Kooperativer Bibliotheksverbund

UID:

Format: 1 online resource (552 pages)

Edition: 1st ed.

ISBN: 0-443-13292-5

Note: Intro -- Intelligence Systems for Earth, Environmental and Planetary Sciences -- Copyright -- Dedication -- Contents -- Contributors -- Acknowledgments -- Chapter 1: Partial least squares regression to explore and predict environmental data -- 1. Introduction -- 1.1. The basis of PLS: Understanding the fundamentals -- 1.2. Examples of PLSR application in environmental science: Demonstrating its relevance and effectiveness -- 1.3. State-of-the-art algorithms for improved PLS analysis: Enhancing accuracy and efficiency -- 1.3.1. NIPALS for principal component analysis -- 1.3.2. NIPALS for PLS -- 1.3.3. Variants of NIPALS -- 1.3.4. Selecting the number of latent variables to include -- 2. PLSR case study: A step-by-step illustration of applying PLSR to environmental data using R -- 2.1. Step 1: Install and load necessary packages -- 2.2. Step 2: Load the dataset -- 2.3. Step 3: Join and select the data in the dataset -- 2.4. Step 4: Preprocess the data -- 2.5. Step 5: Split the data into training and testing datasets -- 2.6. Step 6: Train the PLS model -- 2.7. Step 7: Make predictions on the test dataset -- 2.8. Step 8: Evaluate the model performance -- 2.9. Step 9: Visualize the results -- 3. Key practical considerations for PLS applications -- 3.1. Data quality -- 3.2. Variables selection -- 3.3. Assessment of variable importance -- 3.4. Retrospective analysis of PLSR performance for the case study -- 4. Future perspectives with PLSR -- Acknowledgment -- References -- Chapter 2: Study of solute dynamics in unsaturated sandy soil under controlled irrigation -- 1. Introduction -- 2. Materials and method -- 2.1. Data -- 2.2. The experiment -- 2.3. Climate scenarios -- 2.4. Convective lognormal transfer function -- 2.4.1. Estimation of the parameters of the model -- 2.5. Statistical method -- 2.6. Hydrus 2D -- 2.6.1. Flow and transport equations. , 2.6.2. Model parameters -- 2.6.3. Initial and boundary conditions of the model -- 2.7. Evaluation of models performance -- 3. Results and discussion -- 3.1. Measured data -- 3.2. Convective lognormal transfer function -- 3.2.1. Parameters estimated by the model and elution curves -- 3.3. Hydrus 2D -- 3.3.1. Calibration of soil hydrodynamic parameters -- 3.3.2. Simulation of bromide concentration by the Hydrus-2D model for the different climate scenarios -- 4. Conclusion -- Appendix -- References -- Chapter 3: Prediction of hydraulic heights water table for irrigation management in cranberry fields with Random Forest: ... -- 1. Introduction -- 2. Methodology -- 2.1. Study area -- 2.2. Random Forest model -- 2.3. Data -- 2.3.1. Model development -- Data splitting -- Training and test subsets -- Input selection -- Autoregressive values of the training dataset -- Autoregressive values of the test dataset -- 2.3.2. Model calibration and validation -- 2.4. Importance of input variables -- 2.5. Performance criteria -- 3. Results and discussion -- 3.1. Simulation of water table heights -- 3.2. Predictive variables relative importance -- 3.3. Forecasting water table heights -- 4. Conclusion -- References -- Chapter 4: Spatial intelligence in AI applications for assessing soil health to monitor farming systems and associated ES ... -- 1. Introduction -- 2. Data and methodology -- 2.1. Study area -- 2.2. Description of the dataset -- 2.3. Soil erosion hazard inventory preparation -- 2.4. Factors causing soil erosion hazards -- 2.4.1. Soil erodibility -- 2.4.2. Slope length and steepness -- 2.4.3. Land-use and land-cover -- 2.4.4. Rainfall erosivity -- 2.4.5. Aspect -- 2.4.6. Distance to stream -- 2.4.7. Stream power index -- 2.5. Climate variables and CMIP6 model simulation -- 2.6. Delineation of farming systems. , 2.7. AI tools for soil erosion susceptibility modeling -- 2.7.1. Artificial neural network -- 2.7.2. Adaptive neuro-fuzzy inference system -- 2.7.3. Support vector machine -- 2.7.4. Random forest -- 2.7.5. Combined long short-term memory and revised universal soil loss equation model -- 2.8. Performance analysis of the models -- 3. Results -- 3.1. Land-use and land-cover map and farming system delineation -- 3.2. Soil erosion factor map generation -- 3.3. Rainfall erosivity with climate change scenarios -- 3.4. Soil erosion susceptibility with AI tools -- 3.5. Selection of important variables for soil erosion hazards -- 3.6. Model validation for susceptibility maps -- 4. Discussion -- Conclusion -- Funding -- Author contributions -- Declaration of competing interest -- References -- Chapter 5: Drought forecasting based on machine learning techniques -- 1. Introduction -- 2. Background -- 2.1. Drought -- 2.2. Difference between drought and aridity -- 3. Frequency, severity, and areal extend of drought -- 3.1. Frequency of drought -- 3.2. Severity of drought -- 3.3. Areal extent of drought -- 4. Return period and indexes of drought -- 4.1. Return period -- 4.2. Different form of drought and their indexes -- 4.3. Indices of drought -- 4.3.1. SPI -- 4.3.2. SPEI -- 4.3.3. PDSI -- 4.3.4. SDI -- 4.3.5. SMAI -- 4.3.6. SMI -- 4.3.7. SWSDI -- 4.3.8. MSDI -- 4.4. Using AI methods for drought forecasting -- 4.4.1. Extract dry years -- 4.4.2. Using AI for drought forecasting -- 5. Conclusion -- References -- Chapter 6: How to use artificial intelligence to downscale climate change models data -- 1. Introduction -- 2. Background -- 2.1. Global circulation models -- 2.2. Climate sensitivity balance -- 3. The coupled model intercomparison project -- 3.1. Introduction to CMIP -- 3.2. Goals of CMIP -- 3.3. CMIP5 series -- 3.4. CMIP6 series. , 3.5. Representative concentration pathways and shared socioeconomic pathways -- 3.5.1. Representative concentration pathways -- 3.5.2. Shared socioeconomic pathways -- 3.6. Difference between CMIP5 & -- CMIP6 -- 3.6.1. Socioeconomic scenarios -- 3.6.2. Climate model improvements -- 3.6.3. Earth system components -- 3.6.4. Aerosol and chemistry representation -- 3.6.5. Data accessibility and intercomparison -- 3.6.6. Increased international collaboration -- 3.7. Downscaling -- 3.8. Statistical downscaling -- 3.9. Dynamical downscaling -- 3.10. Bias correction -- 3.11. Weather generator -- 3.12. Ensemble techniques -- 4. How to use -- 4.1. Data preparation -- 4.2. Model training -- 4.3. Model evaluation and validation -- 4.4. Downscaling process -- 4.5. Uncertainty estimation -- 4.5.1. Ensemble modeling -- 4.5.2. Bootstrap resampling -- 4.6. Model refinement and iteration -- 5. Conclusion -- References -- Chapter 7: Advanced methods of soil quality assessment for sustainable agriculture -- Abbreviations -- 1. Introduction -- 2. The concepts and applications of soil quality -- 3. Soil quality datasets -- 3.1. Soil quality indicators -- 3.1.1. Physical indicators -- 3.1.2. Chemical indicators -- 3.1.3. Biological indicators -- 3.2. Minimum dataset -- 4. Soil quality assessment methods -- 4.1. Farmer-based assessment of soil quality scorecards and soil quality test kits -- 4.2. Visual assessment -- 4.3. Soil quality indices -- 4.4. Advanced methods for soil quality assessment -- 4.4.1. Geospatial approach -- 4.4.2. Soil quality modeling -- 4.4.3. Digital soil mapping and machine learning -- 5. Radio isotope technique to assess soil quality -- 5.1. Fallout radionuclides-137Cs as a soil quality indicator -- 5.2. 137Cs to study erosion-induced soil carbon dynamics -- 6. Conclusion -- Acknowledgments -- References. , Chapter 8: Combining the RUSLE approach and GIS tools in soil water erosion monitoring and mapping (Northeastern Algeria) -- 1. Introduction -- 2. Location and geological setting -- 3. Materials and methods -- 3.1. Data source -- 3.2. Method -- 4. Results and discussion -- 4.1. Rainfall erosivity factor (R) -- 4.2. Soil erodibility (K) -- 4.3. Topographic factor (LS) -- 4.4. Land-use factor-plant cover (C) -- 4.5. The anti-erosion practical factor (P) -- 5. Conclusions -- Acknowledgments -- References -- Chapter 9: Leveraging the use of mechanistic and machine learning models to assess interactions between ammonia concentra ... -- 1. Introduction -- 2. Materials and methods -- 2.1. Data collection and experimental setup -- 2.2. Modeling frameworks -- 2.2.1. Mechanistic models -- Equilibrium gas phase NH3 concentration in laying-hen manure -- Carbon-dioxide accelerated ammonia release model -- Mechanistic NH3 emission model -- 2.2.2. Data-driven model: Random forest -- 2.3. Model performance metrics -- 2.4. Importance features -- 3. Results -- 3.1. Laying-hen manure characteristics -- 3.2. Hygrothermal conditions -- 3.3. CO2, CH4, and H2O gas concentrations -- 3.4. Observed NH3-N concentrations -- 3.5. Simulation of NH3 concentrations -- 3.5.1. Mechanistic NH3 emission model -- 3.5.2. Random forest -- Optimal model parameters -- NH3-N predictions -- 3.6. Importance features -- 4. Discussion -- 4.1. Properties of LHM samples -- 4.2. Estimation of NH3-N concentrations -- 4.3. Interactions between manure characteristics, atmospheric conditions, and simulated NH3 concentrations -- 5. Conclusions -- Appendix A: Carbon-dioxide accelerated ammonia release model -- A.1. Diffusion mass transfer module -- A.2. Chemistry of the ammonia at the manure surface -- A.3. Convection mass transfer module -- A.4. State space model. , Appendix B: Mechanistic NH3 emission model.

Additional Edition: ISBN 0-443-13293-3

Language: English