Greenhouse gas emissions in croplands and downslope riparian buffer strips with varying vegetation

Abstract

Vegetated riparian buffer areas are implemented as an intervention to intercept, retain and process diffuse non-point source pollutants (NPS), including nitrate (NO3-), phosphorus (P), herbicides, and pesticides emitted from immediately adjacent agricultural lands. Due to their location in landscapes, riparian buffers maintain high organic matter (OM), increasing soil carbon (C) contents, retain high soil moisture from the periodically high water table, and contain higher nitrogen (N) content intercepted from upslope agricultural land. In turn, these conditions promote the production of gases including nitric oxide (NO), nitrous oxide (N2O), nitrogen gas (N2), methane (CH4), and carbon dioxide (CO2) through various soil processes, including denitrification, methanogenesis, and respiration. Earlier recommendations to the Intergovernmental Panel for Climate Change (IPCC) suggested that greenhouse gas (GHG) i.e., N2O, inventories for agricultural lands could be improved by including gas measurements from riparian buffer areas. However, little information is available regarding the dynamics of unintended emissions of soil gases in these commonplace features of agroecosystems primarily installed for water quality functions and how the dynamics compare to those for adjacent croplands. In order to understand the dynamics and drivers of N2O in riparian buffers and their immediately adjacent croplands, a meta-analysis of studies comparing N2O emissions and their drivers from riparian buffers and adjacent croplands was conducted. It was identified that while there is growing literature quantifying N2O emissions from different types of riparian buffer vegetation, comparative assessments of N2O emissions from riparian buffers and croplands remain limited. This was evident in the literature database summarizing data from 13 studies, with 44 data points from croplands: n = 22, and riparian buffers: n = 22, published between 1980 and 2021. The meta-analysis showed that the croplands generated significantly greater N2O emissions than the riparian buffers. In both the croplands and riparian buffers, emissions of N2O were mainly driven by elevation (p = 0.029), land use (p <0.0001), soil texture (p = 0.018), soil bulk density (BD) (p <0.0001), total carbon (C) (p <0.0001), C: N ratio (p <0.0001), and NH4+-N (p <0.0001). To further understand the dynamics of soil N2O, CH4, and CO2 fluxes from permanent pasture and maize (Zea mays L.) and contiguous riparian buffer strips with different (grass, willow, and woodland) vegetation as well control with no buffer vegetation, measurements on an existing experiment were carried out using the static chamber technique on a replicated plot-scale facility. For the experiment, including a permanent pasture, gas fluxes were measured periodically with soil and environmental variables between June 2018 and February 2019 at North Wyke, United Kingdom. During most of the sampling days, the no-buffer control treatment showed significantly (p <0.05) greater N2O fluxes and cumulative N2O emissions than the other treatments. The results also showed that the grass riparian buffer strip was a sink of N2O equivalent to -2310.2 g N ha-1 (95% confidence intervals: -535.5-492). Event-based water quality results obtained during storms showed that the willow riparian buffer treatment had the highest flow-weighted mean N concentrations (FWMC-N) of 0.041 ± 0.022 and 0.031 ± 0.015 mg N L-1 during the 12 November 2018 and 11 February 2019 storms, respectively, when compared to the other treatments. Soils under all treatments were sinks of soil CH4, with the willow riparian buffer strip (-2555 ± 318.7 g ha-1) being the largest soil CH4 sink followed by the grass riparian buffer (-2532 ± 318.7 g ha-1), woodland riparian buffer (-2318.0 ± 246.4 g ha-1), no-buffer strip control (-1938.0 ± 374.4 g ha-1), and lastly the upslope pasture (-1328.0 ± 89.0 g ha-1). Cumulative soil CO2 fluxes as affected by the treatments followed the descending order: woodland riparian buffer; 11927.8 ± 1987.9 kg ha-1 > no buffer control; 11101.3 ± 3700.4 kg ha-1 > grass riparian buffer; 10826.4 ± 2551.8 kg ha-1 > upslope pasture; 10554.6 ± 879.5 kg ha-1 > willow riparian buffer; 9294.9 ± 1549.2 5 kg ha-1. There was, however, no evidence of significant differences in cumulative CO2 amongst all treatments of the current study. Despite the lack of significant differences, the results showed that the woodland riparian buffer might emit the larger soil CO2 compared to the remainder of the other riparian buffers as well as the upslope pasture it serves. After the upslope area was converted from permanent pasture to maize (riparian buffers left untouched), the N2O, CH4, and CO2 measurements done between May and October 2019 showed that the no-buffer control generated the largest cumulative N2O emissions of 18 929 g N ha-1 (95% confidence intervals: 524.1 - 63 643) while the maize crop upslope generated the largest cumulative CH4 emissions of 5 050 ± 875 g ha-1. On the other hand, the woodland (322.9 ± 3.1 kg ha-1) and grass (285 ± 2.7 kg ha-1) riparian buffer treatments (not significant to each other) generated significantly (p =<0.0001) the largest CO2 compared to the rest of the treatments. A laboratory incubation experiment of the soils sourced from cropland, grass riparian buffer, willow riparian, and woodland riparian buffer using a specialized Denitrification System (DENIS) showed that the grass riparian buffer soils had the highest potential for emissions of NO (2.9 ± 0.31 mg m-2), and N2O (1413.4 ± 448.3 mg m-2), and the willow riparian buffer treatment showed the highest potential for N2 (698.1 ± 270.3 mg N m-2), and CO2 (27558.3 ± 128.9 mg m-2) emissions. The results on permanent pasture show that the grass riparian buffer may be an N2O sink, and on the other hand, the willow riparian buffer may be a larger sink for CH4 and emit lower CO2 when introduced for water quality protection measures in an ungrazed permanent pasture. On the other hand, the results of the fodder maize experiments suggest that fodder maize production in general, and situations where such cropping is not undertaken in tandem with a riparian buffer strip, result in atmospheric CH4 and N2O concerns. Also, the woodland and grass riparian buffers serving a maize crop may pose a CO2 threat. Lastly, the laboratory incubation results show that soils developed under the grass riparian buffer can potentially emit higher NO and N2O, while soils under willow riparian buffers can potentially emit larger N2 and CO2. The results of the study point to the need to take into consideration the disbenefits of greenhouse gas emissions by mitigation measures conventionally implemented for improving the sustainability of water resources. Further experiments are suggested particularly with different crops under varying soils and environmental settings to validate the findings of the current study.