About

Arthur Lee is the Chair of the Department of Computer Science at SUNY Korea. He received his BS degree from the University of Utah, MS from Stanford University, and PhD from the University of Utah, all in Computer Science. He has held academic appointments at Korea University for seven years, including two years as Chair, at the University of Utah for five years, and at Claremont McKenna College for twelve years, including two years as Chair. His professional experience also includes over 15 years of work in research labs and industry, such as Sandia National Laboratories, Xerox Palo Alto Research Center, Evans & Sutherland Computer Corp., and SAP Labs. His research interests encompass programming languages, database systems, distributed systems, and data science, with particular focus on object-oriented and dynamic languages, main-memory database management systems for big data in distributed environments, and data management supporting data science needs. Arthur Lee's research has been sponsored by government agencies and industry in Korea and the USA. He joined SUNY Korea in 2016 and has contributed significantly to the academic community through his research, teaching, and leadership roles.

Research topics

• Astrophysics
• Astronomy
• Physics

Selected publications

• ALMA-FACTS: Variations in CO(2-1)/CO(1-0) Across Twelve Nearby Spiral Galaxies

  Open MIND · 2026-02-03

  article1st authorCorresponding

  We will present early results from a high-resolution analysis of the CO(2-1)/CO(1-0) line ratio, R21, across the disks of twelve nearby spiral galaxies. We use new ALMA CO(1-0) observations from the Fundamental CO(1-0) Transition Survey (FACTS), and re-image CO(2-1) data from PHANGS. Higher ratios suggest higher densities and temperatures. The sample includes barred and unbarred, and flocculent galaxies. We find that R21 varies systematically as a function of structure. The barred spiral galaxies follow a general trend: R21 is high in the center, low along the bar, increases at the bar ends, and then declines towards the outskirts of the disk. R21 in unbarred galaxies fluctuate along the spiral arms. The barred galaxies also have spiral arms beyond the bar, which show similar fluctuations. The gas around HII regions typically have higher ratios. These results show that R21 varies systematically as a function of galactic structure and star formation activity, which should be taken into account when we interpret the relation between gas and star formation.

  DOI

• ALMA-FACTS: Variations in CO(2-1)/CO(1-0) Across Twelve Nearby Spiral Galaxies

  Zenodo (CERN European Organization for Nuclear Research) · 2026-02-03

  articleOpen access1st authorCorresponding

  We will present early results from a high-resolution analysis of the CO(2-1)/CO(1-0) line ratio, R21, across the disks of twelve nearby spiral galaxies. We use new ALMA CO(1-0) observations from the Fundamental CO(1-0) Transition Survey (FACTS), and re-image CO(2-1) data from PHANGS. Higher ratios suggest higher densities and temperatures. The sample includes barred and unbarred, and flocculent galaxies. We find that R21 varies systematically as a function of structure. The barred spiral galaxies follow a general trend: R21 is high in the center, low along the bar, increases at the bar ends, and then declines towards the outskirts of the disk. R21 in unbarred galaxies fluctuate along the spiral arms. The barred galaxies also have spiral arms beyond the bar, which show similar fluctuations. The gas around HII regions typically have higher ratios. These results show that R21 varies systematically as a function of galactic structure and star formation activity, which should be taken into account when we interpret the relation between gas and star formation.

  PublisherDOI

• ALMA FACTS. III. High-resolution CO(2–1)/CO(1–0) Maps of Twelve Nearby Galaxies

  The Astrophysical Journal · 2026-04-02

  articleOpen access1st authorCorresponding

  Abstract We present early results from a high-resolution analysis (∼100–200 pc) of the CO(2–1)/CO(1–0) line ratio in 12 nearby galaxies. We use new Atacama Large Millimeter/submillimeter Array (ALMA) CO(1–0) observations from the Fundamental CO(1–0) Transition Survey (FACTS), and re-imaged CO(2–1) data from PHANGS. We make empirical classifications based on the optical and molecular gas morphologies, which show clear systematic trends in the variation of R 21 as a function of galactic structure. The sample includes barred, unbarred, and flocculent galaxies. The barred spiral galaxies follow a general trend when the gas exists significantly: R 21 is high in the center, low along the bar, increases at the bar ends, and then lowers beyond the bar end or flattens in the outer parts of the disk. The structure dependence suggests the importance of galactic dynamics on molecular gas evolution, and consequently on star formation, in galaxies. R 21 fluctuates in the spiral arms for both barred and unbarred galaxies. Areas around HII regions in some cases appear to show more high-ratio gas. Together, R 21 varies systematically as a function of galactic structure, dynamics, and star formation activity.

  PublisherDOI

• ALMA FACTS. II. Large Scale Variations in the ¹²CO(J = 2 – 1) to ¹²CO(J = 1 – 0) Line Ratio in Nearby Galaxies

  The Astrophysical Journal · 2025-02-07 · 3 citations

  articleOpen accessSenior author

  Abstract We present 12 CO( J = 1 – 0) mapping observations over ∼1/2 of the optical disk of 12 nearby galaxies from the Fundamental CO 1–0 Transition Survey of nearby galaxies (FACTS), using the ALMA Total Power array. Variations in the 12 CO( J = 2 − 1)/ 12 CO( J = 1 − 0) line ratio r 21 are investigated. The luminosity-weighted r 21 of the 11 sample galaxies ranges from 0.52 to 0.69 with an average of 0.61. We use position–velocity diagrams along the major axis and tilted ring models to separate the normal rotating galactic disk from kinematic outliers that deviate from pure circular rotation. We find that r 21 is systematically higher in outliers compared to the disk. We compare r 21 between SA, SAB, and SB galaxies, and find no significant difference in the average r 21 depending on the presence of galactic bars. We find, however, that the radial gradient in r 21 is bimodal, where a group containing all SA galaxies prefer constant or very shallow r 21 gradients out to 40% of the optical radius, while another group containing all SB galaxies have a steep r 21 gradient, decreasing by ∼20% before 40% of the optical radius, which also corresponds to the radius of the stellar bar. After this radius, these galaxies become consistent with a constant or shallow trend in r 21 . The large scale trend in r 21 can have implications for how we interpret observations made solely in the 12 CO( J = 2 − 1) line.

  PublisherDOI

• Dynamically-Driven Evolution of Molecular Gas in M83 Traced by CO 2-1/1-0 Line Ratio Variations

  ArXiv.org · 2025-05-13

  preprintOpen access

  We show the variations of the CO J=2-1/1-0 line ratio (R21) across the barred spiral galaxy M83, using the 46 pc resolution data from ALMA. The R21 map clearly evidences the systematic large-scale variations as a function of galactic structures. Azimuthally, it starts from low R21<~0.7 in the interarm regions and becomes high ~>0.7 in the bar and spiral arms, suggesting that the density and/or kinetic temperature of molecular gas increase by about a factor of 2-3. This evolution is seen even in the parts of spiral arms without star formation, and R21 is often elevated even higher to ~0.8-1.0 when HII regions exist in the vicinity. Radially, R21 starts very high >~1.0 at the galactic center, remains low <~0.7 in the bar region, increases to >~0.7 around the bar end, and again decreases to <~0.7 in the rest of disk where the spiral arms dominate. The evolutionary sequence is synchronized with galactic rotation, and therefore, it is determined largely by the galactic structures and dynamics and is governed by the galactic rotation timescales. The R21 map also shows that the influence of stellar feedback is localized and limited. Massive, large, and non-star forming molecular structures have low R21, which also suggests that the bulk molecular gas in the disk is not regulated by stellar feedback, but more likely by galactic structures and dynamics. These results are consistent with suggestions by the earlier studies of the Milky Way and other barred spiral galaxies, and thus, are likely general among barred spiral galaxies in the local Universe.

  PublisherOA PDFDOI

• Dynamically Driven Evolution of Molecular Gas in the Barred Spiral Galaxy M83 Traced by CO <i>J</i> = 2–1/1–0 Line Ratio Variations

  The Astrophysical Journal · 2025-06-04 · 2 citations

  articleOpen access

  Abstract We show the variations of the CO J = 2–1/1–0 line ratio ( R 21/10 ) across the barred spiral galaxy M83, using the 46 pc resolution data from the Atacama Large Millimeter/submillimeter Array. The R 21/10 map clearly evidences the systematic large-scale variations as a function of galactic structures. Azimuthally, it starts from low R 21/10 ≲ 0.7 in the interarm regions and becomes high ≳0.7 in the bar and spiral arms, suggesting that the density and/or kinetic temperature of molecular gas increase by about a factor of 2–3. This evolution is seen even in the parts of spiral arms without star formation, and R 21/10 is often elevated even higher to ∼0.8–1.0 when H ii regions exist in the vicinity. Radially, R 21/10 starts very high ≳1.0 at the galactic center, remains low ≲0.7 in the bar region, increases to ≳0.7 around the bar end, and again decreases to ≲0.7 in the rest of disk where the spiral arms dominate. The evolutionary sequence is synchronized with galactic rotation, and therefore, it is determined largely by the galactic structures and dynamics and is governed by the galactic rotation timescales. The R 21/10 map also shows that the influence of stellar feedback is localized and limited. Massive, large, and non-star-forming molecular structures have low R 21/10 , which also suggests that the bulk molecular gas in the disk is not regulated by stellar feedback, but more likely by galactic structures and dynamics. These results are consistent with suggestions by the earlier studies of the Milky Way and other barred spiral galaxies, and thus, are likely general among barred spiral galaxies in the local Universe.

  PublisherDOI

• The CO-to-H$_2$ Conversion Factor in the Barred Spiral Galaxy M83

  arXiv (Cornell University) · 2024-04-22

  preprintOpen access1st authorCorresponding

  We analyze the CO-to-H$_2$ conversion factor ($α_{\rm{CO}}$) in the nearby barred spiral galaxy M83. We present new HI observations from the JVLA and single-dish GBT in the disk of the galaxy, and combine them with maps of CO(1-0) integrated intensity and dust surface density from the literature. $α_{\rm{CO}}$ and the gas-to-dust ratio ($δ_{\rm{GDR}}$) are simultaneously derived in annuli of 2 kpc width from R = 1-7 kpc. We find that $α_{\rm{CO}}$ and $δ_{\rm{GDR}}$ both increase radially, by a factor of $\sim$ 2-3 from the center to the outskirts of the disk. The luminosity-weighted averages over the disk are $α_{\rm{CO}} = 3.14$ (2.06, 4.96) M$_{\odot}$ pc$^{-2}$[K$\cdot$ km s$^{-1}$]$^{-1}$ and $δ_{\rm{GDR}}$ = 137 (111, 182) at the 68% (1$σ$) confidence level. These are consistent with the $α_{\rm{CO}}$ and $δ_{\rm{GDR}}$ values measured in the Milky Way. In addition to possible variations of $α_{\rm{CO}}$ due to the radial metallicity gradient, we test the possibility of variations in $α_{\rm{CO}}$ due to changes in the underlying cloud populations, as a function of galactic radius. Using a truncated power-law molecular cloud CO luminosity function and an empirical power-law relation for cloud-mass and luminosity, we show that the changes in the underlying cloud population may account for a factor of $\sim 1.5-2.0$ radial change in $α_{\rm{CO}}$.

  PublisherOA PDFDOI

• Erratum: “First Detection of the Molecular Cloud Population in the Extended Ultraviolet Disk of M83” (2022, ApJ, 941, 3)

  The Astrophysical Journal · 2024-07-01

  erratumOpen access

  Erratum: “First Detection of the Molecular Cloud Population in the Extended Ultraviolet Disk of M83” (2022, ApJ, 941, 3), Jin Koda, Linda Watson, Françoise Combes, Monica Rubio, Samuel Boissier, Masafumi Yagi, David Thilker, Amanda M Lee, Yutaka Komiyama, Kana Morokuma-Matsui, Celia Verdugo

  PublisherOA PDFDOI

• Whole-disk Sampling of Molecular Clouds in M83

  The Astrophysical Journal · 2024-11-22 · 2 citations

  articleOpen access

  Abstract We present a catalog of clouds identified from the 12 CO (1–0) data of M83, which was observed using the Atacama Large Millimeter/submillimeter Array with a spatial resolution of ∼46 pc and a mass sensitivity of ∼10 4 M ⊙ (3 σ ). The almost full-disk coverage and high sensitivity of the data allowed us to sample 5724 molecular clouds with a median mass of ∼1.9 × 10 5 M ⊙ , which is comparable to the most frequently sampled mass of giant molecular clouds by surveys in the Milky Way (MW). About 60% of the total CO luminosity in M83's disk arises from clouds more massive than 10 6 M ⊙ . Such massive clouds comprise 16% of the total clouds in number and tend to concentrate toward the arm, bar, and center, while smaller clouds are more prevalent in interarm regions. Most &gt;10 6 M ⊙ clouds have peak brightness temperatures T peak above 2 K with the current resolution. Comparing the observed cloud properties with the scaling relations determined by P. M. Solomon et al. (1987, hereafter S87), T peak &gt; 2 K clouds follow the relations, but T peak &lt; 2 K clouds, which are dominant in number, deviate significantly. Without considering the effect of beam dilution, the deviations would suggest modestly high virial parameters (median α vir ∼ 2.7) and low surface mass densities (median Σ ∼ 22 M ⊙ pc −2 ) for the entire cloud samples, which are similar to values found for the MW clouds by T. S. Rice et al. (2016) and M.-A Miville-Deschênes et al. (2017). However, once beam dilution is taken into account, the observed α vir and Σ for a majority of the clouds (mostly T peak &lt;2 K) can be potentially explained with intrinsic Σ of ∼100 M ⊙ pc −2 and α vir of ∼1, which are similar to the clouds of S87.

  PublisherDOI

• The CO-to-H₂ Conversion Factor in the Barred Spiral Galaxy M83

  The Astrophysical Journal · 2024-06-01 · 9 citations

  articleOpen access1st authorCorresponding

  Abstract We analyze the CO-to-H 2 conversion factor ( α CO ) in the nearby barred spiral galaxy M83. We present new H i observations from the VLA and single-dish GBT in the disk of the galaxy, and combine them with maps of CO(1-0) integrated intensity and dust surface density from the literature. α CO and the gas-to-dust ratio ( δ GDR ) are simultaneously derived in annuli of 2 kpc width from R = 1–7 kpc. We find that α CO and δ GDR both increase radially, by a factor of ∼2–3 from the center to the outskirts of the disk. The luminosity-weighted averages over the disk are α CO = 3.14 (2.06, 4.96) <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msub> <mml:mrow> <mml:mi>M</mml:mi> </mml:mrow> <mml:mrow> <mml:mo>⊙</mml:mo> </mml:mrow> </mml:msub> <mml:mspace width="0.25em"/> <mml:mspace width="0.25em"/> <mml:msup> <mml:mrow> <mml:mi>pc</mml:mi> </mml:mrow> <mml:mrow> <mml:mo>−</mml:mo> <mml:mn>2</mml:mn> </mml:mrow> </mml:msup> <mml:msup> <mml:mrow> <mml:mo stretchy="false">[</mml:mo> <mml:mi mathvariant="normal">K</mml:mi> <mml:mspace width="0.25em"/> <mml:mspace width="0.25em"/> <mml:mi>km</mml:mi> <mml:mspace width="0.25em"/> <mml:msup> <mml:mrow> <mml:mi mathvariant="normal">s</mml:mi> </mml:mrow> <mml:mrow> <mml:mo>−</mml:mo> <mml:mn>1</mml:mn> </mml:mrow> </mml:msup> <mml:mo stretchy="false">]</mml:mo> </mml:mrow> <mml:mrow> <mml:mo>−</mml:mo> <mml:mn>1</mml:mn> </mml:mrow> </mml:msup> </mml:math> and δ GDR = 137 (111, 182) at the 68% (1 σ ) confidence level. These are consistent with the α CO and δ GDR values measured in the Milky Way. In addition to possible variations of α CO due to the radial metallicity gradient, we test the possibility of variations in α CO due to changes in the underlying cloud populations, as a function of galactic radius. Using a truncated power-law molecular cloud CO luminosity function and an empirical power-law relation for cloud mass and luminosity, we show that the changes in the underlying cloud population may account for a factor of ∼1.5–2.0 radial change in α CO .

  PublisherOA PDFDOI

Frequent coauthors

• S. Boissier

  15 shared

• Jin Koda

  13 shared

• Akihiko Hirota

  Atacama Large Millimeter Submillimeter Array

  12 shared

• F. Combes

  Collège de France

  11 shared

• M. H. Heyer

  8 shared

• Fumi Egusa

  The University of Tokyo

  8 shared

• Nanase Harada

  The Graduate University for Advanced Studies, SOKENDAI

  7 shared

• S. Martín

  7 shared

Awards & honors

• Inaugural Outstanding Teaching Award in the School of Comput…
• Endowed position (W. M. Keck Associate Professor of Computer…