Agriculture contributes to climate changes through land use changes and greenhouse gas emissions. Models can provide crucial insights into the extent of this contribution; however, their effectiveness relies on proper evaluation within the application context. Moreover, crop models that can simulate greenhouse gas emissions have not been extensively tested for the semi-arid tropics. We calibrated and used the STICS soil-crop model to explore the skills of the model to reproduce observed variations in greenhouse gas emissions for a millet-groundnut rotation in an agro-silvo-pastoral parkland dominated by Faidherbia albida trees located in central Senegal. Model simulations were compared with observations of soil temperature, soil water content, N2O and CO2 emissions, aboveground and belowground biomass of millet and groundnut, collected between 2018 and 2022. CO2 emissions were simulated with a two-step approach. Initially, STICS simulated crop leaf area index and biomass (aboveground and belowground), and soil heterotrophic respiration. These variables were then integrated into an independent autotrophic respiration module, and summed with STICS simulated' heterotrophic respiration. In general, the STICS model tends to underestimate the observed minimum soil water content (wilting point) during the dry season and overestimate the observed soil water content after the wet season. However, the temporal dynamics of the soil temperature in the upper layer (0-30 cm) are generally well-represented by the model throughout the simulation period. Simulated N2O emissions were generally consistent in terms of magnitude compared to on-site measurements, although the model currently does not account for N2O absorption by the soil (i.e. negative fluxes). For instance, the simulated peak reached 0.041 kg N ha-1 d-1, while the observed peak was 0.048 kg N ha-1 d-1. The simulated average annual N2O emissions for the period 2018 to 2022 amounted to 0.368 kg N ha-1 yr-1. Simulated CO2 emissions were also comparable to on-site measurements (2021: EF = 0.63, BIAS = -0.75 kg C ha-1 d-1, and RMSE = 15.01 kg C ha-1 d-1; 2022: EF = 0.56, BIAS = -3.25 kg C ha-1 d-1, and RMSE = 5.01 kg C ha-1 d-1). These results indicate that the STICS model can be used to explore the impact of land use and crop management changes on greenhouse gas emissions in a tropical semi-arid context.

In agroforestry systems, fine roots grow at several depths due to the mixture of trees and annual crops. The decomposition of fine roots contributes to soil organic carbon stocks and may impact soil fertility, particularly in poor soils, such as those encountered in sub-Sahelian regions. The aim of our study was to measure the decomposition rate of root litter from annual and perennial species according to soil depth and location under and far from trees in a sub-Sahelian agroforestry parkland. Soil characteristics under and far from the trees were analysed from topsoil to 200 cm depth. Faidherbia tree, pearl millet and cowpea root litter samples were buried in litterbags for 15 months at 20, 40, 90 and 180 cm depths. Root litter decomposition was mainly impacted by soil moisture and soil depth. Faidherbia decomposed more slowly (36 +/- 12% remaining mass after 15 months) than cowpea and pearl millet roots (23 +/- 7% and 29 +/- 11% respectively). Pearl millet aboveground biomass, at harvesting time, was twice as high under (992 g m & 2) than far (433 g m & 2) from the tree, and belowground biomass (0-200 cm of depth) was 30.9 g m & 2 and 19.6 g m & 2 under and far from the tree, respectively. Faidherbia fine roots contributed slightly (p-value < 0.1) to higher stocks of C under the tree (7761 +/- 346 g m 2) than far from it (5425 +/- 558 g m & 2) and from 0 cm down to 200 cm depth.

The mitigation of climate change by agro-sylvo-pastoral systems is complex to assess or model, owing to high spatial and temporal heterogeneities. We set a new long-term observatory up for the monitoring and modelling of microclimate, GHG and deep SOC in a semi-arid agro-sylvo-pastoral system (Niakhar, Senegal), dominated by the multipurpose Faidherbia albida tree. Crops were mainly millet and peanut, under annual rotation. Transhumant livestock contributed largely to manure, SOM and soil fertility. Early 2018, we installed 3 eddy-covariance towers above (i) the whole mosaic, (ii) millet and (iii) peanut and monitored energy, CO2 balance and evapotranspiration for one full year. The mosaic ecosystem displayed low but significant CO2 and H2O fluxes during the dry season, owing to Faidherbia in leaf (Fig. 1). When rains resumed, the soil bursted a large amount of CO2. Just after the raising of millet, CO2 uptake by photosynthesis increased dramatically, then stabilized before harvest. However, this was compensated by large ecosystem respiration. The annual ecosystem CO2 balance was close to nil. This observatory is currently installing soil chambers for GHG fluxes, studying the horizontal variability of SOC by Vis-NIR and of deep soil roots and C using wells. Microclimate (land surface temperature, energy balance and gas exchanges) and light-use-efficiency will be mapped through 3D modelling (Charbonnier et al., 2017; Vezy et al., 2018). This observatory is open for collaboration.

The TTD10 method is an empirical evolution of the constant heat dissipation method of Granier (1985). By contrast to Granier, it uses a transient heating of 10 minutes, and it can be applied to a single-needle probe (Do et al., 2011). This system saves energy and cost and reduces thermal interference due to heat storage and passive thermal gradients. However, the heating duration increases the time resolution of measurement, i.e., 20 min with cooling, and assumes a relative stability of the flow rate over 10 min. Hence the objective was to reduce the heating time and measurement cycle. Previous multi-media laboratory data with both dual-needle and single-needle probes were used to test a generic calibration with the maximum temperature taken at 5 min. A calibration with similar sensitivity and accuracy to TTD10 was obtained under two conditions. The first condition used the difference between the maximum temperature and an offset temperature taken at 30 s. The second defined a new thermal index called K2. This new TTD5 system with single-needle probe enhances the previous advantages of low energy, low cost and simplicity of TTD method, well adapted to large sampling experiments.