Amplitude-Position Formulation of Data Assimilation

Ravela, Sai

doi:10.1007/11758532_66

Cited by 7 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some advances have been made in downscaling future projections to high spatial resolutions, e.g., using physics-based Markov Chain and Random Field models [73], but much remains to be done. Further, the data is sparse, and the uncertainty distributions remain poorly sampled [74], [75]. The heavy-tailed nature of extreme events such as cyclones and floods further exacerbates the challenges in producing their long-term forecasts.…”

Section: Long-term Forecasting Of Geoscience Variablesmentioning

confidence: 99%

Machine Learning for the Geosciences: Challenges and Opportunities

Karpatne

Ebert‐Uphoff

Ravela

et al. 2019

IEEE Trans. Knowl. Data Eng.

Self Cite

399

197

View full text Add to dashboard Cite

Geosciences is a field of great societal relevance that requires solutions to several urgent problems facing our humanity and the planet. As geosciences enters the era of big data, machine learning (ML)-that has been widely successful in commercial domains-offers immense potential to contribute to problems in geosciences. However, problems in geosciences have several unique challenges that are seldom found in traditional applications, requiring novel problem formulations and methodologies in machine learning. This article introduces researchers in the machine learning (ML) community to these challenges offered by geoscience problems and the opportunities that exist for advancing both machine learning and geosciences. We first highlight typical sources of geoscience data and describe their properties that make it challenging to use traditional machine learning techniques. We then describe some of the common categories of geoscience problems where machine learning can play a role, and discuss some of the existing efforts and promising directions for methodological development in machine learning. We conclude by discussing some of the emerging research themes in machine learning that are applicable across all problems in the geosciences, and the importance of a deep collaboration between machine learning and geosciences for synergistic advancements in both disciplines.

show abstract

Section: Long-term Forecasting Of Geoscience Variablesmentioning

confidence: 99%

Machine Learning for the Geosciences: Challenges and Opportunities

Karpatne

Ebert‐Uphoff

Ravela

et al. 2019

IEEE Trans. Knowl. Data Eng.

Self Cite

399

197

View full text Add to dashboard Cite

show abstract

“…It is a position adjustment technique 35,37 . It iteratively solves for the position error problem by minimizing an adjustment function based on gradient and divergence terms.…”

Section: Multi-resolution Viscous Alignmentmentioning

confidence: 99%

Untitled

Lowanichchai¹,

Weesakul²,

Chatdarong³

et al. 2010

ScienceAsia

View full text Add to dashboard Cite

ABSTRACT:The main objective of this paper is to improve the accuracy of radar rainfall estimation by accounting for a storm movement into a radar rainfall accumulation process. The multi-resolution viscous alignment (MVA) technique was used to estimate the velocity of a rain field from two consecutively measured radar images. The analysis used the 10-min radar reflectivity of the Pasicharoen radar and the corresponding 47 rain gauges measurements of 41 rainfall events that occurred in Bangkok during [2005][2006][2007]. The 28 rainfall events occurring during 2005-2006 were used for calibration, and the 13 rainfall events recorded in 2007 were used for validation. Finer temporal resolutions of radar reflectivity data, taken at 1-9 min intervals, were generated using the MVA technique in order to investigate the optimal temporal resolution of the Pasicharoen radar when the MVA technique was integrated into an hourly radar rainfall estimation algorithm to account for a storm movement within a sampling interval. The results showed that using the generated 5-min MVA reflectivity data for estimating hourly radar rainfall gave the smallest root mean square error (RMSE) between hourly radar rainfall estimates and corresponding rain gauge data when compared to other temporal resolutions of generated MVA reflectivity. Hourly radar rainfall obtained from the proposed algorithm, which integrates the MVA technique into the accumulation approach, was compared with the traditional simple linear interpolation (SLI) technique and conventional method. Using the 5-min generated MVA reflectivity data to estimate hourly radar rainfall can reduce RMSEs between hourly radar and rain gauge rainfall by 10% and 17% for the calibration period, and by 27% and 29% for the validation period when compared to the SLI and conventional methods, respectively.

show abstract

“…In front-tracking problems, they perform poorly when the simulated contour is displaced from its observation. They compensate position errors by adjusting amplitudes, which can produce non-physical solutions and do not preserve coherent structures [4,7,21,22,46]. Our objective is therefore to design a shape-oriented data assimilation approach that addresses position errors, in which the innovation measures the observation-simulation distance in terms of shape discrepancies.…”

Section: Introductionmentioning

confidence: 99%

Front Shape Similarity Measure for Shape-Oriented Sensitivity Analysis and Data Assimilation for Eikonal Equation

et al. 2018

View full text Add to dashboard Cite

We present a shape-oriented data assimilation strategy suitable for front-tracking problems through the example of wildfire. The concept of “front” is used to model, at regional scales, the burning area delimitation that moves, undergoes shape and topological changes under heterogeneous orography, biomass fuel and micrometeorology. The simulation-observation discrepancies are represented using a front shape similarity measure deriving from image processing and based on the Chan-Vese contour fitting functional. We show that consistent corrections of the front location and uncertain physical parameters can be obtained using this measure applied on a level-set fire growth model solving for an eikonal equation. This study involves a Luenberger observer for state estimation, including a topological gradient term to track multiple fronts, and of a reduced-order Kalman filter for joint parameter estimation. We also highlight the need – prior to parameter estimation – for sensitivity analysis based on the same discrepancy measure, and for instance using polynomial chaos metamodels, to ensure a meaningful inverse solution is achieved. The performance of the shape-oriented data assimilation strategy is assessed on a synthetic configuration subject to uncertainties in front initial position, near-surface wind magnitude and direction. The use of a robust front shape similarity measure paves the way toward the direct assimilation of infrared images and is a valuable asset in the perspective of data-driven wildfire modeling.

show abstract

Amplitude-Position Formulation of Data Assimilation

Cited by 7 publications

References 14 publications

Machine Learning for the Geosciences: Challenges and Opportunities

Machine Learning for the Geosciences: Challenges and Opportunities

Untitled

Front Shape Similarity Measure for Shape-Oriented Sensitivity Analysis and Data Assimilation for Eikonal Equation

Contact Info

Product

Resources

About