scholarly journals Joint Interpolation and Representation Learning for Irregularly Sampled Satellite-Derived Geophysical Fields

2021 ◽  
Vol 7 ◽  
Author(s):  
Ronan Fablet ◽  
Maxime Beauchamp ◽  
Lucas Drumetz ◽  
François Rousseau
Keyword(s):  
Representation Learning ◽  
Earth Observation ◽  
Optimal Interpolation ◽  
Satellite Orbits ◽  
Atmospheric Conditions ◽  
Geophysical Field ◽  
Geophysical Fields ◽  
Neural Network Learning ◽  
End To End ◽  
Underlying Processes

Earth observation satellite missions provide invaluable global observations of geophysical processes in play in the atmosphere and the oceans. Due to sensor technologies (e.g., infrared satellite sensors), atmospheric conditions (e.g., clouds and heavy rains), and satellite orbits (e.g., polar-orbiting satellites), satellite-derived observations often involve irregular space–time sampling patterns and large missing data rates. Given the current development of learning-based schemes for earth observation, the question naturally arises whether one might learn some representation of the underlying processes as well as solve interpolation issues directly from these observation datasets. In this article, we address these issues and introduce an end-to-end neural network learning scheme, which relies on an energy-based formulation of the interpolation problem. This scheme investigates different learning-based priors for the underlying geophysical field of interest. The end-to-end learning procedure jointly solves the reconstruction of gap-free fields and the training of the considered priors. Through different case studies, including observing system simulation experiments for sea surface geophysical fields, we demonstrate the relevance of the proposed framework compared with optimal interpolation and other state-of-the-art data-driven schemes. These experiments also support the relevance of energy-based representations learned to characterize the underlying processes.

scholarly journals END-TO-END PHYSICS-INFORMED REPRESENTATION LEARNING FOR SATELLITE OCEAN REMOTE SENSING DATA: APPLICATIONS TO SATELLITE ALTIMETRY AND SEA SURFACE CURRENTS

2021 ◽  
Vol V-3-2021 ◽  
pp. 295-302
Author(s):  
R. Fablet ◽  
M. M. Amar ◽  
Q. Febvre ◽  
M. Beauchamp ◽  
B. Chapron
Keyword(s):  
Remote Sensing ◽  
Satellite Altimetry ◽  
Remote Sensing Data ◽  
Representation Learning ◽  
Space Time ◽  
Satellite Orbits ◽  
Sensing Data ◽  
End To End ◽  
Ocean Remote Sensing ◽  
Learning Schemes

Abstract. This paper addresses physics-informed deep learning schemes for satellite ocean remote sensing data. Such observation datasets are characterized by the irregular space-time sampling of the ocean surface due to sensors’ characteristics and satellite orbits. With a focus on satellite altimetry, we show that end-to-end learning schemes based on variational formulations provide new means to explore and exploit such observation datasets. Through Observing System Simulation Experiments (OSSE) using numerical ocean simulations and real nadir and wide-swath altimeter sampling patterns, we demonstrate their relevance w.r.t. state-of-the-art and operational methods for space-time interpolation and short-term forecasting issues. We also stress and discuss how they could contribute to the design and calibration of ocean observing systems.

Unified, high resolution water quality retrievals from Earth Observation satellites

2021 ◽  
Author(s):  
James Harding
Keyword(s):  
Water Quality ◽  
Water Resource Management ◽  
State Of The Art ◽  
Earth Observation ◽  
Water Quality Parameter ◽  
Landsat 8 ◽  
Optical Parameters ◽  
Atmospheric Conditions ◽  
Mixture Density ◽  
Wide Range

<p>Earth Observation (EO) satellites are drawing considerable attention in areas of water resource management, given their potential to provide unprecedented information on the condition of aquatic ecosystems. Despite ocean colours long history; water quality parameter retrievals from shallow and inland waters remains a complex undertaking. Consistent, cross-mission retrievals of the primary optical parameters using state-of-the-art algorithms are limited by the added optical complexity of these waters. Less work has acknowledged their non- or weakly optical parameter counterparts. These can be more informative than their vivid counterparts, their potential covariance would be regionally specific. Here, we introduce a multi-input, multi-output Mixture Density Network (MDN), that largely outperforms existing algorithms when applied across different bio-optical regimes in shallow and inland water bodies. The model is trained and validated using a sizeable historical database in excess of 1,000,000 samples across 38 optical and non-optical parameters, spanning 20 years across 500 surface waters in Scotland. The single network learns to predict concurrently Chlorophyll-a, Colour, Turbidity, pH, Calcium, Total Phosphorous, Total Organic Carbon, Temperature, Dissolved Oxygen and Suspended Solids from real Landsat 7, Landsat 8, and Sentinel 2 spectra. The MDN is found to fully preserve the covariances of the optical and non-optical parameters, while known one-to-many mappings within the non-optical parameters are retained. Initial performance evaluations suggest significant improvements in Chl-a retrievals from existing state-of-the-art algorithms. MDNs characteristically provide a means of quantifying the noise variance around a prediction for a given input, now pertaining to real data under a wide range of atmospheric conditions. We find this to be informative for example in detecting outlier pixels such as clouds, and may similarly be used to guide or inform future work in academic or industrial contexts. </p>

Guessing the missing half of a geophysical field with blunt extension of discrete Universal Multifractal cascades

2021 ◽  
Author(s):  
Auguste Gires ◽  
Ioulia Tchiguirinskaia ◽  
Daniel Schertzer
Keyword(s):  
Time Series ◽  
Financial Support ◽  
Space Time ◽  
Rainfall Data ◽  
Second Step ◽  
Geophysical Field ◽  
Geophysical Fields ◽  
Wide Range ◽  
The Common ◽  
Partial Financial Support

<p>Universal Multifractals have been widely used to characterize and simulate geophysical fields extremely variable over a wide range of scales such as rainfall. Despite strong limitations, notably its non-stationnarity, discrete cascades are often used to simulate such fields. Recently, blunt cascades have been introduced in 1D and 2D to cope with this issue while remaining in the simple framework of discrete cascades. It basically consists in geometrically interpolating over moving windows the multiplicative increments at each cascade steps.</p><p> </p><p>In this paper, we first suggest an extension of this blunt cascades to space-time processes. Multifractal expected behaviour is theoretically established and numerically confirmed. In a second step, a methodology to address the common issue of guessing the missing half of a field is developed using this framework. It basically consists in reconstructing the increments of the known portion of the field, and then stochastically simulating the ones for the new portion, while ensuring the blunting the increments on the portion joining the two parts of the fields. The approach is tested with time series, maps and in a space-time framework. Initial tests with rainfall data are presented.</p><p> </p><p>Authors acknowledge the RW-Turb project (supported by the French National Research Agency - ANR-19-CE05-0022), for partial financial support.</p>

Sun-synchronous repeat ground tracks and other useful orbits for future space missions

2020 ◽  
Vol 124 (1276) ◽  
pp. 917-939 ◽  
Author(s):  
S.W. Paek ◽  
S. Kim ◽  
L. Kronig ◽  
O. de Weck
Keyword(s):  
Space Missions ◽  
Earth Observation ◽  
Satellite Orbits ◽  
Satellite Constellations ◽  
National Space ◽  
Future Space ◽  
Infrared Wavelengths ◽  
Design Requirements ◽  
Definition Of ◽  
The Individual

ABSTRACTThe development of oceanography and meteorology has greatly benefited from satellite-based data of Earth’s atmosphere and ocean. Traditional Earth observation missions have utilised Sun-synchronous orbits with repeat ground tracks due to their advantages in visible and infrared wavelengths. However, diversification of observation wavelengths and massive deployment of miniaturised satellites are both enabling and necessitating new kinds of space missions. This paper proposes several unconventional satellite orbits intended for use in, but not limited to, Earth observation. This ‘toolbox’ of orbits and taxonomy thereof will thus support the definition of design requirements for the individual satellites in nano-satellite constellations developed by national space agencies, industries and academia.

scholarly journals Multipath Cross Graph Convolution for Knowledge Representation Learning

2021 ◽  
Vol 2021 ◽  
pp. 1-13
Author(s):  
Luogeng Tian ◽  
Bailong Yang ◽  
Xinli Yin ◽  
Kai Kang ◽  
Jing Wu
Keyword(s):  
Representation Learning ◽  
Knowledge Graph ◽  
Baseline Model ◽  
Convolutional Network ◽  
Redundant Feature ◽  
Feature Information ◽  
Semantic Spaces ◽  
Local Graph ◽  
End To End ◽  
Representation Method

In the past, most of the entity prediction methods based on embedding lacked the training of local core relationships, resulting in a deficiency in the end-to-end training. Aiming at this problem, we propose an end-to-end knowledge graph embedding representation method. It involves local graph convolution and global cross learning in this paper, which is called the TransC graph convolutional network (TransC-GCN). Firstly, multiple local semantic spaces are divided according to the largest neighbor. Secondly, a translation model is used to map the local entities and relationships into a cross vector, which serves as the input of GCN. Thirdly, through training and learning of local semantic relations, the best entities and strongest relations are found. The optimal entity relation combination ranking is obtained by evaluating the posterior loss function based on the mutual information entropy. Experiments show that this paper can obtain local entity feature information more accurately through the convolution operation of the lightweight convolutional neural network. Also, the maximum pooling operation helps to grasp the strong signal on the local feature, thereby avoiding the globally redundant feature. Compared with the mainstream triad prediction baseline model, the proposed algorithm can effectively reduce the computational complexity while achieving strong robustness. It also increases the inference accuracy of entities and relations by 8.1% and 4.4%, respectively. In short, this new method can not only effectively extract the local nodes and relationship features of the knowledge graph but also satisfy the requirements of multilayer penetration and relationship derivation of a knowledge graph.

Sign in / Sign up

Export Citation Format

Share Document