# Systematic biases in determining dust attenuation curves through galaxy SED fitting

**Authors:** Jianbo Qin, Xian Zhong Zheng, Min Fang, Zhizheng Pan, Stijn Wuyts, Yong Shi, Yingjie Peng, Valentino Gonzalez, Fuyan Bian, Jia-Sheng Huang, Qiu-Sheng Gu, Wenhao Liu, Qinghua Tan, Dong Dong Shi, Jian Ren, Yuheng Zhang, Man Qiao, Run Wen, Shuang Liu  
**arXiv:** [2201.05467](https://arxiv.org/abs/2201.05467)  
**Date:** 2022-01-14  
**Summary page:** https://www.summarizepaper.com/en/paper/2201.05467v1/

## Key points

- Study investigates systematic biases in determining dust attenuation curves through galaxy spectral energy distribution (SED) fitting
- Focus on understanding relationship between slope of dust attenuation curve ($\delta$) and effective dust attenuation ($A_V$)
- Analyze sample of local star-forming galaxies (SFGs) from Galaxy And Mass Assembly survey and mock galaxy SEDs
- Best-fitting $\beta_0$ for SFGs shows wider coverage compared to starburst with constant star formation
- Strong degeneracies between $\beta_0$, $\delta$, and $A_V$ in SED fitting process
- Degeneracies induce systematic biases leading to false correlation between $A_V$ and $\delta$
- Simulation tests using input model galaxy SEDs with flat $A_V$-$\delta$ relation reproduce observed correlation
- Variations in best-fitting $\delta$ primarily due to fitting errors rather than true physical differences
- Assuming starburst with constant star formation results in steeper attenuation curve, smaller degeneracy errors, and stronger $A_V$-$\delta$ relation
- Apparent $A_V$-$\delta$ relation likely driven by systematic biases induced by fitting degeneracies
- Importance of considering degeneracy effects in SED fitting when determining dust attenuation curves

## Summary

This study investigates the systematic biases in determining dust attenuation curves through galaxy spectral energy distribution (SED) fitting. The researchers focus on understanding how the fitting degeneracies shape the relationship between the slope of the dust attenuation curve ($\delta$) and effective dust attenuation ($A_V$). They analyze a sample of local star-forming galaxies (SFGs) selected from the Galaxy And Mass Assembly survey, along with mock galaxy SEDs of known attenuation parameters. To generate model SED templates, the researchers adopt a well-designed declining starburst star formation history and vary the intrinsic UV slope ($\beta_0$) over a wide range. Surprisingly, they find that the best-fitting $\beta_0$ for their sample SFGs shows a much wider coverage compared to a starburst with constant star formation, which has a limited range of $\beta_0<-2.2$. This discrepancy suggests strong degeneracies between $\beta_0$, $\delta$, and $A_V$ in the SED fitting process. The results reveal that these degeneracies induce systematic biases leading to a false correlation between $A_V$ and $\delta$. To further investigate this relationship, simulation tests are conducted using input model galaxy SEDs with a flat $A_V$-$\delta$ relation. Remarkably, even with this assumption, the researchers find that they can reproduce the observed $A_V$-$\delta$ correlation. Therefore, they conclude that variations in best-fitting $\delta$ are primarily dominated by fitting errors rather than true physical differences. Additionally, assuming a starburst with constant star formation in SED fitting results in a steeper attenuation curve, smaller degeneracy errors, and a stronger $A_V$-$\delta$ relation. These findings confirm that the apparent $A_V$-$\delta$ relation obtained through SED fitting is likely driven by systematic biases induced by the fitting degeneracies between $\beta_0$, $\delta$, and $A_V$. Overall, this study highlights the importance of considering the degeneracy effects in SED fitting when determining dust attenuation curves. The results provide valuable insights into the limitations and potential biases associated with inferring dust properties from galaxy SEDs.