ML Analysis of JWST Images

By -
ML Analysis of JWST Images

What is JWST?

The James Webb Space Telescope (JWST) is a highly anticipated and technologically advanced space observatory to revolutionize our understanding of the universe. Equipped with a massive segmented primary mirror, the JWST is designed to observe infrared light, allowing it to penetrate dust clouds and reveal hidden celestial objects and events. Its advanced suite of scientific instruments will enable scientists to study distant galaxies, star formation, exoplanets, and more, with remarkable clarity and sensitivity. The images produced by the JWST are expected to surpass those of its predecessor, offering intricate details and revealing the universe in a completely new light. 

By capturing stunning visuals of distant galaxies, stellar nurseries, and other cosmic wonders, these images will help astronomers unravel the mysteries of our universe, deepen our knowledge of its origins, and expand our understanding of the fundamental laws that govern it. Recently, researchers have applied AI/ML techniques to observe the images generated by JWST. The introduction of ML to the JWST image data is expected to uncover numerous new properties of galaxies and clusters.

Introduction and Background

The formation and evolution of disk galaxies within the current cosmological model framework, known as ΛCDM (Lambda Cold Dark Matter), is a complex puzzle in astrophysics. The hierarchical process of galaxy mass assembly during early cosmic epochs can disrupt dynamically cold disks. However, despite this disruptive nature, thin-disk galaxies are abundant in the local universe. Understanding the era when the first disk galaxies emerged is crucial for unraveling the development of galaxy morphology over cosmic time.

The researchers aim to identify distant disk galaxies within the JWST data by employing the Morpheus deep-learning framework. Previous observations of disk galaxy formation in the distant universe have revealed significant differences in the ionized gas dynamics compared to the local universe. Early-forming disks exhibited lower ratios of rotational velocity to velocity dispersion (V/σ) than present-day disk galaxies. Large, star-forming disk galaxies at redshifts around z ∼ 1-3 can be dominated by baryonic matter, with dark matter playing a relatively minor role in their rotation curves.

The new JWST data offer an opportunity to address unresolved questions raised by earlier studies. When did the first candidates for disk galaxies appear in the universe? What is the balance between gas and stellar mass in these galaxies, and how does it impact their formation? The remarkable capabilities of the JWST, including its spectroscopic capabilities, can significantly contribute to answering these questions by providing better insights into the stellar population properties of early disks and kinematic measures.

What is Morpheus Framework?

Morpheus is a deep learning framework developed for pixel-level analysis of astronomical image data. It is specifically designed to tackle challenges in the analysis of large-scale astronomical surveys. The primary goal of Morpheus is to automate the process of identifying and characterizing celestial objects in astronomical images. This includes tasks such as identifying galaxies, stars, and other astrophysical phenomena, as well as measuring their properties like positions, sizes, and shapes. Morpheus utilizes deep learning techniques, specifically convolutional neural networks (CNNs), to analyze the pixel-level information in astronomical images. 

CNNs are a type of artificial neural network that excels at processing grid-like data, such as images. By training on labeled data, Morpheus can learn to recognize different types of astronomical objects and perform automated analysis on large volumes of data. The framework provides a range of functionalities for image preprocessing, model training, and inference. It includes features for data augmentation, network architecture customization, and evaluation of model performance.

Applying Morpheus on JWST Images

The researchers perform a detailed pixel-level analysis of the JWST Cosmic Evolution Early Release Science Survey (CEERS) data in the Extended Groth Strip (EGS) using the Morpheus deep-learning model (H20). Without modification, the model is directly applied to the JWST F150W mosaic of the EGS. The researchers identify galaxies with photometric redshifts greater than 2 and exhibiting dominant disk morphologies based on Morpheus's classification. 

By fitting surface brightness models, they confirm that the majority of these candidates are structurally consistent with high-redshift disks. These results demonstrate the effectiveness of AI/ML methods like Morpheus in identifying potential disk galaxies at distant redshifts, which can then be further investigated through kinematic follow-up observations using spectroscopy.

The filter used for this study was F150W, as mentioned above. JWST F150W images refer to the observations captured by the James Webb Space Telescope (JWST) using its F150W filter. The F150W filter is one of the imaging filters available on the JWST's Near Infrared Camera (NIRCam), which is one of its scientific instruments. 

The F150W filter is designed to capture near-infrared light with a central wavelength of approximately 1.5 micrometers (hence the designation "F150W"). Near-infrared observations are valuable in studying various astronomical objects and phenomena, including distant galaxies, star-forming regions, exoplanets, and more. Near-infrared light can penetrate dust clouds more effectively than visible light, allowing astronomers to observe objects that may be obscured in other wavelengths.

Results

During the analysis of JWST NIRCam images using Morpheus, a set of selection criteria was applied to identify potential disk galaxy candidates. Out of the total catalog matches of 2507 objects, 202 were classified as disk galaxy candidates based on these criteria. These candidates had redshifts (z) greater than or equal to 2.

The average photometric redshift (〈z〉) of the selected sample was found to be 2.67. Within the sample, there were 42 disk candidates with redshifts above 3 (out of a total of 988 catalog matches), indicating they are at higher redshifts. Additionally, there were 10 disk candidates with redshifts greater than 4 (out of 353 catalog matches), representing even more distant objects.

The analysis identified 202 potential disk galaxy candidates in the JWST NIRCam images using Morpheus, out of which the average photometric redshift was 2.67. Among these candidates, 42 were at redshifts greater than 3, and 10 were at redshifts higher than 4. These findings contribute to the understanding of the presence and characteristics of disk galaxies at different redshifts in the early universe.

Conclusion

The findings of the study indicate that Morpheus successfully identifies galaxies as disks in a manner consistent with traditional methods. In fact, Morpheus demonstrates remarkable effectiveness in classifying galaxies as disk-like structures.

Moreover, the identification of high-redshift disk galaxy candidates in the JWST imagery suggests that disk galaxies might have formed at early cosmic epochs. This discovery challenges previous assumptions and contributes to our understanding of the formation and evolution of galaxies throughout cosmic history.

Comments

Post a Comment

Please do not use foul language while commenting. We will appreciate your suggestions and reviews in the comments.

Read Also

Deep Neural Networks for ADMET properties' prediction

By -
Image
Introduction The absorption, distribution, metabolism, excretion and toxicity properties of any compound contributes to its potential as an orally administered drug. A set of criteria defined by Rule of Five or the Pfizer's Rule of Five stats that any compound meeting these criteria are more likely to have ADMET properties. The importance of these properties has increased in modern medicinal chemistry and new machine learning techniques are employed to predict these properties. The predictive ADMET models have helped in discovery of small molecules with improved safety and dose. Deep Neural Networks are showing usefulness in such predictions due to improved computational efficiency, larger datasets and adaption of image processing in the chemistry. Early Models The early DNNs employed showed improved prediction performance. Many researchers raised concerned that this improvement is in fact has no relation with better computational capacity, rather it is result of mere memorizing of
Read more

Marine eDNA Analysis using DL techniques

By -
Image
Introduction Environmental DNA (eDNA) is the genetic information or material released by an organism into its surroundings through activities like skin shedding, mucus, urine or other secretions. These materials contain DNA fragments that can be found in soil, water or air samples. The concept of eDNA is based on the fact that organisms constantly leave organic traces behind them. By analysing the eDNA, researchers can detect the presence of any species in that area. This method is a valuable technique for biodiversity monitoring and ecological research.    Unlike traditional methods that involve direct observation or capturing of organisms, eDNA analysis allows researchers to study biodiversity without directly interacting with the species. This non-invasive approach is particularly useful for studying elusive or endangered species. eDNA analysis has been applied in various ecosystems, including freshwater, marine, and terrestrial environments. It is used to study aquatic organisms, t
Read more

Assisting Neuroimaging through DL

By -
Image
Introduction: In recent years, neuroimaging has been a research attraction, especially due to fMRI (Functional Magnetic Resonance Imaging) advancements. The fMRI data comprises of often several individuals. This data is used by researchers to study the association between the cognitive states of an individual and the underlying brain activities. The fMRI data used for neuroimaging is ideally suited for deep learning applications. The large, structured datasets of fMRI can be used for representation-learning methods of the DL. Generally, DL can be defined as a learning method with multiple levels of abstraction, where at each level the input data is transformed by a simple non-linear function which enables the model to recognize complex patterns. With higher-level representation, DL methods can associate a target variable with variable patterns in input data. Also, DL techniques can independently acquire these transformations from the data, eliminating the need for a comprehensive preex
Read more

What is Terraforming?

By -
Image
  Humans are working more and more on extraterrestrial habitations. A few decades back we were planning to settle on the Moon and now we are planning the same on Mars. Many researchers also proposed many theories on such habitations as O'Neill Cylinder, McKendree Cylinder, and many more. One such famous hypothetical theory is Terraforming. But what does the term terraforming mean? Terraforming is the process of transforming the atmosphere, topography, temperature and other factors of any moon or planet to make it habitable and similar to the Earth. It was Carl Sagan who first proposed the theory of terraforming. He proposed the terraforming of Venus. He thought of seeding Venus with algae. The algae will convert nitrogen and carbon dioxide to organic matter and eventually, the greenhouse effect will reduce to a comfortable, habitual level.  However, later studies revealed that the clouds of Venus are full of sulphuric acid. Also, the atmosphere of Venus is very thick, any carbon co
Read more