Content uploaded by Fabien Charbonnier
Author content
All content in this area was uploaded by Fabien Charbonnier on May 20, 2014
Content may be subject to copyright.
A preview of the PDF is not available
Measuring and modeling light, water and carbon budgets and net primary productivity in a coffee-based agroforestry system of Costa Rica
Abstract
Compared to monocultures, agroforestry systems (AFS) are expected to provide enhanced resource-use efficiency and larger ecosystem services. However, due to the complexity of the interactions occurring in AFS, it is challenging to quantify and decompose the effects of shade trees on the main crop net primary productivity (NPP). Few process-based models are able to analyze the interactions between crop and shade trees for carbon and water. Interactions for light, water and energy occurring between tree and crops might have counterintuitive effects on photosynthesis, light use efficiency (LUE), transpiration efficiency and microclimate. We showed that a 3D process-based model, MAESPA, was able to quantitatively describe the spatial variability of those processes from the plant to the plot, and from hourly to yearly timescales. MAESPA simulated satisfactorily light interception in a 2-layer heterogeneous coffee AFS. It was used to produce powerful explanatory variables in AFS experiments and to analyze the determinants of coffee plant NPP. LUE displayed a 2-fold increase for shaded coffee plants totally compensating the expected decrease of local irradiance interception, and coffee plant ANPP was the same below shade trees or in the open. MAESPA also simulated satisfactorily carbon exchange at whole plant and plot scales, when compared to gas exchange records in a whole-plant chamber, or with eddy-covariance records above the canopy. We used MAESPA to simulate the spatial variability of photosynthesis and LUE. Overall, MAESPA proved to be a relevant model to quantify spatial interactions. The next very relevant development would be to couple it to a model of carbon allocation among organs in the coffee plants.
... As an example of how the required trade-offs might be made successfully, consider the simulation of heterogeneous stands of trees. Radiation interception and microclimate (i.e. the microclimate below, above or within the canopy and soil) are two key processes that must be carefully accounted for in those simulations (Charbonnier, 2013;Luedeling et al., 2016;Singh et al., 2012) because they become more heterogeneous as the canopy structure becomes more complex. Two-dimensional multi-layer models struggle to simulate the light interception of such ecosystems (Luedeling et al., 2016), which propagate into simulations errors of transpiration and photosynthesis. ...
... MASTRO, MAESTRA and then MAESPA), mainly for radiation and CO 2 fluxes. While the model showed excellent agreement between modeled and measured plant transpiration in an Oak forest (Hanson et al., 2004) under optimum soil water content, a native Eucalyptus forest (Medlyn et al., 2007) and a planted Eucalyptus stand (Christina et al., 2017;Christina et al., 2016), it has also been found to under-estimate high evapotranspiration rates on Coffea agroforestry systems (Charbonnier, 2013), on Pinus and Eucalyptus stands (Moreaux, 2012). Preliminary investigations suggested that the underestimation of evapotranspiration in these systems could occur due to unreliable estimation of canopy temperature. ...
... Preliminary investigations suggested that the underestimation of evapotranspiration in these systems could occur due to unreliable estimation of canopy temperature. Leaf temperatures were found to be underestimated by several degrees Celsius under high radiation and evapotranspiration conditions (Charbonnier, 2013). Modeled leaf temperature remained unrealistically close to air temperature within the canopy, itself remaining equal to the air temperature given as input to the model, generally taken from a meteorological station located outside the canopy. ...
... At the plant scale, coffee is an evergreen-broadleaved perennial plant, but its LAI varies seasonally (van Kanten and Vaast, 2006;Siles et al., 2010b) due to abiotic factors such as drought, shade, temperature (Matoso-Campanha et al., 2004;Righi et al., 2007), biological factors such as diseases and over-production (Avelino et al., 1991;Avelino et al., 2007;Lopez-Bravo et al., 2012), or management such as pruning and fertilizing. LAI was reported as a good indicator of coffee vigour (Charmetant et al., 2007), or of energy and gas exchanges (Charbonnier, 2013). At the plot scale and for various ecosystems, LAI was long reported to affect microclimate (Ong et al., 2000), evapotranspiration , hydrological services (Gómez-Delgado, 2010), erosion control (Ataroff and Monasterio, 1997), biomass and growth (Marsden et al., 2010;le Maire et al., 2011a), gross and net primary productivities (Gower et al., 1999;Roy et al., 2001;Beer et al., 2010). ...
... Apart from its impact on yield and on hydrological services which were presented here as examples, we stress that LAI is also a key factor for other ecosystem services such as Gross Primary Productivity (GPP or canopy photosynthesis), which is driving the net primary productivity and the Provisioning Services (Charbonnier, 2013). It is also likely that correlations could be evidenced between LAI and soil organic matter. ...
Scalable indicators are promising to assess ecosystem services. In a large (660 ha) coffee agroforestry farm, we calibrated the relationship between the Normalized Difference Vegetation Index (NDVI), calculated on a High Resolution (HR) satellite image and ground-truth LAI, providing a 2-layer (shade trees and coffee) LAI calibration with LAI 2000 and a new technique based on the cumulative distribution of LAI along transects. The effective and apparent clumping of coffee leaves were computed (0.76 and 0.89, respectively). We also calibrated the relationship between the derived HR-LAI farm map and NDVI from the Moderate Resolution Imaging Spectroradiometer (MODIS) in order to re-construct LAI time-series (2001–2011).
... Finally, we see how well LAI (observed and retrieved) correlates with other important crop development variables: canopy height and biomass. To further assess the performance of the studied LAI models against the measured LAI, the measured LAI data were divided into three groups: low (0-2), medium (2)(3)(4), and high (>4) LAI ranges. ...
Leaf area index (LAI) is an essential indicator of crop development and growth. For many agricultural applications, satellite-based LAI estimates at the farm-level often require near-daily imagery at medium to high spatial resolution. The combination of data from different ongoing satellite missions, Sentinel 2 (ESA) and Landsat 8 (NASA), provides this opportunity. In this study, we evaluated the leaf area index generated from three methods, namely, existing vegetation index (VI) relationships applied to Harmonized Landsat-8 and Sentinel-2 (HLS) surface reflectance produced by NASA, the SNAP biophysical model, and the THEIA L2A surface reflectance products from Sentinel-2. The intercomparison was conducted over the agricultural scheme in Bekaa (Lebanon) using a large set of in-field LAIs and other biophysical measurements collected in a wide variety of canopy structures during the 2018 and 2019 growing seasons. The major studied crops include herbs (e.g., cannabis: Cannabis sativa, mint: Mentha, and others), potato (Solanum tuberosum), and vegetables (e.g., bean: Phaseolus vulgaris, cabbage: Brassica oleracea, carrot: Daucus carota subsp. sativus, and others). Additionally, crop-specific height and above-ground biomass relationships with LAIs were investigated. Results show that of the empirical VI relationships tested, the EVI2-based HLS models statistically performed the best, specifically, the LAI models originally developed for wheat (RMSE:1.27), maize (RMSE:1.34), and row crops (RMSE:1.38). LAI derived through European Space Agency's (ESA) Sentinel Application Platform (SNAP) biophysical processor underestimated LAI and provided less accurate estimates (RMSE of 1.72). Additionally, the S2 SeLI LAI algorithm (from SNAP biophysical processor) produced an acceptable accuracy level compared to HLS-EVI2 models (RMSE of 1.38) but with significant underestimation at high LAI values. Our findings show that the LAI-VI relationship, in general, is crop-specific with both linear and non-linear regression forms. Among the examined indices, EVI2 outperformed other vegetation indices when all crops were combined, and therefore it can be identified as an index that is best suited for a unified algorithm for crops in semi-arid irrigated regions with heterogeneous landscapes. Furthermore, our analysis shows that the observed height-LAI relationship is crop-specific and essentially linear with an R 2 value of 0.82 for potato, 0.79 for wheat, and 0.50 for both cannabis and tobacco. The ability of the linear regression to estimate the fresh and dry above-ground biomass of potato from both observed height and LAI was reasonable, yielding R 2 :~0.60.
... What agroforestry research lacks desperately, in order to move beyond the classical dichotomy between shaded and non-shaded plots (Charbonnier et al., 2014), are maps of random variables inside heterogeneous agroforestry systems, for whatever tree spacing. For instance, the MAESPA 3D model (Duursma and Medlyn, 2012) has been applied in agroforestry and 2D horizontal maps have been proposed recently for the light distribution within the crop , for the crop's surface temperature (Vezy et al., 2018), for crop photosynthesis, transpiration, and water-use efficiency (Charbonnier, 2013), and for light-use-efficiency (LUE) (Charbonnier et al., 2017). 2D and 3D maps of root distribution have been proposed as well, with the uptake of water and nutrients (van Noordwijk and Lusiana, 1998;Dupraz et al., 2019). ...
The trees in agroforestry plots create spatial heterogeneity of high interest for adaptation, mitigation, and the provision of ecosystem services. But to what distance, exactly, from the tree? We tested a novel approach, based upon geostatistics and Unmanned Aerial Vehicle (UAV) sensing, to infer the distance at which a single agroforestry tree affects the surrounding under-crop, to map yield, litter (i.e. stover) and compute crop-partial Land Equivalent Ratio (LERcp) at the whole-plot level.
In an agro-silvo-pastoral parkland of semi-arid western Africa dominated by the multi-purpose tree Faidherbia albida, we harvested the pearl-millet under-crop at the whole-plot scale (ca. 1 ha) and also in subplot transects, at three distances from the trunks. We observed that the yield was three times higher below the tree crown (135.6 g m⁻²) than at a distance of five tree-crown radii from the trunk (47.7 g m⁻²). Through geostatistical analysis of multi-spectral, centimetric-resolution images obtained from an UAV overflight of the entire plot, we determined that the ‘Range’ parameter of the semi-variogram (assumed to be the distance of influence of the trees on the Normalized difference vegetation index (NDVI)) was 17 m. We correlated the yield (r² = 0.41; RRMSE = 48 %) and litter production (r² = 0.46; RRMSE = 35 %) in subplots with NDVI, and generated yield and litter maps at the whole-plot scale. The measured whole-plot yield (0.73 t ha⁻¹) differed from the one estimated via the UAV mapping by only 20 %, thereby validating the overall approach. The litter was estimated similarly at 1.05 tC ha⁻¹ yr⁻¹ and mapped. Using a geostatistical proxy for the sole crop, LERcp was estimated 1.16, despite the low tree density.
This new method to handle heterogeneity in agroforestry systems is a first application. We also propose strategies for extension to the landscape level.
The DynACof model was designed to model coffee agroforestry systems and study the trade-offs to e.g. optimize the system facing climate changes. The model simulates net primary productivity (NPP), growth, yield, mortality, energy and water balance of coffee agroforestry systems according to shade tree species and management. Several plot-scale ecosystem services are simulated by the model, such as production, canopy cooling effect, or potential C sequestration. DynACof uses metamodels derived from a detailed 3D process-based model (MAESPA) to account for complex spatial effects, while running fast. It also includes a coffee flower bud and fruit cohort module to better distribute fruit carbon demand over the year, a key feature to obtain a realistic competition between sinks.
The model was parameterized and evaluated using a highly comprehensive database on a coffee agroforestry experimental site in Costa Rica. The fluxes simulated by the model were close to the measurements over a 5-year period (nRMSE = 26.27 for gross primary productivity; 28.22 for actual evapo-transpiration, 53.91 for sensible heat flux and 15.26 for net radiation), and DynACof satisfactorily simulated the yield, NPP, mortality and carbon stock for each coffee organ type over a 35-year rotation.
The DynACof model was designed to model coffee agroforestry systems and study the trade-offs to e.g. optimize the system facing climate changes. The model simulates net primary productivity (NPP), growth, yield, mortality, energy and water balance of coffee agroforestry systems according to shade tree species and management. Several plot-scale ecosystem services are simulated by the model, such as production, canopy cooling effect, or potential C sequestration. DynACof uses metamodels derived from a detailed 3D process-based model (MAESPA) to account for complex spatial effects, while running fast. It also includes a coffee flower bud and fruit cohort module to better distribute fruit carbon demand over the year, a key feature to obtain a realistic competition between sinks.
We compared the model outputs with a highly comprehensive database on a coffee agroforestry farm in Costa Rica. The fluxes simulated by the model were close to the measurements over a 5-year period (RMSE= 1.60 gC m-2 d-1 for gross primary productivity; 0.63 mm d-1 for actual evapo-transpiration, 1.34 MJ m-2 d-1 for sensible heat flux and 1.88 MJ m-2 d-1 for net radiation), and DynACof satisfactorily simulated the yield, NPP, mortality and carbon stock for each coffee organ type over a 35-year rotation.
The preprint is archived on ZENODO (https://zenodo.org/record/3246268).
DynACof model website: https://vezy.github.io/DynACof
The profitability of hydropower is affected by soil erosion and sedimentation in dam reservoirs, which are influenced by land use, infiltration and aquifer interactions with surface water. In order to promote the payment of Hydrological Environmental Services (HES) in Costa Rica, a quantitative assessment of the impact of land uses (such as coffee farming) on the functioning of drainage basins is required. This thesis seeks: 1) to study the water balance partitioning in a newly installed coffee agroforestry basin 2) to estimate the water and sediment yield at various spatio-temporal scales: from plot to basin and from event to annual scale; and 3) to simulate the water and sediment yields, at both annual and peakflow scale, by including the surface runoff from hillslopes and roads. The main hydrological, ecophysiological and sediment processes were monitored during one year at the basin (rainfall, streamflow, evapotranspiration, soil humidity, aquifer level, turbidity) and at the plots (surface runoff and erosion). A new eco-hydrological model was developed to close the water balance, and the annual sediment yield was also quantified. Improvements are in progress to take into account the effect of roads in surface runoff generation. The low surface runoff, low plot erosion and low basin sediment yield observed under the current biophysical conditions (andisol) and management practices (no tillage, planted trees, bare soil kept by weeding), offer potential HES by reducing the superficial displacement capacity for fertilizers and pesticides, yielding low sediment loads and regulating streamflow variability through highly efficient mechanisms of aquifer recharge-discharge.
Distribution of organic material and nutrients was determined for leaves, branches, stems and roots of each species. Similar analyses were made for the litter layer and the mineral soil. -from English summary