DESI survey validation data in the COSMOS/HSC field: Cool gas trace main sequence star-forming galaxies at the cosmic noon

TL;DR

This study uses DESI validation data to analyze the gaseous halos of star-forming galaxies at cosmic noon, revealing correlations between gas absorption features and galaxy properties, and evidence of feedback processes.

Contribution

First to utilize DESI validation data for CGM studies at high redshift, providing new insights into gas-galaxy interactions and evolution during cosmic noon.

Findings

Strong MgII absorption correlates with star formation rate within 250 kpc.

Covering fraction of MgII evolves significantly in main-sequence galaxies from z=0.9 to 2.2.

Detection of gas out of virial radius suggests feedback effects.

Abstract

We present the first result in exploring the gaseous halo and galaxy correlation using the Dark Energy Spectroscopic Instrument (DESI) survey validation data in the Cosmic Evolution Survey (COSMOS) and Hyper Suprime-Cam (HSC) field. We obtain the multiphase gaseous halo properties in the circumgalactic medium (CGM) by using 115 quasar spectra (S/N > 3). We detect MgII absorption at redshift 0.6 < z < 2.5, CIV absorption at 1.6 < z < 3.6, and HI absorption associated with the MgII and CIV. By cross-matching the COSMOS2020 catalog, we identify the MgII and CIV host galaxies in ten quasar fields at 0.9 < z < 3.1. We find that within the impact parameter of 250 kpc, a tight correlation is seen between strong MgII equivalent width and the host galaxy star formation rate. The covering fraction fc of strong MgII selected galaxies, which is the ratio of absorbing galaxy in a certain galaxy population, shows significant evolution in the main-sequence galaxies and marginal evolution in all the galaxy populations within 250 kpc at 0.9 < z < 2.2. The fc increase in the main-sequence galaxies likely suggests the co-evolution of strong MgII absorbing gas and the main-sequence galaxies at the cosmic noon. Furthermore, several MgII and CIV absorbing gas is detected out of the galaxy virial radius, tentatively indicating the feedback produced by the star formation and/or the environmental effects.

Tables

Table 1. Table 1: Mg  ii -absorbing galaxy candidates information

QSO	RA	DEC	$D$	$D/R_{vir}$	$z_{abs}$	$W_r$	log(SFR)	log M*/M⊙	$L_v/L_{\odot }$
			(kpc)			(Å)	(M∗/yr)
J100219.49+015536.84	150.5812	1.9269	218.53	1.79	0.9520	1.462$\pm$0.316	0.54${}_{-0.07}{}^{}{}_{}{}^{0.07}$	9.6${}_{-0.03}{}^{}{}_{}{}^{0.04}$	10.04
J100219.49+015536.84	150.5812	1.9269	102.62	0.31	1.2090	0.932$\pm$0.260	0.0${}_{-0.068}{}^{}{}_{}{}^{0.068}$	10.11${}_{-0.03}{}^{}{}_{}{}^{0.04}$	10.14
J095834.03+024426.88	149.6418	2.7408	181.26	2.21	1.2756	1.005$\pm$0.126	–0.17${}_{-0.297}{}^{}{}_{}{}^{0.072}$	8.95${}_{-0.13}{}^{}{}_{}{}^{0.09}$	8.7
J095749.98+013354.10	149.4582	1.5650	144.51	0.73	1.5660	1.169$\pm$0.202	0.19${}_{-0.13}{}^{}{}_{}{}^{0.18}$	9.16${}_{-0.11}{}^{}{}_{}{}^{0.10}$	9.18
J095749.98+013354.10	149.4582	1.5650	153.39	0.73	1.6428	1.223$\pm$0.162	0.21${}_{-0.14}{}^{}{}_{}{}^{0.19}$	9.21${}_{-0.10}{}^{}{}_{}{}^{0.09}$	9.24
J100031.61+014757.48	150.1317	1.7993	150.82	1.89	1.6625	1.836$\pm$0.257	0.73${}_{-0.09}{}^{}{}_{}{}^{0.08}$	9.41${}_{0.06}{}^{}{}_{}{}^{0.05}$	9.88
J095949.39+020140.80	149.9558	2.0280	137.37	0.74	1.7372	1.152$\pm$0.175	0.91${}_{-0.07}{}^{}{}_{}{}^{0.09}$	9.68${}_{-0.07}{}^{}{}_{}{}^{0.05}$	9.9
J095749.98+013354.10	149.4582	1.5650	146.41	0.43	1.8770	1.234$\pm$0.114	0.29${}_{0.829}{}^{}{}_{}{}^{1.262}$	9.50${}_{-0.09}{}^{}{}_{}{}^{0.07}$	9.51
J100014.14+020054.36	150.0589	2.0151	149.83	1.85	1.9813	0.300$\pm$0.117	1.17${}_{-0.22}{}^{}{}_{}{}^{0.32}$	9.56${}_{-0.04}{}^{}{}_{}{}^{0.04}$	10.14
J100105.30+021348.00	150.2703	2.2312	66.70	0.48	2.1680	0.800$\pm$0.235	0.36${}_{0.12}{}^{}{}_{}{}^{0.1}$	9.01${}_{0.09}{}^{}{}_{}{}^{0.08}$	9.46

Table 2. Table 2: C  iv -absorbing galaxy candidates information

QSO	RA	DEC	$D$	$D/R_{vir}$	$z_{abs}$	$W_r$	log(SFR)	log M*/M⊙	$Lv/L_{\odot }$
			(kpc)			(Å)	(M∗/yr)
J095749.98+013354.1	149.4582	1.5650	150.82	0.73	1.6428	0.738$\pm$0.200	0.21${}_{-0.14}{}^{}{}_{}{}^{0.19}$	9.24${}_{9.15}{}^{}{}_{}{}^{9.37}$	9.24
J100031.61+014757.48	150.1317	1.7993	129.28	1.97	1.6625	1.828$\pm$0.270	0.73${}_{-0.09}{}^{}{}_{}{}^{0.08}$	9.88${}_{9.55}{}^{}{}_{}{}^{9.66}$	9.88
J095949.39+020140.80	149.9558	2.0280	137.37	1.87	1.7372	1.126$\pm$0.080	0.91${}_{-0.07}{}^{}{}_{}{}^{0.09}$	9.9${}_{10.03}{}^{}{}_{}{}^{10.16}$	9.9
J100302.90+015208.40	150.7621	1.8690	178.71	0.75	1.7970	0.531$\pm$0.104	0.22${}_{-0.31}{}^{}{}_{}{}^{0.11}$	9.36${}_{8.6}{}^{}{}_{}{}^{8.83}$	9.36
J100014.14+020054.36	150.0589	2.0151	137.45	1.83	1.8400	0.574$\pm$0.147	0.8${}_{-0.5}{}^{}{}_{}{}^{0.25}$	9.95${}_{7.94}{}^{}{}_{}{}^{8.23}$	9.95
J095834.03+024426.88	149.6418	2.7408	142.13	1.30	1.8555	0.252$\pm$0.100	0.55${}_{-0.5}{}^{}{}_{}{}^{0.25}$	9.22${}_{8.81}{}^{}{}_{}{}^{9.12}$	9.22
J095749.98+013354.10	149.4582	1.5650	146.41	2.33	1.8770	0.716$\pm$0.142	0.29${}_{-0.22}{}^{}{}_{}{}^{0.32}$	9.51${}_{10.38}{}^{}{}_{}{}^{10.54}$	9.51
J100014.14+020054.36	150.0589	2.0151	152.60	0.46	1.9450	0.620$\pm$0.099	1.14${}_{0.07}{}^{}{}_{}{}^{0.07}$	10.1${}_{8.97}{}^{}{}_{}{}^{9.12}$	10.1
J095752.30+022021.12	149.4679	2.3392	177.02	0.46	1.9810	0.546$\pm$0.200	0.66${}_{-0.28}{}^{}{}_{}{}^{0.11}$	9.64${}_{8.62}{}^{}{}_{}{}^{8.87}$	9.64
J100014.14+020054.36	150.0589	2.0151	149.84	1.83	1.9813	0.453$\pm$0.114	1.17${}_{-0.07}{}^{}{}_{}{}^{0.07}$	10.14${}_{8.97}{}^{}{}_{}{}^{9.12}$	10.14
J100014.14+020054.36	150.0589	2.0151	135.46	2.19	2.1305	0.434$\pm$0.071	0.80${}_{0.1}{}^{}{}_{}{}^{0.09}$	9.95${}_{10.98}{}^{}{}_{}{}^{11.08}$	9.95
J100105.30+021348.00	150.2721	2.2300	133.99	0.49	2.1530	0.518$\pm$0.341	0.35${}_{-0.13}{}^{}{}_{}{}^{0.1}$	9.45${}_{9.314}{}^{}{}_{}{}^{9.432}$	9.45
J100105.30+021348.00	150.2721	2.2300	137.10	0.71	2.1680	1.102$\pm$0.221	0.36${}_{-0.12}{}^{}{}_{}{}^{0.1}$	9.46${}_{9.314}{}^{}{}_{}{}^{9.432}$	9.46
J095806.96+022248.36	149.5209	2.3814	117.77	0.42	3.0880	1.207$\pm$0.211	1.37${}_{-0.09}{}^{}{}_{}{}^{0.09}$	9.74${}_{8.67}{}^{}{}_{}{}^{8.74}$	9.74

Table 3. Table 3: The covering fraction of Mg  ii  and C  iv in all the galaxies and main-sequence galaxies at different impact parameter bins (50-100, 100-200, 200-250 kpc) respectively. The galaxies are searched within 30 arcseconds offset by the quasar(absorbing system) using the COSMOS2020 catalog.

$D$	$f_c$ Mg ii	$f_c$ Mg ii (MSG)	$f_c$ C iv	$f_c$ C iv (MSG)
50 – 100 kpc	0.152 $\pm$ 0.076	0.290 $\pm$ 0.127	0.176 $\pm$ 0.024	0.333 $\pm$ 0.083
100 – 200 kpc	0.074 $\pm$ 0.119	0.154 $\pm$ 0.293	0.071 $\pm$ 0.022	0.167 $\pm$ 0.051
200 – 250 kpc	0.061 $\pm$ 0.187	0.103 $\pm$ 0.141	0.055 $\pm$ 0.017	0.100 $\pm$ 0.038

Table 4. Table 4: Incidence rate d ​ N / d ​ z 𝑑 𝑁 𝑑 𝑧 dN/dz and comoving density of Mg  ii  and C  iv  absorbers in this work.

$\Delta z$	Mg ii  $dN/dz$	Mg ii  $dN/dz$	$\Delta z$	C iv  $dN/dz$	C iv $dN/dz$
	($W_r>$ 0.3 Å)	($W_r>$ 1.0 Å)		($W_r>$ 0.2 Å)	($W_r>$ 1.0 Å)
0.60 – 1.00	0.606 $\pm$ 0.076	0.275 $\pm$ 0.127	1.30 – 1.50	1.675 $\pm$ 0.149	0.494 $\pm$ 0.094
1.00 – 1.50	1.079 $\pm$ 0.119	0.484 $\pm$ 0.293	1.50 – 2.00	1.549 $\pm$ 0.206	0.420 $\pm$ 0.122
1.50 – 2.00	1.330 $\pm$ 0.187	0.602 $\pm$ 0.141	2.00 – 2.50	2.311 $\pm$ 0.447	0.800 $\pm$ 0.298
2.00 – 2.50	1.440 $\pm$ 0.298	0.403 $\pm$ 0.149	2.50 – 3.20	3.689 $\pm$ 0.1.07	1.618 $\pm$ 0.799

Table 5. Table 5: Quasars used in this work. (1) Quasar name (2) Quasar emission redshift (3) RA (4) DEC (5) The quasar effective exposure time (6) The mean signal-to-noise measured from three continuum region without obvious emission or absorption.

TARGETID(1)	$z_{em}$(2)	RA(3)	DEC(4)	EXPTIME(5)	S/N(6)
J095426.83+025022.92	1.3422	148.6118	2.0897	22884.44	8
J095430.38+021525.92	2.0498	148.6266	2.2572	7302.19	17
J095435.15+023142.24	1.3138	148.6465	2.5284	22884.44	7
J095436.38+027031.08	1.8071	148.6516	2.1253	22884.44	7
J095446.03+014639.00	1.1153	148.6918	1.7775	7302.19	3
J095458.24+015616.79	0.7464	148.7427	1.9380	22884.44	29
J095504.24+026052.92	3.0927	148.7677	2.1147	22884.44	6
J095505.44+021028.92	1.2235	148.7727	2.1747	22884.44	4
J095525.79+028011.76	3.1414	148.8575	2.1366	2884.447	3
J095615.91+024652.67	1.6542	149.0663	2.7813	22884.44	14
J095656.18+021314.88	1.1215	149.2341	2.2208	22884.44	37
J095712.88+014917.39	1.1818	149.3037	1.8215	22884.44	12
J095726.32+024027.83	0.9583	149.3597	2.0744	22884.44	22
J095739.24+015533.23	1.8146	149.4135	1.9259	22884.44	3
J095749.96+013353.99	2.0055	149.4582	1.5650	7302.19	18
J095752.29+022021.11	2.0490	149.4679	2.3392	7302.19	41
J095806.96+022248.36	3.0956	149.5290	2.3801	7302.19	6
J095820.44+023003.95	1.3578	149.5852	2.0511	22884.44	36
J095834.03+024426.88	1.8927	149.6418	2.7408	7302.19	33
J095834.51+034338.63	1.2651	149.6438	3.7274	7302.19	7
J095847.11+035003.84	1.8582	149.6963	3.8344	22884.44	15
J095900.11+033651.83	1.5873	149.7505	3.6144	7302.19	4
J095910.94+019046.80	2.6908	149.7956	1.1630	22884.44	5
J095911.63+033442.95	1.8081	149.7985	3.5786	22884.44	3
J095915.28+034033.23	0.6818	149.8137	3.6759	7302.19	19
J095922.27+034046.56	4.0667	149.8428	3.6796	7302.19	3
J095923.80+003853.16	0.7838	149.8492	0.6481	7302.19	3
J095931.72+033710.19	1.1300	149.8822	3.6195	6402.13	4
J095933.12+033118.12	2.1464	149.8880	3.5217	7302.19	6
J095946.82+004918.47	2.2488	149.9451	0.8218	22884.44	70
J095949.39+021040.80	1.7533	149.9558	2.0280	7302.19	26
J095956.56+004301.55	1.9471	149.9857	0.7171	7302.19	4
J100009.35+005311.03	0.9103	150.0390	0.8864	22884.44	6
J100011.59+004154.24	1.3761	150.0483	0.6984	7302.19	3
J100014.13+020054.35	2.4968	150.0589	2.0151	22884.44	32
J100017.87+005400.00	0.7289	150.0745	0.9000	7302.19	5
J100020.49+015011.40	1.5242	150.0854	1.0865	22884.44	8
J100020.49+033247.76	2.0233	150.0854	3.5466	22884.44	3
J100022.72+033724.96	0.9057	150.0947	3.6236	7302.19	6
J100023.78+035500.12	1.1259	150.0991	3.9167	22884.44	16
J100025.32+034823.76	2.4103	150.1055	3.8066	22884.44	15
J100029.13+011044.75	1.0591	150.1214	1.0291	7302.19	37
J100029.68+035023.27	1.6433	150.1237	3.0898	22884.44	21
J100031.60+014757.47	1.683	150.1317	1.7993	22884.44	17
J100032.80+033458.43	1.724	150.1367	3.5829	7302.19	6
J100036.98+018003.12	1.8355	150.1541	1.1342	7302.19	3
J100037.39+034455.68	3.2336	150.1558	3.7488	7302.19	15
J100038.47+015009.24	2.1797	150.1603	1.0859	22884.44	9
J100039.23+012006.35	3.8343	150.1635	1.0351	7302.19	5
J100039.55+033216.44	0.9768	150.1648	3.5379	22884.44	11
J100042.11+034911.27	1.6542	150.1755	3.8198	22884.44	12
J100042.98+004438.76	1.452	150.1791	0.7441	7302.19	5
J100043.39+033217.16	2.2667	150.1808	3.5381	22884.44	3
J100047.92+034700.95	1.9006	150.1997	3.7836	7302.19	3
J100048.83+005925.79	0.8704	150.2035	0.9905	7302.19	3
J100048.83+033039.23	3.3652	150.2035	3.5109	22884.44	7
J100052.29+005121.59	1.9461	150.2179	0.8560	7302.19	7
J100053.80+033105.87	2.4587	150.2242	3.5183	22884.44	3
J100055.00+005508.40	2.0652	150.2292	0.9190	7302.19	5
J100058.53+004837.44	1.7392	150.2439	0.8104	7302.19	3
J100058.82+015359.99	1.5619	150.2451	1.9000	7302.19	16
J100101.03+035233.96	2.7714	150.2543	3.8761	22884.44	1
J100104.60+004648.00	2.5726	150.2692	0.7800	7302.19	3
J100105.30+021347.99	2.6066	150.2721	2.2300	7302.19	8
J100106.43+033650.39	2.0931	150.2768	3.6140	7302.19	4
J100108.66+005730.59	2.0411	150.2861	0.9585	7302.19	25
J100109.21+004859.40	0.6574	150.2884	0.8165	7302.19	8
J100111.40+034100.23	2.2668	150.2975	3.6834	7302.19	29
J100111.49+033506.36	2.4574	150.2979	3.5851	7302.19	21
J100113.39+005009.95	2.5853	150.3058	0.8361	7302.19	4
J100118.31+035101.07	2.8947	150.3263	3.8503	22884.44	6
J100125.46+005205.15	0.7802	150.3561	0.8681	22884.44	56
J100126.59+004648.71	1.7273	150.3608	0.7802	7302.19	8
J100127.55+034434.07	2.8133	150.3648	3.7428	22884.44	22
J100129.44+003813.55	2.9074	150.3727	0.6371	4602.01	3
J100132.08+005259.15	1.0669	150.3837	0.8831	22884.44	8
J100133.36+005118.71	1.403	150.3890	0.8552	22884.44	11
J100134.15+011021.72	1.7589	150.3923	1.1727	7302.19	3
J100136.36+034309.48	1.1248	150.4015	3.7193	5502.06	5
J100137.19+021612.35	1.65	150.4050	2.2701	2884.447	11
J100137.77+004655.56	2.5841	150.4074	0.7821	7302.19	25
J100138.97+033616.19	1.2473	150.4124	3.6045	7302.19	10
J100140.31+003947.52	1.6027	150.4180	0.6632	7302.19	4
J100142.04+003907.55	1.3516	150.4252	0.6521	7302.19	3
J100142.55+015031.20	1.8225	150.4273	1.0920	7302.19	6
J100147.88+021447.03	0.8804	150.4495	2.2464	22884.44	17
J100205.23+004249.68	1.7855	150.5218	0.7138	7302.19	11
J100217.87+004252.20	1.2343	150.5745	0.7145	7302.19	10
J100219.48+015536.84	1.5087	150.5812	1.9269	7302.19	17
J100236.69+015948.47	1.5192	150.6529	1.9968	7302.19	19
J100302.90+015208.40	1.8026	150.7621	1.8690	7302.19	22
J100344.35+025002.03	2.9914	150.9348	2.0839	7302.19	8
J100348.67+021044.76	1.3908	150.9528	2.1791	7302.19	4
J100417.61+021330.35	3.1056	151.0734	2.2251	22884.44	3
J100441.78+021147.04	1.828	151.1741	2.1964	22884.44	6
J100449.99+021641.52	1.7882	151.2083	2.2782	22884.44	3
J100505.03+021519.08	2.8624	151.2710	2.2553	22884.44	9
J100520.87+021112.84	2.3843	151.3370	2.1869	7302.19	7
J100523.85+015920.40	1.7769	151.3494	1.9890	22884.44	18
J100524.86+025047.76	1.0844	151.3536	2.0966	22884.44	6
J100527.09+027025.31	1.5706	151.3629	2.1237	22884.44	5
J100534.43+021015.96	1.8209	151.3935	2.1711	7302.19	18
J100541.51+015950.64	1.7298	151.4230	1.9974	22884.44	16
J100542.69+021516.92	1.1336	151.4279	2.2547	22884.44	4
J100546.20+027052.67	1.7086	151.4425	2.1313	22884.44	4
J100547.68+021221.59	0.9066	151.4487	2.2060	22884.44	11
J100606.45+021445.23	1.1864	151.5269	2.2459	22884.44	6
J100624.45+014758.20	1.0138	151.6019	1.7995	22884.44	4
J100632.71+013955.43	2.4867	151.6363	1.6654	6402.13	3
J100634.60+026014.40	2.3479	151.6442	2.1040	7302.19	3
J100638.88+014941.16	2.1134	151.6620	1.8281	7302.19	3
J100641.11+021658.07	2.0098	151.6713	2.2828	22884.44	17
J100039.25+010206.47	3.8344	150.1635	1.0350	7302.19	18
J100015.90+010801.75	2.0098	151.6713	2.2828	22884.44	17

Table 6. Table 6: Metal absorbers Measurements (1) Quasar name (2) Absorber redshift (3) Velocity width of Mg  ii  ( λ 𝜆 \lambda 2796) line (4) Equivalent width of the Mg  ii  ( λ 𝜆 \lambda 2796) line (5) Velocity width of C  iv  ( λ 𝜆 \lambda 1548) line (6) Equivalent width of the C  iv  ( λ 𝜆 \lambda 1548) line.

TARGETID(1)	$z_{abs}$(2)	$\Delta v\left(\lambda 2796\right)$(3)	$W_r\left(\lambda 2796\right)$(4)	$\Delta \left(\lambda 1548\right)$(5)	$W_r\left(\lambda 1548\right)$(6)
		(km s-1)	(Å)	(km s-1   )	(Å)
J100219.48+015536.84	0.9520	495.462	1.462$\pm$0.316	nan	nan
J100219.48+015536.84	1.2090	206.951	0.631$\pm$0.096	nan	nan
J100302.90+015208.40	1.7970	411.110	0.533$\pm$0.091	nan	nan
J095749.96+013353.99	1.6428	379.000	1.223$\pm$0.162	419.351	0.738$\pm$0.2
J095749.96+013353.99	1.5660	465.260	1.169$\pm$0.202	375.754	0.716$\pm$0.142
J095749.96+013353.99	1.8770	314.938	1.234$\pm$0.114	nan	nan
J095834.03+024426.88	1.2756	393.437	1.005$\pm$0.126	nan	nan
J095820.44+023003.95	0.7505	480.986	0.9 $\pm$0.087	nan	nan
J095808.15+015425.20	2.4845	119.614	1.676$\pm$0.666	nan	nan
J100113.39+005009.95	1.0620	294.506	1.853$\pm$0.636	nan	nan
J100137.77+004655.56	2.1385	528.224	1.656$\pm$0.278	nan	nan
J100137.77+004655.56	1.3445	235.737	0.41 $\pm$0.108	nan	nan
J100205.23+004249.68	1.2510	410.364	1.286$\pm$0.392	nan	nan
J100039.25+010206.47	1.3450	219.637	0.74 $\pm$0.385	nan	nan
J100108.66+005730.59	2.0250	nan	nan	402.453	0.386$\pm$0.065
J100108.66+005730.59	1.7378	486.468	0.570$\pm$0.196	449.654	0.503$\pm$0.193
J095430.38+021525.92	1.8315	636.834	0.636$\pm$0.43	364.219	0.84 $\pm$0.189
J095430.38+021525.92	1.1072	336.301	0.346$\pm$0.22	nan	nan
J100111.40+034100.23	2.2660	nan	nan	606.366	1.114$\pm$0.055
J100111.40+034100.23	1.4660	nan	nan	976.981	0.625$\pm$0.408
J100111.40+034100.23	1.5475	431.816	0.476$\pm$0.157	nan	nan
J100111.40+034100.23	1.6715	352.731	0.336$\pm$0.139	nan	nan
J100534.43+021015.96	1.4990	693.396	4.891$\pm$0.283	nan	nan
J100534.43+021015.96	1.1940	391.394	0.719$\pm$0.195	nan	nan
J100111.49+033506.36	1.4140	nan	nan	643.073	0.504$\pm$0.217
J100111.49+033506.36	0.6605	785.408	0.728$\pm$0.36	nan	nan
J100111.49+033506.36	1.4140	444.960	0.712$\pm$0.16	nan	nan
J100541.51+015950.64	0.8280	420.354	0.908$\pm$0.278	nan	nan
J100541.51+015950.64	1.4480	316.473	1.408$\pm$0.155	451.940	1.444$\pm$0.268
J100527.09+027025.31	1.4846	360.323	1.699$\pm$0.574	464.818	1.591$\pm$0.559
J100042.11+034911.27	1.6425	544.964	2.611$\pm$0.141	610.488	1.907$\pm$0.261
J095946.82+004918.47	2.1544	359.402	0.399$\pm$0.034	482.318	0.227$\pm$0.043
J095847.11+035003.84	1.3310	453.723	0.856$\pm$0.203	nan	nan
J100025.32+034823.76	1.5140	349.751	0.741$\pm$0.205	374.901	0.627$\pm$0.079
J100029.68+035023.27	0.9760	423.160	1.01 $\pm$0.196	nan	nan
J100217.87+004252.20	1.2230	213.961	0.274$\pm$0.149	nan	nan
J100015.90+010801.75	1.4780	312.211	2.061$\pm$0.8	nan	nan
J100632.71+013955.43	2.4886	146.668	0.848$\pm$0.249	nan	nan
J095949.39+021040.80	1.7372	530.321	1.152$\pm$0.175	522.252	1.126$\pm$0.08
J100058.82+015359.99	0.6715	603.396	2.235$\pm$0.36	nan	nan
J100105.30+021347.99	2.1680	nan	nan	585.025	1.102$\pm$0.421
J100105.30+021347.99	2.1530	212.247	0.799$\pm$0.212	478.487	0.518$\pm$0.341
J100014.13+020054.35	1.4700	241.531	0.804$\pm$0.332	nan	nan
J100133.36+005118.71	0.7838	580.924	1.552$\pm$0.531	nan	nan
J100031.60+014757.47	1.6625	689.316	1.836$\pm$0.257	722.683	1.828$\pm$0.27
J095426.83+025022.92	1.3500	403.319	1.279$\pm$0.253	504.614	0.877$\pm$0.281
J095726.32+024027.83	0.7575	381.720	0.622$\pm$0.158	nan	nan
J100014.13+020054.35	2.1305	nan	nan	395.208	0.434$\pm$0.071
J100014.13+020054.35	1.9813	342.842	0.300$\pm$0.117	514.078	0.453$\pm$0.148
J100014.13+020054.35	1.9450	nan	nan	558.430	0.62 $\pm$0.099
J100014.13+020054.35	1.8400	nan	nan	568.266	0.574$\pm$0.147
J100137.19+021612.35	1.6360	497.133	2.682$\pm$0.205	669.046	2.911$\pm$0.99
J100441.78+021147.04	1.4015	416.102	1.794$\pm$0.521	nan	nan
J095435.15+023142.24	1.3010	443.636	1.971$\pm$0.336	nan	nan
J100127.55+034434.07	2.7740	nan	nan	422.899	1.34$\pm$0.09
J100127.55+034434.07	2.2294	284.891	0.327$\pm$0.122	468.897	1.08$\pm$0.125
J100055.00+005508.40	1.9490	447.044	1.538$\pm$0.517	nan	nan
J095752.29+022021.11	1.9810	nan	nan	523.836	0.546$\pm$0.210
J095834.03+024426.88	1.8555	nan	nan	508.977	0.252$\pm$0.103
J095826.64+024228.00	2.4520	nan	nan	482.868	0.923$\pm$0.296
J095839.84+024424.00	3.1620	nan	nan	220.482	0.530$\pm$0.100
J095839.84+024424.00	3.1800	nan	nan	152.901	0.217$\pm$0.100
J100038.47+015009.24	2.1774	nan	nan	510.516	1.124$\pm$0.121
J100104.60+004648.00	2.5350	nan	nan	602.860	3.052$\pm$0.991
J100129.44+003813.55	2.9100	nan	nan	515.093	1.92 $\pm$0.482
J100055.00+005508.40	2.0710	nan	nan	464.685	1.506$\pm$0.271
J100638.88+014941.16	2.1080	nan	nan	323.381	0.922$\pm$0.214
J100344.35+025002.03	3.0290	nan	nan	481.671	1.477$\pm$0.258
J095806.96+022248.36	3.0880	nan	nan	369.762	1.207$\pm$0.211
J095933.12+033118.12	2.1020	nan	nan	504.720	1.098$\pm$0.35
J095933.12+033118.12	2.1320	248.939	0.859$\pm$0.217	581.047	1.143$\pm$0.202
J100037.39+034455.68	2.8240	nan	nan	400.547	0.56 $\pm$0.123
J095504.24+026052.92	2.2050	nan	nan	347.999	0.694$\pm$0.103
J095436.38+027031.08	1.7950	nan	nan	721.820	2.092$\pm$0.246
J095525.79+028011.76	3.1390	nan	nan	261.151	0.684$\pm$0.133
J100449.99+021641.52	1.7362	nan	nan	457.749	1.303$\pm$0.137
J100020.49+033247.76	2.0110	nan	nan	360.158	1.18 $\pm$0.639
J100043.39+033217.16	2.2674	nan	nan	480.610	1.871$\pm$0.278

Equations

$$f_c(W_r>W_{lim})=\frac{N_{abs}}{N_{tot}}$$

$$F={\rho }_{SFR}^{MgII}/{\rho }_{SFR}=<SFR>\times \frac{dN}{dz}{\frac{dl}{dz}}^{-1}{\sigma }^{-1}/{\rho }_{SFR}$$

Full text

Quasar absorption line spectroscopy (1317); Circumgalactic medium (1879); High-redshift galaxies (734)

††journal: ApJ††facilities: Mayall (DESI)

eh

1 Introduction

The baryon cycle in and around a galaxy is of critical importance in understanding the cosmic star formation history and galaxy evolution. The so-called interstellar medium (ISM) and circumgalactic medium (CGM) play a key role in regulating the baryon cycle and thus galaxy evolution (see reviews of Tumlinson et al. 2017; Péroux & Howk 2020 and the references therein). The absorption lines produced by the intervening medium toward bright background quasars provide a sensitive measurement of the multi-phase gas properties in the transverse direction of the gaseous halos. Different phases (density, temperature, ionization parameter) of the gas in the ISM and CGM can be characterized by different ions. For instance, neutral atomic carbon (C i) is used to trace cold, metal-enriched (Ledoux et al., 2015; Zou et al., 2018) and molecular gas (Noterdaeme et al., 2018). The Ly$\alpha$ absorption systems can probe optically thick neutral gas, known as the damped Lyman$\alpha$ systems (neutral hydrogen column density N (H i) $>10^{20.3}$ cm*-2*, DLA) (Petitjean et al. 2000; Wolfe et al. 2005; Prochter et al. 2006; Noterdaeme et al. 2011; Rafelski et al. 2012; Krogager et al. 2017; Neeleman et al. 2019; Lin et al. 2022). Mg ii ($\lambda\lambda$2796,2803) and C iv doublets are found to reside in the cool (T$\sim$ 104 K) (Bergeron & Boissé 1991; Steidel et al. 2002) and warm-hot gas (T$\sim$ 105.5-6 K) gas, respectively (Bordoloi et al., 2014a; Burchett et al., 2016). Additionally, peculiar abundance patterns of DLA or Lyman-Limit-Systems potentially exhibit the signature of the old generation of stars (e.g., Zou et al. 2020; Welsh et al. 2022).

Extensive experiments have been designed to explore the CGM-galaxy correlation and its role in galaxy evolution at redshift $z<$ 1. The CGM has been found to co-rotate with the host galaxy along the major and minor axis (Nielsen et al., 2013b). The strength (equivalent width or column density) of the cool gas tentatively correlates with the luminosity ($L_{B}$ of the host galaxy (Chen et al., 2010a) and is anti-correlated with the impact parameter of the galaxy (Bouché et al., 2006; Nielsen et al., 2013a). Emission from the diffuse gas region has also been detected in recent work (e.g., Feltre et al. 2018; Leclercq et al. 2022).

Observations are still limited in exploring the CGM-galaxy correlation at $z>$ 1 because direct galaxy observation is very challenging for ground-based telescopes. Stacking of a large quantity of Sloan Digital Sky Survey data (SDSS) quasar spectra can be used to probe the weak galaxy emission at high redshift (Joshi et al., 2017) and CGM distribution at a large scale (Pieri et al., 2014). Recently, integral field spectroscopy such as Keck Cosmic Web Imager (KCWI) and VLT/Multi Unit Spectroscopic Explorer (MUSE) provides an efficient tool to provide a 3D view in both the large and small-scale of the CGM at $z<1.5$ and $z>2$. Samples of Mg ii-galaxy pairs at $z<1.5$ are built using VLT/MUSE (Zabl et al., 2019; Schroetter et al., 2019, 2021; Dutta et al., 2020). A bimodality of azimuthal angle-metallicity relation has been found in MusE GAs FLOw and Wind (MEGAFLOW) survey and at $z\sim$1 to trace either inflow (Zabl et al., 2019) or galactic outflow (Schroetter et al., 2019). The environmental effects of cool gas traced by Mg ii have been reported in the MUSE Analysis of Gas around Galaxies (MAGG) survey (Dutta et al., 2020).

At $z$ = 2–4, the CGM can be probed by the emission lines near bright quasars (e.g., Cantalupo et al. 2014; Martin et al. 2014; Cai et al. 2019; Fossati et al. 2021) and overdense regions (Cai et al., 2017a, b). Surveys to study the small-scale CGM structure and its correlation with the host galax(ies) are somehow scarce (e.g., the Keck Baryonic Structure Survey, Rudie et al. 2012, 2019). The cosmic star formation rate and baryon accretion peak around $z\sim$ 2 (Madau & Dickinson, 2014). In order to trace the multiphase gas-galaxy co-evolution at $z$ = 1–3, especially towards the cosmic noon ($z=$ 2–3), we present a pilot study that takes advantage of the 30 multiband photometry in the Cosmological Evolution Survey (COSMOS) field to search for the CGM host galaxies. In Zou et al. (2021) (hereafter Z21), the authors tentatively search for strong Mg ii absorbers (2 $<z<$ 6) counterparts from Hubble Space Telescope/Canada France Hawaii Telescope/The Dark Energy Camera Legacy Survey deep images. The result indicates that the strong Mg ii absorbing gas tends to have a smaller halo size but a more disturbed environment than that at lower redshift.

In this paper, we will first present CGM gas properties by using the quasars observed in the Dark Energy Spectroscopic Instrument (DESI) survey validation (SV) in the COSMOS and Hyper Suprime-Cam Subaru Strategic Program (HSC) fields, then we will particularly study the CGM-galaxy correlation in the COSMOS field. This paper is presented as follows: we introduce the observation and data analysis in Section 2. The multiphase gas properties are presented in Section 3, and the gas-galaxy correlation is presented in Section 4. We discuss and summarize the implication of this work in Section 5 and 6.

2 Observation and Data Processing

DESI is a Stage IV ground-based dark energy experiment that studies baryon acoustic oscillations and the growth of structure through redshift-space distortions with a wide-area galaxy and quasar redshift survey (DESI Collaboration et al., 2016a, b; DESI collaboration et al., 2022). Full details of the DESI early data release and secondary projects are described in DESI Collaboration et al. (2023a, b). Descriptions of the SV data and data reduction pipeline are presented in Myers et al. (2023) and Guy et al. (2022), respectively. Target selection and samples of quasars, bright galaxies, emission-line galaxies, and luminous red galaxies from the SV data can be found in Yèche et al. (2020); Chaussidon et al. (2022); Raichoor et al. (2022); Yang et al. (2023); Zhou et al. (2023); Lan et al. (2023). Full Mg ii absorber catalog of the DESI EDR data is presented in Napolitano et al. (2023).

This project is a DESI secondary program in the SV. The quasars are selected from the overlapping region of COSMOS and HSC Ultra-Deep (Aihara et al., 2018) fields (2 deg2 and having the tangent point RA, DEC at 150.116, 2.210). Observations of quasars in this work were conducted at Kitt Peak by DESI between March 2021 to May 2021. The average effective exposure time is around 5.5 hours. The DESI SV quasar catalog is created with three algorithms: the DESI pipeline classifier Redrock (RR, Bailey et al. in prep), a broad Mg ii line finder, and a machine learning-based classifier QuasarNET (Busca & Balland, 2018; Farr et al., 2020). The RR algorithm is a template-fitting classifier to classify the quasars using the templates of the different targets (stars, galaxies, and quasars) generated from the SDSS spectra. The Mg ii line finder is an afterburner algorithm using the inputs and outputs from RR. QuasarNET is a deep convolutional neural network classifier. Details of the quasar visual inspection result are presented in Alexander et al. (2022).

From all the observed quasars in the COSMOS/HSC field, we selected 115 quasars each with a mean signal-to-noise ratio (S/N) greater than 3. We calculate the mean S/N of one spectrum as the mean S/N per pixel from three different intervals, where there are no significant emission and absorption lines (flux residuals smaller than 0.5). Details of all quasars used in this work are presented in Appendix Table 5. Among these 114 quasars, 30 quasars are included in the quasar catalog of SDSS release 16 (Lyke et al., 2020) and the rest 84 are new quasars. We show the spectra comparison of one quasar (J095749.98+013354.1) taken by DESI and SDSS is presented in Figure 1. The DESI spectra resolution at 3600 – 9800 Å ranges from 2000 to 5000.

2.1 Normalization and metal lines finder

To detect the intervening absorption systems, we first normalize the quasar spectrum by dividing the flux with the continuum. The continuum defined in this work is the spline line without a significant absorption feature (see the red curve line in Figure 1 for an example). We use two methods to estimate the quasar continuum: the principle component analysis (PCA) method (see e.g., Pâris et al. 2011; Guo et al. 2018) and an unsupervised probabilistic continuum fitting method with uncertainty quantification (Sun et al., 2022)111https://github.com/ZechangSun/QFA/. The fitting spectra redward Ly$\alpha$ emission from these two methods are consistent. The fitting result is then visually checked to avoid any significant improper fitting.

After obtaining the normalized spectra, we use the algorithm described in Z21 to automatically search for metal lines redward of Ly$\alpha$ emission, i.e., rest frame $\lambda$ 1216 – 3000 Å. We briefly introduce the algorithm here. We identify the absorption feature from the normalized spectrum with a Gaussian kernel filter having rest-frame velocity FWHM smaller than 600 km s*-1*. When the rest-frame equivalent width $W_{r}$ of this Gaussian kernel is greater than our detection limit (0.3 Å), this kernel is then identified as an absorption feature. We search for the Mg ii (C iv) doublets with two Gaussian kernels that are separated around 770 km s*-1* (500 km s*-1*). The self-blending of Mg ii and C iv systems can be mitigated by the two Gaussian kernels. We then visually inspect all the detected absorbers to ensure there is no significant blending from absorber systems at other redshifts. The error of the equivalent width is calculated by the flux variance summation over the search boxcar. We also add the SNR constraint in the nearby region of the absorption feature. The final absorber sample is visually checked. The criteria to select the Mg ii (C iv) doublets are as follows:

(1) the local S/N $>3$; where the local S/N is the mean S/N per pixel around $\pm$ 10 pixels adjacent to the search boxcar center.

(2) $W_{r}(\lambda 2796)>$ 0.3 Å and $W_{r}(\lambda 2803)>$ 0.15 Å ($W_{r}(\lambda 1548)>$ 0.3 Å and $W_{r}(\lambda 1550)>$ 0.15 Å) for C iv;

(3) $W_{r}(\lambda 2796)$ / $\sigma$ ($W_{r}$) $>$ 3 and $W_{r}(\lambda 2803)$ / $\sigma$ ($W_{r}$) $>$ 3 ($W_{r}(\lambda 1548)$ / $\sigma$ ($W_{r}$) $>$ 3 and $W_{r}(\lambda 1550)$ / $\sigma$ ($W_{r}$) $>$ 3 for C iv) indicating a 3$\sigma$ detection.

3 Multi-phase gas

The multiplicity of the absorbers is detected (H i, Mg ii, C iv, Al ii, Al iii, Si ii, Si iii and Si iv) from the DESI SV quasar spectra. In this work, we focus on the discussion of H i, Mg ii and C iv. We detect Mg ii and C iv independently using the algorithm described in Section 2.1, the H i systems are then searched in the same system once the Mg ii or C iv are detected.

Quantitatively, we detect 51 Mg ii absorption (0.66 $<z<$ 2.49, 0.27 $<W_{r}<$ 4.89 Å) and 50 C iv absorption (1.35 $<z<$ 3.18, 0.22 $<W_{r}<$ 3.05 Å). In the overlapping redshift coverage between Mg ii and H i ($z=$ 1.95 – 2.50); C iv and H i ($z=$ 1.95 – 3.20), 8 out of 10 (80%) Mg ii systems have H i detection, and 24 out of 29 (82.75%) C iv systems have H i detection. In the redshift coverage of both Mg ii and C iv ($z=$ 1.3 – 2.5), 20 out of 34 (58.82%) Mg ii systems have C iv detection. We found a relatively larger median $W_{r}$ of the C iv systems with Mg ii absorption (0.859 Å) than C iv systems without Mg ii absorption (0.694 Å).

We plot all systems having H i detection in the upper panel of Figure 2. The figure indicates that all of the detected H i absorption associated with Mg ii and C iv systems have log $N$(H i)/cm2 $>$ 14.0. The column density of H i is measured using the Voigt profile (package Voigtfit, Krogager 2018) when there are strong Lyman series lines (e.g., $W_{r}(\lambda 1215)>$ 10 Å) detected at the redshifts of metal line absorber. The total column density of H i is derived from both the Ly$\alpha$ and Ly$\beta$ lines (when detected). Both Ly$\alpha$ and Ly$\beta$ lines are fitted with the same Doppler parameter $b$, which is varied between 40-200 km s*-1*. The final $N$ value and its associated error are taken when the smallest $\chi^{2}$ value is obtained. The complete fit results of Mg ii, C iv and H i systems are presented in Appendix Table 6. It suggests that Mg ii absorbers probe a relatively higher $N$(H i) gas than the C iv absorbers. All of the Mg ii systems have log $N$(H i)/cm2 $>$ 16.0, which is consistent with what previous detections have claimed for Mg ii absorption at $z$ $<$ 1 (e.g., Churchill et al. 2000; Nielsen et al. 2013b). Particularly, five out of seven (71.43%) Mg ii systems are associated with DLA and sub-DLA (19.0 $<$ log $N$(H i)/cm2 $<$ 20.3) systems, indicating a significant correlation between Mg ii systems and large $N$(H i) gas. The C iv systems are likely to probe wider range of $N$(H i) gas than Mg ii: one is associated with a more neutral phase having 18.5 $<$ log $N$(H i)/cm2 $<$ 20 and one is associated with a lower density phase (14.0 $<$ log $N$(H i)/cm2 $<$ 17.0). Note that the system with the highest $N$(H i) has Mg ii but no 3$\sigma$ C iv detection.

3.1 Line density evolution of different ions

In this paragraph, we present the statistical properties of the metal absorbers. Firstly, we calculate the survey completeness by taking into account the contamination such as the skylines and spikes in the spectra. To calculate the survey completeness, we follow and update the algorithm described in Z21. We briefly introduce the algorithm here. The method is to sample uniformly mock Mg ii and C iv doublets in each spectrum using a Monte Carlo simulation. Then, we use the detection algorithm in Section 2.1 to calculate the recovery rate. We generate 10,000 doublets with equivalent widths that vary uniformly distributed between 0.3 and 4.0 Å. We measure the real velocity spread of the absorption profile, $\Delta v_{90}$, defined as the velocity spread containing 90% of the total optical depth of the absorption line (e.g., Prochaska et al. 2008). We fit a relation between $W_{r}$ and velocity spread $\Delta v_{90}$ with a polynomial curve fitting technique considering the errors from two variables of Mg ii and C iv absorbers: $\Delta v_{90}$ = 77 $\times W_{r}(\lambda 2796)$ + 315 km s*-1* and $\Delta v_{90}$ = 81 $\times W_{r}(\lambda 1548)$ + 403 km s*-1*. The $\Delta v_{90}$–$W_{r}$ relation of Mg ii and C iv is plotted in the lower pane of Figure 2. The inserted observed velocity of the absorber is given by the $\Delta v_{90}$–$W_{r}$ relation. We then insert the mock doublets in 10,000 uniformly distributed redshifts in our searching path length (0.5 to 2.6 for Mg ii and 1.3 to 3.2 for C iv). The detection result of every inserted absorber is described as a Heaviside function $H(z,W_{r})$. The path density $g(z,W_{r})$ is the integral of $H(z,W_{r})$ over the total sightline number $N$. The average completeness is then calculated by the average of path density $C=\int_{W_{lim}}^{\infty}g(z,W_{r})/N$ as a function of redshift and $W_{r}$ limit. (see Figure 3).

After applying the survey completeness correction, we calculate the line density per redshift bin ($dN/dz$) of Mg ii and C iv. The line density of Mg ii and C iv follow the Poisson distribution (Zhu & Ménard, 2013), thus, the error of $dN/dz$ is given by 1/$\sqrt{N}$. The final line density of Mg ii are presented in Table 4 and Figure 4. We compare our calculated $dN/dz$ of Mg ii with other surveys at $z=$ 1–2 (Zhu & Ménard, 2013) and $z>2$ (Chen et al., 2017; Zou et al., 2021). Figure 4 clearly shows that the line density evolution of strong Mg ii ($W_{r}>$ 1 Å) is consistent with the cosmic star formation history from the local Universe to $z\sim$ 6. Particularly, the trend has a tentative turnover at $z\sim$ 2. The line densities of weaker Mg ii systems (0.3 Å$<W_{r}<$ 0.6 Å, 0.6 Å$<W_{r}<$ 1.0 Å) increase with the increasing redshift from $z=0.6-2.5$.

4 Absorbing gas-host galaxy correlation

In this section, we first discuss the method of identifying the absorbing gas host galaxy. We then discuss the correlation between the gas phase properties (absorber strengths and ionization states) and the host galaxy properties (stellar mass, halo mass, Virial radius, impact parameter, and star formation rate (SFR)). Fourteen quasar fields among the whole sample (115 quasars) are covered by the COSMOS2020 catalog (Weaver et al., 2022). We focus on the gas-galaxy correlation in the following discussion of these fourteen quasar fields. We compare our results with other surveys such as the MAGIICAT (Nielsen et al., 2013b; Churchill et al., 2013a, b) and the MEGAFLOW (Zabl et al., 2019; Schroetter et al., 2019, 2021) surveys (for Mg ii ) at $z<1$ and $z=1-1.5$, and the Cosmic Origin Spectrograph(COS) survey (for C iv ) at $z<$ 1 for low-mass galaxies (Borthakur et al., 2013; Bordoloi et al., 2014a).

4.1 Galaxy Identification

The identification of the CGM gas host galaxies is one of the key processes in revealing the correlation between the multiphase gas and the host galax(ies). To compare our result with the MUSE surveys, we search for galaxy candidates within a radius of 30$\arcsec$ (257.51 kpc at $z=2.0$) using the COSMOS2020 galaxy catalog (Weaver et al., 2022). COSMOS is the survey that contains half a million galaxies with a limiting magnitude of 27.2 AB in the F814W band. The radius of 30$\arcsec$ is determined to match the MUSE field of view of 1 arcmin. The sky coverage of our survey is 2 deg2. Galaxies are detected in the combined $zYJHK_{s}$ images from the Subaru and Visible Infrared Survey Telescope for Astronomy (VISTA) telescopes (Scoville et al. 2007; Laigle et al. 2016; Weaver et al. 2022). By adding particularly the ultra-VISTA DR4 data and HSC PDR2 (Aihara et al., 2018), the photometric redshift ($z_{phot}$) uncertainties in COSMOS2020 can match those of galaxies 0.7 magnitudes brighter in COSMOS2015 (Laigle et al., 2016). The photometric redshift is measured by the LePhare (Arnouts et al., 2002; Ilbert et al., 2006) and EAZY (Brammer et al., 2008). The likelihood distribution of the $z_{phot}$ is given by the observed photometry and redshift $\mathcal{L}$(data$|z$). The final $z_{phot}$ is the median of the likelihood distribution ($z_{pdf}$) and the error bar is the 1$\sigma$ interval. Details of the photo-$z$ measurement are presented in Weaver et al. (2022). The galaxy stellar mass and SFR are computed using LePhare with the same configuration as COSMOS2015 (see Laigle et al. (2016)). Template library generated from the stellar population synthesis models in Bruzual & Charlot (2003) are fitted to the observed photometry to obtain the galaxy properties. We adopt the ‘best’ value in the catalog which are taken at the minimum chi2. The stellar mass and SFR uncertainties are within the 68% confidence level. In this work, we use THE FARMER v2.1 catalog222cosmos2020.calet.org.

The precision of $z_{phot}$ is quantified with the difference between spectroscopic redshift $z_{spec}$ and photometric redshift: $\sigma=|z_{phot}-z_{spec}|/(1+z_{spec})$. We measure the $z_{phot}$ precision in the COSMOS field with the galaxy $z_{spec}$ from DESI SV data. We find that the median $\sigma$ is around 0.006. We select the absorbing gas host galaxies using criteria as follows: (a) ** $\sigma<$ 0.01; (b) we search the galaxy candidates within a $\pm$1000 km s-1 velocity window with the absorber at $z_{abs}$ in the center, i.e., $c\times|z_{phot}$-$z_{abs}|$/(1+$z_{abs})<$ 1000 km s*-1*. Considering the photometric redshift uncertainty is still large for this velocity window ($\pm$1000 km s*-1*), we include $all$ the galaxies whose $z_{pdf}$ distribution across this velocity window around the absorber’s redshift instead of finding $the$ host galaxy. After applying these criteria, we pre-selected the Mg ii and C iv host galaxy candidates. We plot all the pre-selected galaxy candidates in Figure 5. The blue dashed line in Figure 5 is the stellar mass completeness calculated in Weaver et al. (2022). We also test different galaxy selection criteria by cutting the stellar mass limit (blue dashed line in Figure 5), SFR (log SFR $>$ –1) and $L_{v}$ ($L_{v}/L_{*}>$ 0.01, $L_{v}$ is the V-band luminosity and $L_{*}$ is the characteristic galaxy luminosity at $z\sim$ 2) limits, and find that the major result discussed in the following does not change.

4.2 Galaxy counterparts properties

We first present the Mg ii and C iv host galaxies properties in this section. In Section 4.1, we select multiple galaxy candidates which are potentially associated with the absorbers. In order to minimize the uncertainty given by $z_{phot}$ in finding $the$ most likely absorber host galaxy, we measure a weighted average physical parameter (stellar mass, SFR, and impact parameter $D$) of all the galaxy candidates instead. The stellar mass, SFR, and luminosity contribution from a single galaxy candidate to the average parameter are weighted by their $z_{phot}$ likelihood distribution area under the $\pm$1000 km s*-1* velocity window at the absorber redshift.

In Figure 6, we plot the $M_{*}$-SFR relation of Mg ii and C iv absorbing galaxies. The blue hollow circles represent all the selected galaxy candidates colored by their $z_{phot}$ distribution area probability within the velocity window of $\pm$1000 km s*-1* of the absorber. The filled blue circles are the weighted average parameters in each quasar field. The gray lines are the $M_{*}$-SFR relation of the main-sequence galaxies (MSG) at $z=1.6$ from Speagle et al. (2014) with 3 dexes of scatter. The pre-selected Mg ii and C iv galaxies have stellar masses ranging from 10*∼8.5-11* M⊙. Most of the weighted stellar masses are in the 109-10.5 M⊙ range.

4.3 Correlation between gas and galaxy properties

We then explore the dependence of absorbing gas strengths and ionization state on the host galaxy properties such as SFR, $M_{*}$, V-band luminosity $L_{v}$ and the impact parameter $D$. Empirical correlations between the Mg ii absorbing gas and host galaxy properties at $z<1$ have been reported in the literature, such as the extent of Mg ii absorbing gas $R_{gas}$ with the stellar mass relation (Chen et al., 2010b); the $R_{gas}$ with galaxy luminosity $L_{B}$ relation(Chen et al., 2010a); the $W_{r}$ of Mg ii and the luminosity of [O ii] emission in the host galaxy relation by stacking the SDSS spectra (Ménard et al., 2011) and the Mg ii $W_{r}$-$M_{*}$ relation (Bordoloi et al., 2011). The $W_{r}$-$D$ anti-correlation is reported in extensive work (e.g., Nielsen et al. 2013a; Rubin et al. 2018), and the scatter in this relation is then explained by the segregation in the gas halo mass difference (Churchill et al., 2013a).

In figure 7 and 8, we plot the weighted average $W_{r}$-$M_{*}$, $W_{r}$-$SFR$ , $W_{r}$-$L_{v}/L_{\odot}$ and $W_{r}$-$D/R_{vir}$ relations of Mg ii and C iv  absorber and its host galaxies. The $R_{vir}$ is the virial radius, defined by 200 times overdensity of the critical cosmic density $\rho_{crit}$: $R_{vir}$ = (Mh/(3/4$\pi$)200$\rho_{crit}$)1/3. We calculate the virial radius by converting the stellar mass into the halo mass using the stellar-halo mass relation in Girelli et al. (2020). We present the values of the weighted average galaxy parameters in Table 1 and 2 as well.

We note that for Mg ii, a tighter correlation is seen between the absorber strength with the average galaxy SFR than its luminosity and stellar mass. We assume a power-law correlation between $W_{r}$ and SFR and $L/L_{*}$: log $W_{r}$(Mg ii) = a$\times$log(SFR) + b1 and log $W_{r}$(Mg ii) = a$\times$log($L/L_{*}$) + b2. The maximum likelihood estimation result is log $W_{r}$(Mg ii) = 0.14$\times$log(SFR) + 0.038 (1 $\sigma$). Different than the works in the local universe, we do not see a significant correlation between the Mg ii $W_{r}$ against the stellar mass, $D/R_{vir}$ or luminosity within $D$ = 250 kpc. Furthermore, we detect several Mg ii systems at $D/R_{vir}>$ 1.

For C iv, no significant correlations are seen between the overall C iv equivalent width and the galaxy candidates’ physical parameters. Additionally, we note that the C iv (with Mg ii) systems and C iv-only systems host galaxies do not exhibit significant differences in $M_{*}$, SFR, and $L_{v}$. A marginal trend is seen that the C iv (no Mg ii) systems tentatively reside in a larger impact parameter than C iv (with Mg ii) systems.

To further strengthen our results about the correlation between the absorber and its environment, we generate the posterior distribution $P(r|X$) of the Pearson correlation coefficient $r$ between log$W_{r}$-$M_{*}$, log$W_{r}$-log SFR, and log$W_{r}$-$L_{v}$, where $X$ is the galaxy dataset. We assume there are $n$ galaxies in one quasar field that are associated with the absorber. The $P(r|X)=\sum_{X_{n}}P(r|X_{n})P(X_{n}|X)$. For one quasar field, we have $n$ galaxies. We perform a Markov chain Monte Carlo to sample the dataset. We present the result in Appendix Figure 11. From the figure, we can tell that, for Mg ii, the $W_{r}$ has a more significant correlation with SFR than stellar mass. For C iv, no significant difference is seen between the $W_{r}$-SFR, $W_{r}$-$M_{*}$, and $W_{r}$-$L/L_{\odot}$ relation. This is consistent with the results when including the major galaxy candidates. This method is unbiased in selecting the absorbing host galaxies with a velocity window and considering the $z_{phot}$ uncertainties in a statistical way.

4.4 Covering fraction of Mg ii and C iv absorbing gas

Here we discuss the absorbing galaxy fraction among different galaxy populations above a certain detection limit. The covering fraction $f_{c}$ within an impact parameter $D$ is defined as the probability of a galaxy having a metal absorption line ($W_{r}>W_{lim}$) that can be detected. The probability can be calculated by the ratio of absorbing galaxy number versus the total galaxy number within a certain $D$ (Bordoloi et al., 2014a):

[TABLE]

, for which $N_{abs}$ is the number of galaxies having $W_{r}>W_{lim}$ galaxy number and $N_{tot}$ is the total galaxy number in the same field. We use the same method to select the potential absorbing galaxies that are discussed in Section 4.1.

In our case, we count the number of absorbing galaxies weighted by the $z_{pdf}$ distribution within the velocity window $p_{z}$ of each galaxy. The median $p_{z}$ in each quasar field is dominated by the galaxies that are most correlated with the absorber. We neglect the galaxies whose $p_{z}$ is 1 dex smaller than the major galaxi(es). We divide the differential Mg ii and C iv absorbing galaxies covering fraction at 0.9 $<z<$ 2.2 into three impact parameter bins (50-100, 100-200, and 200-250 kpc). The covering fraction is plotted in Figure 9. We find that within 250 kpc, an anti-correlation between the $f_{c}$ and the $D$ is seen for both Mg ii and C iv. In Section 4, we present that the strength of our Mg ii absorption has the most significant dependence on the SFR of host galaxies, therefore, we check the $f_{c}$ of strong Mg ii in the main sequence galaxies, $f_{c}(MSG)$ = $N_{MgII}/N_{MSG}$. We select the main sequence galaxies based on the relation in Speagle et al. (2014) within 0.3 dex. The $z_{pdf}$ of each galaxy is used for the target selection. We find that the strong Mg ii covering fraction in main-sequence galaxies (dark green circles) is two times higher (0.30 within 100 kpc) than that in the whole galaxy population (blue circles) in Figure 9. Given the arbitrariness of the definition of main sequence galaxies within a 0.3 dex scatter and the uncertainties associated with estimating the main sequence, we further include two subsamples of galaxies with SFRs greater and smaller than the median SFR (green and red circles in Figure 9). The $f_{c}$ value of Mg ii galaxies in the subsample with SFR $>$ median SFR is similar to $f_{c}(MSG)$ and exhibits a significant evolution compared to those in the local star-forming galaxies in Huang et al. (2021).

4.5 Comparison with other surveys

We compare our CGM-galaxy correlation results with other surveys in this section. We are cautious about comparing surveys having different depths and sample selection methods. Therefore we only discuss the trends rather than the quantitative differences here.

For Mg ii, we compare our results with the strong Mg ii-galaxy correlation in the MAGIICAT (Nielsen et al., 2013b), MAGAFLOW survey (Schroetter et al., 2019; Zabl et al., 2019) and the DECaLS imaging survey (Lan, 2020). The galaxies selected in MAGIICAT have a similar depth as ours (B-band magnitude limit of -16.1, i.e., around 24.5 mag at $z=$ 0.34). The DECaLS imagings used in Lan (2020) have $g$ and $z$ band limits of 24.2 and 23 mag, respectively. We perform the Kolmogorov–Smirnov test between the $M_{*}$-SFR relation of the sample and $M_{*}$-SFR of galaxies in the MAGAFLOW inflowing-mode (upward triangles) and outflowing-mode (downward triangles). The $p$-values are 0.014, and 0.0153, respectively, suggesting that our Mg ii gas does not exhibit obvious inflowing or outflowing features. In other words, the strong Mg ii gas at $z=$ 1–2 tentatively exhibits complex kinematics, which may be a combination of effects by gas accretion, galactic outflows and gas recycling, etc.

We compare the Mg ii having $W_{r}>$ 1 Å  gas $f_{c}$ with that at $z=$ 1–1.5 (Lan, 2020) (the grey dashed line and grey line, respectively) and $z<1$ (Nielsen et al., 2013a) (grey triangles). We note that our $f_{c}$ of our strong Mg ii absorbing gas shows a marginal evolution than that at $z<1$ (Nielsen et al., 2013a). The Mg ii-absorbing gas covering fraction in either main-sequence galaxies or star-forming galaxies has a significant evoulution from $z=0$ to $z=$ 1–2.5. Lan (2020) also find that the covering fraction of strong Mg ii systems in star-forming galaxies is higher than that in passive galaxies, exhibiting a significant evolution from $z$ = 0.4 to $z$ = 1.3. Schroetter et al. (2021) provide a novel Hamiltonian Monte Carlo model optimizing for estimating the $f_{c}$ with limited samples. Their $f_{c}$ is for Mg ii systems having $W_{r}>$ 0.6 Å, it is clear that our strong Mg ii gas $f_{c}$ is higher than their $f_{c}$ beyond 100 kpc.

We compare our C iv $M_{*}$-SFR relation with C iv host galaxy from the COS-Dwarfs (Bordoloi et al., 2014b) and COS-Halo surveys (Borthakur et al., 2013) at $z<1$. Bordoloi et al. (2014b) study a sample of 43 sub-$L_{*}$/dwarf galaxies at $z<$ 0.1. A tentative correlation between the C iv $W_{r}$ and host galaxy SFR has been reported in Bordoloi et al. (2014b) at $z<1$. Borthakur et al. (2013) study 20 galaxies at $z<$ 0.2 and find a high detection rate of C iv systems in starburst galaxies. The detection rate of C iv is 4/5 in starburst galaxies(blue triangles in Fig. 8) and 2/12 in the control sample comprising of normal/passive galaxies (grey triangles).

We note that our C iv $f_{c}$, unlike the strong Mg ii gas, is lower than that in the MAGAFLOW survey Schroetter et al. (2021) at $z=$1–1.5 within 250 kpc. This may be because that half of our C iv systems are Mg ii bearing halos, and the galaxies associated with weaker C iv-only halos are underestimated. In Schroetter et al. (2021), the authors suggest that at $z$ = 1–1.5, the covering fraction of C iv-only gas is larger than that of Mg ii and C ivgas within 250 kpc and it likely resides a broader radius in the IGM (out to 250 kpc).

5 Discussion

5.1 Strong Mg ii absorbing gas contribution to global SFR at $z$ = 1–2

We follow the method described in Nestor et al. (2011) to estimate the strong Mg ii gas contribution to the global SFR density:

[TABLE]

, where $<SFR>$ is the mean SFR of Mg ii host galaxies, the dN/dz is the line density of Mg ii absorbers, $dl/dz$ is the differential proper distance, $\sigma$ is the Mg ii gas cross-section and the $\rho_{SFR}$ is the global star formation density. We take our detected strong Mg ii galaxy $<SFR>$ of 7.44 M*⊙* yr*-1*, the dN/dz at 1.5 – 2.0 = 0.753 $\pm$ 0.141, the cross-section of Mg ii gas as (250 kpc)2. The $\rho_{SFR}$ at $z=$ 2.0 calculated from Equation 15 in Madau & Dickinson (2014) is 0.132 M*⊙* yr*-1* Mpc*-3*. We then estimate the contribution of strong Mg ii absorbing gas to the global SFR density at $z\sim$ 2.0 is 0.068 by assuming a gas cross-section of (250 kpc)2. This fraction is consistent with our measured covering fraction of strong Mg ii in star forming galaxy subsample within 250 kpc, further suggesting the co-evolution of cool gas probed by the strong Mg ii systems and the cosmic star formation activity.

5.2 Mg ii and C iv gas origins

Here we discuss how the Mg ii and C iv absorbing gas fuel the galaxy star formation and are affected by the galactic feedback towards the comic noon. According to the results above, we find that two gas phases co-exist in the CGM in this work: one is with detectable Mg ii and C iv absorption in the same system, i.e., same physical position and shares similar kinematics profiles, which probes a relatively higher $N$(H i); one is with the C iv-only systems, probing a lower $N$(H i) gas and a tentative larger impact parameter. Such multiphase structure of the CGM is clearly seen in the hydrodynamic simulations (Ford et al., 2013; Suresh et al., 2017). In Ford et al. (2013), the simulation reveals that within the 300 kpc of galaxies at $z=$ 0.25, the multiphase CGM generally probes a 104-4.5K photo-ionized gas. Low ions such as Mg ii and Si iv probe a denser gas and are closer to galaxies, while C iv can be associated with the cool gas in the inner region or the collisionally dominated highly ionized gas in a broader region. In this work, we mainly consider the CGM within 250 kpc of the galaxy, the gas mainly exhibits “halo fountains” region properties suggested by the simulation of Oppenheimer & Davé (2008). This region is first fueled by the metal-poor inflowing gas but later enriched by the metal-rich and momentum-driven outflows, where the gas has complex kinematics. Recent TNG simulation has the resolution to resolve small-scale structures. The Mg ii halos are found to be highly structured, clumpy, asymmetric (Nelson et al., 2020) and with a variety of kinematics (DeFelippis et al., 2020). This likely explains why our Mg ii absorbing gas does not exhibit obvious inflowing or outflowing features.

We find a tight correlation between the strong Mg ii equivalent width and the host galaxy SFR at $z=$1–2, suggesting the co-evolution of strong Mg ii absorbing gas and the main-sequence galaxies towards the cosmic noon. The conclusion is strengthened by the fact that the covering fraction of Mg ii absorbing gas in the main-sequence galaxies is two times higher than the covering fraction in all the galaxies. Particularly, we detect several Mg ii and C iv systems having D/R${}_{vir}>1$, indicating that the cool gas still exists out of the virial radius. These unbounded gas may be driven by the star formation activity at the cosmic noon. In Rudie et al. (2019), they found that 70% of their galaxies have some unbounded metal-enriched gas, suggesting galactic winds may commonly eject gas from halos at $z\sim 2$.

The C iv only systems, as suggested in the COS-Halos survey in Borthakur et al. (2013) and Bordoloi et al. (2014a), likely trace the dwarf galaxies and/or starburst galaxies. The sub-L∗ galaxies undergo extended bursty star formation rather than the continuous ’normal’ star formation in the super $L_{*}$ galaxies. Though we do not find the host galaxies exhibit significant differences between the C iv-only systems and the C iv(with Mg ii) systems. This may be due to that our galaxy properties are weighted average values, representing a feature of multiple star-forming galaxies.

5.3 Environmental effects

Environmental effects can also give rise to the optically thick cool gas. Environmental effects of the Mg ii absorbing gas are detected in the MAGG survey at $z\sim$ 1–1.5 (Dutta et al., 2020, 2021) and the MAGIICAT survey $z<$ 1 (Nielsen et al., 2018; Huang et al., 2021). Particularly, Nielsen et al. (2018) point out that the covering fraction and median $W_{r}$ of Mg ii absorbing gas residing in a group environment is larger than that of an isolated galaxy. The group environment kinematics are with more power for high-velocity dispersion similar to outflow kinematics. Lofthouse et al. (2023) study the correlation between the log $N$(H i)/cm2 $>$ 19 absorbing gas and Ly$\alpha$ emitters at $z=$ 3–4 and find that the optically thick gas covering fraction in galaxy group is three times higher than that in isolated systems.

In Chen et al. (2010a) and Huang et al. (2021), the authors find that the Mg ii gas that does not show obvious anti-correlation with $D$ resides in a group environment rather than associated with a single galaxy. Thus, the environmental effect is also a plausible explanation of our Mg ii and C iv systems detected out of the virial radius, and no obvious anti-correlation is seen in our Mg ii $W_{f}$ with $D/R_{vir}$. Additionally, the C iv system with the largest equivalent width has three major components in the absorption profile of C iv, Mg ii and Si iv absorption. This system also has a large $D/R_{vir}$ value (1.97), which may indicate it is affected by the outflows or reside in a group or disturbing environment.

5.4 Absorbing gas and non-absorbing gas counterparts

In order to further test if there is an intrinsic correlation between the absorbing gas with the galaxy overdensity, we compare the galaxy density at the redshift of absorbers and non-absorbers. In Appendix Figure 12, we plot the normalized galaxy $z_{phot}$ density P($z$) at 0.0 $<z<$ 6.0. The Mg ii and C iv absorber redshifts are labeled by red and blue lines, respectively. The P($z$) is a normalized photo-$z$ probability distribution by taking around all the galaxies in the COSMOS2020 catalog within 30 $\arcsec$ offset the quasar sightline. We find that the Mg ii and Mg ii(with C iv) absorbers likely occur at the redshift where the galaxy density is higher than the redshifts where there are no absorbers between $z=1-2.5$. The C iv-only systems, i.e., the relatively lower $N$ (H i) systems, exhibit a mild correlation with the galaxy density. We carry out a cross-correlation analysis between the incidences of Mg ii and C iv absorbers, and their rest-frame equivalent width $W_{r}$, with galaxy overdensity. The methodology is similar as which presented in Dutta et al. (2021). The galaxy overdensity, denoted as $\delta$, is defined by the relation $\delta=(\rho_{abs}-\rho_{0})/\rho_{0}$. Here, $\rho_{abs}$ and $\rho_{0}$ are the absorbing-related and field galaxy number density, respectively. We calculate the $\rho_{abs}$ within a velocity window of $\pm$1000 km/s centered on the absorber redshif. The projected area is defined using an annulus with a radius of 250 kpc. The field galaxy number density in a cube volume of 300$\times$300 kpc and d$z$ = 0.2. We plot the relation between Mg ii and C iv $W_{r}$ and incidence with the galaxy number overdensity in Figure 10. From the figure we can tell that the galaxy density is higher when the Mg ii absorber incidence goes higher. We perform a Pearson correlation analysis between Mg ii and C iv incidence between the galaxy overdesnity. The $p$ value for Mg ii and C iv incidence with galaxy number overdenstiy are 0.047 and 0.590, respectively. The $p$ value for Mg ii and C iv $W_{r}$ with galaxy number overdenstiy are 0.838 and 0.175, respectively. This tentatively suggest that the environmental effects indeed play an active role in the origin of strong Mg ii absorbing gas at the cosmic noon. We are somewhat cautious about drawing firm conclusions given the limited sample size. The galaxy density in the large-scale structure may play a more significant role in the birth and evolution of the multiphase CGM than the specific ’host’ galaxy properties.

6 Summary

1. We detect 51 Mg ii and 50 C iv absorption systems separately from 115 DESI SV quasars in the COSMOS+HSC fields. All of the systems having Ly$\alpha$ within the detection limit are with log $N$(H i)/cm2 $>$ 14.0. In the redshift coverage of both Mg ii and C iv ($z$ = 1.3 – 2.5), 20 out of 34 (58.82%) Mg ii systems have C iv  detection.

2. Fourteen quasars are covered by the COSMOS2020 catalog. By cross-matching the COSMOS2020 catalog, we select the Mg ii and C iv-galaxies above the mass limit. The majority of the Mg ii galaxies and C iv  galaxies are classified as the main-sequence star-forming galaxies within 0.3 dex scatter. A tight correlation between the Mg ii equivalent width and the weighted average galaxy SFR is found. The C iv-only galaxies tentatively reside in a larger impact parameter than the systems having both Mg ii and C iv absorption.

3. We find that the covering fraction of strong Mg ii absorbing gas selected galaxies in main sequence galaxies is two times higher than that in all the galaxy populations within 250 kpc. The strong Mg ii contributes $\sim$0.068 star formation to the global star formation at $z\sim$ 2, which is consistent with the covering fraction of the Mg ii gas. The result suggests the co-evolution of cool gas probed by strong Mg ii and the main sequence galaxies at the cosmic noon.

4. We find that the environmental effects and the galaxy density in a large-scale structure tentatively play an active role in the origin of multiphase CGM at $z=$ 1–3.0.

Future JWST observations in the COSMOS field will also provide more information on the galaxy morphology and star formation history of host galaxies. Methods in analyzing gaseous halo and its host galaxies in this work can be tested in the Rubin era when a large quantity of photo-$z$ will be provided.

We would like to thank the anonymous referee for very sconstructive comments. We also thank Solène Chabanier and J. Xavier Prochaska for the constructive comments in the DESI internal review. We thank Clotilde Laigle, Henry J. McCracken, John Weaver, the COSMOS team, Patrick Petitjean, Luis C. Ho, Hsiao-Wen Chen, Yunjing Wu, Ting-Wen Lan, Abhijeet Anand, Antonella Palmese and all the colleagues at the WMAG22 conference for fruitful discussions. SZ, LJ, and ZP acknowledge support from the National Science Foundation of China (11721303, 11890693). SZ, ZC, and ZS are supported by the National Key R&D Program of China (grant no. 2018YFA0404503), the National Science Foundation of China (grant no. 12073014). The science research grants from the China Manned Space Project with No. CMS-CSST2021-A05, and Tsinghua University Initiative Scientific Research Program (No. 20223080023). HZ acknowledges support from the National Science Foundation of China (grant no. 12120101003). This research is supported by the Director, Office of Science, Office of High Energy Physics of the U.S. Department of Energy under Contract No. DE–AC02–05CH11231, and by the National Energy Research Scientific Computing Center, a DOE Office of Science User Facility under the same contract; additional support for DESI is provided by the U.S. National Science Foundation, Division of Astronomical Sciences under Contract No. AST-0950945 to the NSF’s National Optical-Infrared Astronomy Research Laboratory; the Science and Technologies Facilities Council of the United Kingdom; the Gordon and Betty Moore Foundation; the Heising-Simons Foundation; the French Alternative Energies and Atomic Energy Commission (CEA); the National Council of Science and Technology of Mexico (CONACYT); the Ministry of Science and Innovation of Spain (MICINN), and by the DESI Member Institutions: https://www.desi.lbl.gov/collaborating-institutions. The authors are honored to be permitted to conduct scientific research on Iolkam Du’ag (Kitt Peak), a mountain with particular significance to the Tohono O’odham Nation.