Soil moisture and gaseous N-flux (N2O, N-2) dynamics in Costa Rican coffee plantations were successively simulated using a mechanistic model (PASTIS) and two process-based models (NGAS and NOE). Two fertilized (250 kg N ha(-1) y(-1)) coffee plantations were considered, namely a monoculture and a system shaded by the N-2 fixing legume species Inga densiflora. In situ N2O fluxes were previously measured in these plantations. NGAS and NOE used specific microbial activities for the soils. To parameterize NGAS, we estimated N mineralization via in situ incubations and the contribution of heterotrophic soil respiration to total soil respiration. Potential denitrification rates and the proportion of denitrified N emitted as N2O were measured in the laboratory to define the values of NOE parameters, as well as nitrification rates and related N2O production rates for parameterizing both models. Soil moisture and both NGAS and NOE N2O fluxes were best modelled on an hourly time step. Soil moisture dynamics were satisfactorily simulated by PASTIS. Simulated N2O fluxes by both NGAS and NOE (3.2 and 2.1 kg N ha(-1) y(-1) for NGAS; 7.1 and 3.7 kg N ha(-1) y(-1) for NOE, for the monoculture and shaded plantations respectively) were within a factor of about 2 of the observed annual fluxes (4.3 and 5.8 kg N ha(-1) y(-1), for the monoculture and shaded plantations respectively). Statistical indicators of association and coincidence between simulated and measured values were satisfactory for both models. Nevertheless, the two models differed greatly in describing the nitrification and denitrification processes. Some of the algorithms in the model NGAS were apparently not applicable to these tropical acidic Andosols. Therefore, more detailed information about microbial processes in different agroecosystems would be needed, notably if process-oriented models were to be used for testing strategies for mitigating N2O emissions. (C) 2009 Elsevier Ltd. All rights reserved.
Tag: algorithms
Ensemble Modeling on Near-Infrared Spectra as Rapid Tool for Assessment of Soil Health Indicators for Sustainable Food Production Systems
A novel total ensemble (TE) algorithm was developed and compared with random forest optimization (RFO), gradient boosted machines (GBM), partial least squares (PLS), Cubist and Bayesian additive regression tree (BART) algorithms to predict numerous soil health indicators in soils with diverse climate-smart land uses at different soil depths. The study investigated how land-use practices affect several soil health indicators. Good predictions using the ensemble method were obtained for total carbon (R2 = 0.87; RMSE = 0.39; RPIQ = 1.36 and RPD = 1.51), total nitrogen (R2 = 0.82; RMSE = 0.03; RPIQ = 2.00 and RPD = 1.60), and exchangeable bases, m3. Cu, m3. Fe, m3. B, m3. Mn, exchangeable Na, Ca (R2 > 0.70). The performances of algorithms were in order of TE > Cubist > BART > PLS > GBM > RFO. Soil properties differed significantly among land uses and between soil depths. In Kenya, however, soil pH was not significant, except at depths of 45–100 cm, while the Fe levels in Tanzanian grassland were significantly high at all depths. Ugandan agroforestry had a substantially high concentration of ExCa at 0–15 cm. The total ensemble method showed better predictions as compared to other algorithms. Climate-smart land-use practices to preserve soil quality can be adopted for sustainable food production systems.