scholarly journals Areal Precipitation Coverage Ratio for Enhanced AI Modelling of Monthly Runoff: A New Satellite Data-Driven Scheme for Semi-Arid Mountainous Climate

2022 ◽  
Vol 14 (2) ◽  
pp. 270
Author(s):  
Seyyed Hasan Hosseini ◽  
Hossein Hashemi ◽  
Ahmad Fakheri Fard ◽  
Ronny Berndtsson
Keyword(s):  
Satellite Data ◽  
Precipitation Variability ◽  
Data Partitioning ◽  
Performance Criteria ◽  
Data Driven ◽  
Coverage Ratio ◽  
Areal Precipitation ◽  
Input Variables ◽  
Monthly Runoff ◽  
New Variables

Satellite remote sensing provides useful gridded data for the conceptual modelling of hydrological processes such as precipitation–runoff relationship. Structurally flexible and computationally advanced AI-assisted data-driven (DD) models foster these applications. However, without linking concepts between variables from many grids, the DD models can be too large to be calibrated efficiently. Therefore, effectively formulized, collective input variables and robust verification of the calibrated models are desired to leverage satellite data for the strategic DD modelling of catchment runoff. This study formulates new satellite-based input variables, namely, catchment- and event-specific areal precipitation coverage ratios (CCOVs and ECOVs, respectively) from the Global Precipitation Mission (GPM) and evaluates their usefulness for monthly runoff modelling from five mountainous Karkheh sub-catchments of 5000–43,000 km2 size in west Iran. Accordingly, 12 different input combinations from GPM and MODIS products were introduced to a generalized deep learning scheme using artificial neural networks (ANNs). Using an adjusted five-fold cross-validation process, 420 different ANN configurations per fold choice and 10 different random initial parameterizations per configuration were tested. Runoff estimates from five hybrid models, each an average of six top-ranked ANNs based on six statistical criteria in calibration, indicated obvious improvements for all sub-catchments using the new variables. Particularly, ECOVs were most efficient for the most challenging sub-catchment, Kashkan, having the highest spacetime precipitation variability. However, better performance criteria were found for sub-catchments with lower precipitation variability. The modelling performance for Kashkan indicated a higher dependency on data partitioning, suggesting that long-term data representativity is important for modelling reliability.

scholarly journals An Invariance Property of Predictors in Kernel-Induced Hypothesis Spaces

2006 ◽  
Vol 18 (4) ◽  
pp. 749-759 ◽  
Author(s):  
Nicola Ancona ◽  
Sebastiano Stramaglia
Keyword(s):  
Time Series ◽  
Radial Basis Function ◽  
Basis Function ◽  
Sufficient Conditions ◽  
Kernel Functions ◽  
Invariance Property ◽  
Risk Minimization ◽  
Independent Variables ◽  
Input Variables ◽  
New Variables

We consider kernel-based learning methods for regression and analyze what happens to the risk minimizer when new variables, statistically independent of input and target variables, are added to the set of input variables. This problem arises, for example, in the detection of causality relations between two time series. We find that the risk minimizer remains unchanged if we constrain the risk minimization to hypothesis spaces induced by suitable kernel functions. We show that not all kernel-induced hypothesis spaces enjoy this property. We present sufficient conditions ensuring that the risk minimizer does not change and show that they hold for inhomogeneous polynomial and gaussian radial basis function kernels. We also provide examples of kernel-induced hypothesis spaces whose risk minimizer changes if independent variables are added as input.

scholarly journals Estimation of Terrestrial Global Gross Primary Production (GPP) with Satellite Data-Driven Models and Eddy Covariance Flux Data

2018 ◽  
Vol 10 (9) ◽  
pp. 1346 ◽  
Author(s):  
Joanna Joiner ◽  
Yasuko Yoshida ◽  
Yao Zhang ◽  
Gregory Duveiller ◽  
Martin Jung ◽  
...  
Keyword(s):  
Primary Production ◽  
Eddy Covariance ◽  
Satellite Data ◽  
Gross Primary Production ◽  
Satellite Observation ◽  
Data Driven ◽  
Spatial And Temporal Variability ◽  
Light Use Efficiency ◽  
High Productivity ◽  
Use Efficiency

We estimate global terrestrial gross primary production (GPP) based on models that use satellite data within a simplified light-use efficiency framework that does not rely upon other meteorological inputs. Satellite-based geometry-adjusted reflectances are from the MODerate-resolution Imaging Spectroradiometer (MODIS) and provide information about vegetation structure and chlorophyll content at both high temporal (daily to monthly) and spatial (∼1 km) resolution. We use satellite-derived solar-induced fluorescence (SIF) to identify regions of high productivity crops and also evaluate the use of downscaled SIF to estimate GPP. We calibrate a set of our satellite-based models with GPP estimates from a subset of distributed eddy covariance flux towers (FLUXNET 2015). The results of the trained models are evaluated using an independent subset of FLUXNET 2015 GPP data. We show that variations in light-use efficiency (LUE) with incident PAR are important and can be easily incorporated into the models. Unlike many LUE-based models, our satellite-based GPP estimates do not use an explicit parameterization of LUE that reduces its value from the potential maximum under limiting conditions such as temperature and water stress. Even without the parameterized downward regulation, our simplified models are shown to perform as well as or better than state-of-the-art satellite data-driven products that incorporate such parameterizations. A significant fraction of both spatial and temporal variability in GPP across plant functional types can be accounted for using our satellite-based models. Our results provide an annual GPP value of ∼140 Pg C year - 1 for 2007 that is within the range of a compilation of observation-based, model, and hybrid results, but is higher than some previous satellite observation-based estimates.

Data-driven impact assessment of multidimensional project complexity on project performance

2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Abroon Qazi
Keyword(s):  
Hybrid Approach ◽  
Bayesian Belief Networks ◽  
Project Performance ◽  
Performance Criteria ◽  
Data Driven ◽  
Ann Model ◽  
Content Type ◽  
Performance Space ◽  
Project Complexity ◽  
And Performance

PurposeThe purpose of this paper is to propose a data-driven scheme for identifying critical project complexity dimensions and establishing the trade-off across multiple project performance criteria.Design/methodology/approachThis paper adopts a hybrid approach using Bayesian Belief Networks (BBNs) and Artificial Neural Networks (ANNs). The output of the ANN model is used as input to the BBN model for prioritizing project complexity dimensions relative to multiple project performance criteria. The proposed process is demonstrated through a real application in the construction industry.FindingsWith a number of nonlinear interactions involved within and across project complexity and performance, it is not feasible to model and assess the strength of these interactions using conventional techniques. The proposed process helps in effectively mapping a “multidimensional complexity” space to a “multidimensional performance” space and makes use of data from past projects for operationalizing this mapping scheme by means of ANNs. This obviates the need for developing a parametric model that is both challenging and computationally cumbersome. The mapping function can be used for generating all possible scenarios required for the development of a data-driven BBN model.Originality/valueThis paper introduces a data-driven process for operationalizing the mapping of project complexity to project performance within a network setting of interacting complexity dimensions and performance criteria. The results of the application study manifest the importance of capturing the interdependency across project complexity and performance. Ignoring the underlying interdependencies and relying exclusively on conventional correlation-based techniques may lead to making suboptimal decisions.

scholarly journals The Plant–Craig Stochastic Convection Scheme in ICON and Its Scale Adaptivity

2014 ◽  
Vol 71 (9) ◽  
pp. 3404-3415 ◽  
Author(s):  
Richard J. Keane ◽  
George C. Craig ◽  
Christian Keil ◽  
Günther Zängl
Keyword(s):  
High Resolution ◽  
Numerical Weather Prediction ◽  
Climate Models ◽  
Weather Prediction ◽  
Precipitation Amount ◽  
Precipitation Variability ◽  
Convection Scheme ◽  
Numerical Weather ◽  
Multiple Resolutions ◽  
Input Variables

Abstract The emergence of numerical weather prediction and climate models with multiple or variable resolutions requires that their parameterizations adapt correctly, with consistent increases in variability as resolution increases. In this study, the stochastic convection scheme of Plant and Craig is tested in the Icosahedral Nonhydrostatic GCM (ICON), which is planned to be used with multiple resolutions. The model is run in an aquaplanet configuration with horizontal resolutions of 160, 80, and 40 km, and frequency histograms of 6-h accumulated precipitation amount are compared. Precipitation variability is found to increase substantially at high resolution, in contrast to results using two reference deterministic schemes in which the distribution is approximately independent of resolution. The consistent scaling of the stochastic scheme with changing resolution is demonstrated by averaging the precipitation fields from the 40- and 80-km runs to the 160-km grid, showing that the variability is then the same as that obtained from the 160-km model run. It is shown that upscale averaging of the input variables for the convective closure is important for producing consistent variability at high resolution.

scholarly journals Data-Driven Optimization Framework for Nonlinear Model Predictive Control

2017 ◽  
Vol 2017 ◽  
pp. 1-15 ◽  
Author(s):  
Shiliang Zhang ◽  
Hui Cao ◽  
Yanbin Zhang ◽  
Lixin Jia ◽  
Zonglin Ye ◽  
...  
Keyword(s):  
Model Predictive Control ◽  
Predictive Control ◽  
Nonlinear Model ◽  
Learning Algorithm ◽  
Nonlinear Model Predictive Control ◽  
Data Driven ◽  
Optimization Framework ◽  
Linguistic Model ◽  
Input Variables ◽  
Control Accuracy

The structure of the optimization procedure may affect the control quality of nonlinear model predictive control (MPC). In this paper, a data-driven optimization framework for nonlinear MPC is proposed, where the linguistic model is employed as the prediction model. The linguistic model consists of a series of fuzzy rules, whose antecedents are the membership functions of the input variables and the consequents are the predicted output represented by linear combinations of the input variables. The linear properties of the consequents lead to a quadratic optimization framework without online linearisation, which has analytical solution in the calculation of control sequence. Both the parameters in the antecedents and the consequents are calculated by a hybrid-learning algorithm based on plant data, and the data-driven determination of the parameters leads to an optimization framework with optimized controller parameters, which could provide higher control accuracy. Experiments are conducted in the process control of biochemical continuous sterilization, and the performance of the proposed method is compared with those of the methods of MPC based on linear model, the nonlinear MPC with neural network approximator, and MPC nonlinear with successive linearisations. The experimental results verify that the proposed framework could achieve higher control accuracy.

Improvement of uncertainty estimation in global hydrological models by using high resolution satellite data as an interpolator

2020 ◽  
Author(s):  
Rogier Westerhoff ◽  
Frederika Mourot ◽  
Conny Tschritter
Keyword(s):  
High Resolution ◽  
Water Balance ◽  
Groundwater Recharge ◽  
Satellite Data ◽  
Spatial Scales ◽  
Remotely Sensed ◽  
Google Earth ◽  
Data Driven ◽  
Hydrological Models ◽  
Water Balance Components

<p>Global hydrological models often ingest remotely-sensed observations supported by ground-truthed data in attempts to better quantify water balance components, e.g. soil water content, evapotranspiration, runoff/discharge, groundwater recharge. However, the scaling up process from local observations to that global, coarse, scale contains large uncertainty, often undermining the relevance of water balance calculations.</p><p>With recent more advanced high-resolution satellite data, freely available at 10m spatial resolution and (sub-) weekly temporal resolution, there is a possibility to reduce uncertainty in that upscaling. However, there are two challenges in doing so when working with global models: exponential increase of computational effort, and the need for quantifying the yet unknown uncertainty of assumptions that coarse global model cells and their underlying equations bring.</p><p>This study hypothesises that a bottom-up approach with high-resolution satellite data and in situ observations will better constrain and quantify uncertainty. By using these more spatially-explicit data, we make the case that the estimation of water balance components should become more data-driven. We propose a more data-driven model that improves uncertainty of estimation and scalability by using more sophisticated, remotely-sensed interpolation layers.</p><p>Our case study shows New Zealand-wide estimates of evapotranspiration and groundwater recharge at two resolutions: 1km x 1km, using an earlier developed model and MODIS satellite data; and a novel approach at 10m x 10m using Sentinel-1 and Sentinel-2 data to better incorporate impervious areas (e.g., anthropogenic urbanised land covers) and small land patches (e.g., small forestry areas). We then study the implications of using different spatial scales and quantify the differences between 10m x 10m versus 1km x 1km model pixel estimates. Our method is applied in the Google Earth Engine, a free-for-research high performance cloud computing facility, hence providing powerful computational resources and making our approach easily scalable to global, regional and catchment scales. </p><p>Finally, we discuss what underlying model assumptions in traditional models could be changed to facilitate a method that works consistently at these different scales.</p>

Sign in / Sign up

Export Citation Format

Share Document