[59][60][61] . The inferred alterations to C and N cycling are also evident in the differing relationship between soil biological health and relative yield in our models: soil biological health was not related to relative yield in the unfertilized plots of the N responsiveness model ( Fig.…”