HERSCHEL OBSERVATIONS OF MAJOR MERGER PAIRS AT z = 0: DUST MASS AND STAR FORMATION

Galaxy interactions and mergers are important external mechanisms for triggering galactic evolution (Kormendy & Kennicutt 2004). Both observations and numerical simulations have shown that dynamical instabilities induced by galaxy interactions can cause cold gas inflow and massive star formation in the central region of disk galaxies (see, e.g., Bahcall et al. 1995; Dasyra et al. 2006; Hopkins et al. 2006). Merger-induced star formation was first predicted by Toomre & Toomre (1972) and confirmed by Larson & Tinsley (1978) in a study of optical colors of Arp galaxies. Many subsequent studies of the H_α emission and Far-infrared (FIR) emission in Arp galaxies and in paired galaxies provided further support of this theory (see Kennicutt 1996 for a review). On the other hand, Haynes & Herter (1988) found little or no enhanced FIR emission in a sample of optically selected pairs compared to a control sample of single galaxies. In a more influential paper, Bergvall et al. (2003) reported a multi-wavelength study in which they found no significant star formation rate (SFR) enhancement for a sample of morphologically selected merger candidates. Apparently, only some merging galaxies have significantly enhanced SFR, with ultra-luminous infrared galaxies (ULIRGs) as the extreme examples, and the others do not. Therefore, whether the mean SFR of a merger sample shows significant enhancement depends very much on how it is selected. Kennicutt et al. (1987) and Bushouse et al. (1988) found that merger candidates that show strong signs of tidal interactions have significantly stronger SFR enhancement than optically selected paired galaxies, the latter being only marginally enhanced (a factor of ∼2) compared to single galaxies. Telesco et al. (1988) found a strong tendency for pairs with the highest FIR color temperatures to have the smallest separation. Xu & Sulentic (1991) showed that the enhancement of the FIR emission of close spiral–spiral (S+S) pairs with separation less than the size of the primary and with signs of interaction is significantly stronger than that of wider pairs and pairs without interaction signs. Sulentic (1989) found that elliptical–elliptical (E+E) pairs are equally quiet in the FIR emission as single ellipticals. Very few E's in S+E pairs are FIR bright, possibly cross-fueled by their S companions (Domingue et al. 2003).

More recently, large digitized surveys (e.g., SDSS, 2MASS, 2df, etc.) enabled large and homogeneously selected pair samples. A clear anti-correlation between the specific SFR (sSFR = SFR/M_star) and the pair separation has been well established (Barton et al. 2000; Lambas et al. 2003; Alonso et al. 2004; Nikolic et al. 2004; Ellison et al. 2008; Li et al. 2008; Scudder et al. 2012; Patton et al. 2013). In particular, close pairs with projected separation ${r}_{{\rm{proj}}}\leqslant 20\;{h}^{-1}\;\mathrm{kpc}$ have relatively strong sSFR enhancement of a factor of ≳2 (Ellison et al. 2008), while the sSFR enhancement of wider pairs is significantly weaker (Scudder et al. 2012; Patton et al. 2013). On the other hand, not all star-forming galaxies (SFGs) in close pairs have enhanced sSFRs. Xu et al. (2010) studied the sSFR enhancement in a sample of K-band-selected close major-merger pairs (primary-to-secondary mass ratio ${m}_{{\rm{pri}}}/{m}_{{\rm{2nd}}}\leqslant 2.5$) using Spitzer FIR observations and found that (1) on average, SFGs in S+E pairs do not show any sSFR enhancement compared to their counterparts in a mass-matched control sample; and (2) the sSFR enhancement in S+S major-merger pairs is mass dependent in the sense that significant sSFR enhancement is confined to massive SFGs while no enhancement is found in low-mass SFGs with nearly equal (low) mass companions. Using data obtained in Herschel (PEP/HerMES) and Spitzer surveys of the COSMOS field (Scoville et al. 2007), Xu et al. (2012a) studied the cosmic evolution of the sSFR enhancement in close major-merger pairs since z  =  1, and found that the sSFR enhancement in massive S+S pairs decreases with increasing redshift, while there is no significant sSFR enhancement in massive S+E pairs at any redshift. Other authors also found evidence for the lack of SFR enhancement in S+E pairs (Hwang et al. 2011; Moon & Yoon 2015) or low SFR activity in spiral galaxies with early-type close neighbors (Park & Choi 2005, 2009; Park et al. 2008). The literature results on the sSFR enhancement in close minor-merger pairs are controversial. Woods & Geller (2007) found that in close minor mergers, only secondary companions are sSFR enhanced, while there is no significant enhancement in primary companions. This is the opposite of the results of Lambas et al. (2003) that show stronger sSFR enhancement in the primaries. More recently, Scudder et al. (2012) found that both the primaries and the secondaries have significantly enhanced sSFRs in close minor-merger pairs. Lanz et al. (2013) studied SEDs for 31 interacting galaxies in 14 major and minor-merger systems, and found increases in SFR but not sSFR as the interaction sequences progress from non-interacting to strongly interacting.

In this paper, we present new FIR imaging observations in the six photometry bands (70, 100, 160, 250, 350, and 500 μm) of Herschel PACS (Poglitsch et al. 2010) and SPIRE (Griffin et al. 2010), for a large and complete sample of close major-merger pairs of SFGs (including both S+S and S+E pairs). The focus of this new study is on the dependence of the SFR enhancement on the dust mass, which is a good proxy of gas mass. Star formation activity in galaxies is fueled by cold gas, which dominates the gas mass, and all SFGs are gas-rich. Does the SFR enhancement depend on the gas content of a paired SFG? Both in simulations (Hopkins et al. 2009; Perret et al. 2014; Scudder et al. 2015) and in observations (Xu et al. 2012a) there have been indications that the SFR enhancement may decrease with increasing gas fraction. According to Hopkins et al. (2009) (see also Mihos & Hernquist 1996), this is because the gravitational torque imposed by the stellar disk on the gas disk is less effective when the gas fraction is high, therefore less disk gas can sink to the nuclear region to fuel the merger-induced starburst by losing angular momentum to stars. Maps of the dust emission in the six Herschel bands enabled us to accurately estimate the dust mass (and its distribution) for individual SFGs. All galaxies in our Herschel sample, a subset of the KPAIR sample of K-band selected major-merger pairs (Domingue et al. 2009; Xu et al. 2012b), have ${M}_{{\rm{star}}}\gt {10}^{10}\ {M}_{\odot }$ and normal metallicity. Therefore their gas mass is likely to be related to dust mass with a rather constant gas-to-dust ratio of ∼100 (Draine & Li 2007), and the sSFR versus M_gas relation for these paired SFGs can be explored using the Herschel observations. It is worth noting that it is difficult to directly measure the gas mass in individual galaxies in close major-merger pairs. The angular resolutions of single-dish HI 21 cm line observations are too coarse ($\mathrm{beam}\gtrsim {5}^{\prime }$) to resolve pairs into individual galaxies, while the interferometry observations using the VLA are very expensive in terms of the integration time.

The rest of the paper is arranged as follows. Section 2 describes the selection of the H-KPAIR sample, and the details of the Herschel observation and photometry are given in Section 3. Section 4 shows the images and catalogs of the H-KPAIR galaxies and describes the estimation of dust mass and SFRs using SED fittings, while we describe the selection, data reduction, and photometry on control sample galaxies in Section 5. Sections 6 and 7 show statistical comparison results on sSFRs, total gas masses, and star formation efficiencies (SFEs) in H-KPAIRs and the control sample. Discussions will be given in Section 8. We summarize our results and briefly introduce future plans in Section 9.

Throughout this paper, we adopt the Λ-cosmology with Ω_m  =  0.3 and ${{\rm{\Omega }}}_{{\rm{\Lambda }}}=0.7$, and ${H}_{{\rm{0}}}=70\ (\mathrm{km}\;{{\rm{s}}}^{-1}\;{\mathrm{Mpc}}^{-1})$.

The local galaxy pairs sample used in this work (hereafter Herschel KPAIR sample, or H-KPAIR) was constructed from the KPAIR, which is a complete and unbiased K-band selected sample of 170 close major-merger galaxy pairs (see details in Domingue et al. 2009; Xu et al. 2010). Pairs in the KPAIR with the following characteristics are excluded from H-KPAIR: (1) Elliptical+Elliptical (E+E) pairs; (2) pairs with only one measured redshift; (3) pairs with recession velocities <2000 km s⁻¹. The resulting sample contains 88 galaxy pairs (176 paired galaxies), of which 44 are Spiral+Spiral (S+S) pairs and 44 are Spiral+Elliptical (S+E) pairs. The interaction morphology of these pairs was visually checked by three of us (C. C., C. K. X., and D. D). Accordingly, pairs are classified into three types: (1) INT (with signs of interacting, e.g., morphological distortions, tidal tails and bridges, plumes etc.), (2) MER (merging systems) and (3) JUS ("just" pairs without clear signs of interaction). All galaxies have $z\lt 0.1$, with a median of z  =  0.04.

3.1. PACS and SPIRE Images

Pairs in H-KPAIR were observed successfully using PACS and SPIRE photometers in the three PACS bands (70, 100, 160 μm) and three SPIRE bands (250, 350, 500 μm). The observations of 83 pairs were carried out within our own proposal (OT2_cxu_2). All but two of these pairs are smaller than 2', while the sizes of two large pairs (J1406+5043 and J2047+0019) are between 3' and 5'.

PACS observations were done in the scanmap mode with medium scan speed (20''/s). Each pair was observed with four concatenated PACS observations. The first two are for the 70 and 160 μm bands with orientation angles of +45° (nominal) and −45° (orthogonal), respectively, and the other two are for the 100 and 160 μm bands of the same cross-scan configuration. For small pairs (<2'), each scan map has six scans with separation of 30'' and scan length = 3. This yields nearly uniform coverage of a 3' × 3' region, leaving enough margins outside the pairs for background determination. The average 4σ sensitivity limits are 36, 42, and 56 mJy for the three PACS bands. For large pairs we used scan number = 10, separation = 36'', and scan length = 6', yielding a uniform coverage of a 6' × 6' region and the average 4σ sensitivity limits of 30, 36, and 48 mJy. Data reduction was carried out using the Herschel Interactive Processing Environment (HIPE; Ott 2010), with the mapmaking done using UNIMAP (Traficante et al. 2011). The mean beam size (FWHM) are 57, 68, and 120 for 70, 100, and 1600 μm images, respectively.

SPIRE observations of small pairs were done in the small-map mode, providing uniformly covered cross-scan maps of 4' × 4' in 3 SPIRE bands. For large pairs, the observations were done in the large-map mode with nominal scan speed (30''/s), both scan length and scan height equal to 6', and in the A and B directions. For both the small and large maps, each observation has 4 repeats, yielding confusion limited 4σ sensitivities of 29, 30, and 34 mJy in the 3 SPIRE bands, respectively. The SPIRE data were reduced using HIPE 10.0.0, which uses the de-striper for the mapmaking. Bad pixels (PSWF8, PSWE9, PSWB5, PSWD11) were masked. The mean beam FWHM are 182, 249, and 363 for 250, 350, and 500 μm images, respectively.

The remaining 5 pairs in H-KPAIR were observed by other OT and KPGT projects, and their Herschel data were taken from the archives. SPIRE images of two pairs (J1101+5720 and J1429+3534) were taken directly from HerMES data release v2. In Table 1 we present the pair and galaxy ID, R.A., and decl., redshift (z), Ks-band magnitude (from 2MASS), morphological types (spiral or elliptical), interaction morphological types (as described in Section 2), and Herschel proposal ID for 176 paired galaxies in the H-KPAIR sample. SDSS, PACS, and SPIRE images of all pairs are shown in Figure 1.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

View figure set (22 panels)

Tarball of figure set images

Table 1.  H-KPAIR Galaxy Sample

(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
Pair ID	Galaxy ID	R.A.	decl.	z	Ks	Type	Int-type	Herschel-ID	s(p)	δ(Vz)
(KPAIR)	(2MASX)	(J2000)	(J2000)	redshift	(mag)				(kpc)	(km s−1)
J0020+0049	J00202580+0049350	5.107446	0.826346	0.014864	10.98	S	JUS	OT2_cxu_2	11.44	402.0
	J00202748+0050009	5.114637	0.833554	0.016204	10.5	E	JUS
J0118−0013	J01183417−0013416	19.642259	−0.228261	0.045344	12.05	S	JUS	OT2_cxu_2	27.23	54.0
	J01183556−0013594	19.648352	−0.233075	0.045524	12.88	S	JUS
J0211−0039	J02110638−0039191	32.776675	−0.655108	0.017746	11.42	S	JUS	OT2_cxu_2	10.63	96.0
	J02110832−0039171	32.784577	−0.654777	0.018066	10.9	S	JUS
J0338+0109	J03381222+0110088	54.550919	1.16912	0.039201	12.42	S	INT	OT2_cxu_2	24.9	432.9
	J03381299+0109414	54.554136	1.161523	0.040644	12.77	E	INT
J0754+1648	J07543194+1648214	118.633109	16.805959	0.045869	12.03	S	MER	OT2_cxu_2	13.89	104.1
	J07543221+1648349	118.634232	16.809722	0.046216	11.55	S	MER
J0808+3854	J08083377+3854534	122.14073	38.914841	0.040173	12.37	S	JUS	OT2_cxu_2	18.71	23.7
	J08083563+3854522	122.148487	38.914504	0.040094	11.88	E	JUS
J0823+2120	J08233266+2120171	125.886114	21.338096	0.018076	12.15	S	JUS	OT2_cxu_2	15.44	0.3
	J08233421+2120515	125.892554	21.34764	0.018075	11.52	S	JUS
J0829+5531	J08291491+5531227	127.312147	55.522992	0.025059	11.73	S	JUS	OT2_cxu_2	27.75	55.2
	J08292083+5531081	127.336797	55.518931	0.025243	11.42	S	JUS
J0836+4722	J08364482+4722188	129.186777	47.371907	0.05256	12.28	S	JUS	OT2_cxu_2	15.8	0.0
	J08364588+4722100	129.191208	47.369469	0.05256	11.9	S	JUS
J0838+3054	J08381759+3054534	129.573312	30.914861	0.047559	12.65	S	INT	OT2_cxu_2	9.09	152.7
	J08381795+3055011	129.5748	30.916973	0.048068	11.94	S	INT
J0839+3613	J08385973+3613164	129.748888	36.221234	0.055618	12.04	E	JUS	OT2_cxu_2	26.58	231.6
	J08390125+3613042	129.755226	36.217846	0.054846	12.55	S	JUS
J0841+2642	J08414959+2642578	130.456662	26.716073	0.084834	12.27	S	JUS	OT2_cxu_2	30.64	300.0
	J08415054+2642475	130.460609	26.713209	0.085834	12.99	E	JUS
J0906+5144	J09060283+5144411	136.511812	51.744702	0.02912	11.68	E	JUS	OT2_cxu_2	24.29	3.3
	J09060498+5144071	136.520782	51.735305	0.029131	11.95	S	JUS
J0912+3547	J09123636+3547180	138.151523	35.788345	0.023532	11.53	E	JUS	OT2_cxu_2	14.24	6.0
	J09123676+3547462	138.153186	35.796182	0.023512	12.34	S	JUS
J0913+4742	J09134461+4742165	138.435884	47.704599	0.051226	11.91	E	INT	OT2_cxu_2	24.32	450.9
	J09134606+4742001	138.441955	47.700051	0.052729	12.1	S	INT
J0915+4419	J09155467+4419510	138.977827	44.33086	0.039568	12.23	S	MER	OT1_dsanders_1	9.69	2.4
	J09155552+4419580	138.981364	44.332785	0.039576	11.31	S	MER
J0926+0447	J09264111+0447247	141.671319	4.790211	0.089132	13.16	S	MER	OT2_cxu_2	8.13	483.6
	J09264137+0447260	141.672412	4.790559	0.090744	12.48	S	MER
J0937+0245	J09374413+0245394	144.433951	2.760819	0.024156	10.01	S	INT	OT2_cxu_2	25.64	197.7
	J09374506+0244504	144.437634	2.74737	0.023497	10.45	E	INT
J1010+5440	J10100079+5440198	152.503332	54.672181	0.046004	12.2	S	MER	OT2_cxu_2	13.92	83.7
	J10100212+5440279	152.508859	54.674434	0.046283	12.34	S	MER
J1015+0657	J10155257+0657330	153.969058	6.959172	0.029931	12.28	S	JUS	OT2_cxu_2	13.02	244.5
	J10155338+0657495	153.972451	6.963758	0.029116	11.3	E	JUS
J1020+4831	J10205188+4831096	155.216286	48.519406	0.052968	13.26	S	INT	OT2_cxu_2	25.94	45.0
	J10205369+4831246	155.223658	48.523383	0.053118	12.27	E	INT
J1022+3446	J10225647+3446564	155.73532	34.78235	0.05537	13.18	S	INT	OT2_cxu_2	11.56	301.5
	J10225655+3446468	155.735626	34.779681	0.056375	12.39	S	INT
J1023+4220	J10233658+4220477	155.902453	42.346583	0.045621	12.36	S	INT	OT2_cxu_2	15.98	53.1
	J10233684+4221037	155.903524	42.351041	0.045444	13.01	S	INT
J1027+0114	J10272950+0114490	156.872904	1.246742	0.023592	11.79	S	INT	OT2_cxu_2	14.21	64.8
	J10272970+0115170	156.873782	1.254605	0.023376	10.9	E	INT
J1032+5306	J10325316+5306536	158.22153	53.114916	0.06403	12.62	S	INT	OT2_cxu_2	8.32	44.7
	J10325321+5306477	158.221747	53.113263	0.063881	12.15	E	INT
J1033+4404	J10332972+4404342	158.373843	44.076176	0.052303	11.91	S	JUS	OT2_cxu_2	27.42	64.2
	J10333162+4404212	158.381778	44.072572	0.052089	12.37	S	JUS
J1036+5447	J10364274+5447356	159.178124	54.79323	0.045841	12.47	S	INT	OT2_cxu_2	16.9	17.1
	J10364400+5447489	159.183338	54.796929	0.045784	11.63	E	INT
J1039+3904	J10392338+3904501	159.847442	39.080611	0.043464	12.49	S	JUS	OT2_cxu_2	20.35	43.2
	J10392515+3904573	159.854824	39.082589	0.04332	12.29	E	JUS
J1043+0645	J10435053+0645466	160.960666	6.76296	0.028694	11.96	S	INT	OT2_cxu_2	23.23	173.7
	J10435268+0645256	160.969501	6.75702	0.028115	12.2	S	INT
J1045+3910	J10452478+3910298	161.353285	39.174955	0.026834	11.63	S	JUS	OT2_cxu_2	22.72	340.5
	J10452496+3909499	161.354008	39.163873	0.025699	11.41	E	JUS
J1051+5101	J10514368+5101195	162.93181	51.022155	0.025027	10.27	E	JUS	OT2_cxu_2	6.99	366.0
	J10514450+5101303	162.935331	51.025075	0.023807	10.97	S	JUS
J1059+0857	J10595869+0857215	164.994547	8.955983	0.063163	12.01	E	JUS	OT2_cxu_2	21.62	140.1
	J10595915+0857357	164.996465	8.95992	0.062696	12.95	S	JUS
J1101+5720	J11014357+5720058	165.431565	57.334969	0.046903	12.46	E	JUS	KPGT_soliver_1	28.03	276.0
	J11014364+5720336	165.431837	57.342686	0.047823	13.17	S	JUS
J1106+4751	J11064944+4751119	166.706021	47.853312	0.064324	12.66	S	INT	OT2_cxu_2	17.92	334.8
	J11065068+4751090	166.711183	47.85252	0.06544	12.44	S	INT
J1120+0028	J11204657+0028142	170.194061	0.470612	0.025534	11.78	S	JUS	OT2_cxu_2	12.33	25.2
	J11204801+0028068	170.200062	0.468579	0.025618	11.0	S	JUS
J1125+0226	J11251704+0227007	171.321025	2.450198	0.050379	12.75	S	INT	OT2_cxu_2	12.99	84.0
	J11251716+0226488	171.321518	2.446913	0.050659	12.14	S	INT
J1127+3604	J11273289+3604168	171.887066	36.071335	0.035053	11.67	S	JUS	OT2_cxu_2	27.43	25.2
	J11273467+3603470	171.894471	36.063083	0.035137	10.82	S	JUS
J1137+4728	J11375476+4727588	174.478176	47.466349	0.034334	11.28	E	JUS	OT2_cxu_2	26.63	137.7
	J11375801+4728143	174.491711	47.470652	0.033875	11.86	S	JUS
J1144+3332	J11440335+3332062	176.013986	33.535072	0.031799	11.74	E	INT	OT2_cxu_2	20.48	100.5
	J11440433+3332339	176.018049	33.54277	0.031464	12.66	S	INT
J1148+3547	J11484370+3547002	177.182095	35.783414	0.064099	12.81	S	INT	OT2_cxu_2	29.11	144.0
	J11484525+3547092	177.188549	35.785896	0.063619	11.94	S	INT
J1150+3746	J11501333+3746107	177.555564	37.769659	0.055042	12.7	S	INT	OT2_cxu_2	25.36	149.4
	J11501399+3746306	177.558293	37.775167	0.05554	12.44	S	INT
J1150+1444	J11505764+1444200	177.740186	14.738912	0.057202	12.69	S	INT	OT2_cxu_2	17.17	300.0
	J11505844+1444124	177.743517	14.736805	0.056202	11.73	E	INT
J1154+4932	J11542299+4932509	178.595801	49.547493	0.070154	13.0	S	INT	OT2_cxu_2	8.23	308.4
	J11542307+4932456	178.596149	49.546019	0.071182	12.02	E	INT
J1202+5342	J12020424+5342317	180.517927	53.708769	0.064651	12.97	S	INT	OT2_cxu_2	27.14	210.3
	J12020537+5342487	180.522233	53.713478	0.06395	12.43	E	INT
J1205+0135	J12054066+0135365	181.419422	1.59349	0.022002	12.1	S	JUS	OT2_cxu_2	30.83	332.4
	J12054073+0134302	181.419711	1.575058	0.020894	11.21	E	JUS
J1211+4039	J12115507+4039182	182.97949	40.655067	0.022854	11.82	S	INT	OT2_cxu_2	7.73	195.3
	J12115648+4039184	182.98535	40.655124	0.023505	11.98	S	INT
J1219+1201	J12191719+1200582	184.821646	12.01619	0.027303	11.83	E	INT	OT2_cxu_2	13.18	160.8
	J12191866+1201054	184.827781	12.018189	0.026767	12.8	S	INT
J1243+4405	J12433887+4405399	190.911978	44.094436	0.041791	12.09	S	JUS	OT2_cxu_2	22.6	249.9
	J12433936+4406046	190.914009	44.101295	0.040958	11.94	E	JUS
J1252+4645	J12525011+4645272	193.208797	46.757577	0.061331	12.42	S	JUS	OT2_cxu_2	27.67	104.7
	J12525212+4645294	193.21717	46.758185	0.060982	12.2	E	JUS
J1301+4803	J13011662+4803366	195.319279	48.06018	0.030278	12.2	S	INT	OT2_cxu_2	11.84	129.6
	J13011835+4803304	195.326491	48.058471	0.029846	12.67	S	INT
J1308+0422	J13082737+0422125	197.114112	4.370179	0.025476	13.39	S	JUS	OT2_cxu_2	18.52	66.0
	J13082964+0422045	197.123418	4.367971	0.025696	12.44	S	JUS
J1313+3910	J13131429+3910360	198.309546	39.176686	0.071586	12.24	E	INT	OT2_cxu_2	8.29	0.0
	J13131470+3910382	198.311265	39.177305	0.071586	13.1	S	INT
J1315+4424	J13151386+4424264	198.807791	44.407356	0.03586	12.01	S	INT	OT1_rmushotz_1	27.84	35.1
	J13151726+4424255	198.821937	44.407105	0.035743	11.23	S	INT
J1315+6207	J13153076+6207447	198.878026	62.129135	0.030566	11.99	S	INT	OT1_dsanders_1	22.39	6.0
	J13153506+6207287	198.896085	62.124613	0.030586	11.54	S	INT
J1332−0301	J13325525−0301347	203.230301	−3.026349	0.049306	12.95	S	INT	OT2_cxu_2	21.45	297.0
	J13325655−0301395	203.235759	−3.027679	0.048316	12.19	S	INT
J1346−0325	J13462001−0325407	206.583483	−3.428114	0.024781	11.19	S	JUS	OT2_cxu_2	25.03	222.0
	J13462215−0325057	206.592397	−3.418283	0.025521	11.66	E	JUS
J1400−0254	J14003661−0254327	210.152556	−2.909102	0.025579	11.49	S	INT	OT2_cxu_2	12.23	391.5
	J14003796−0254227	210.15818	−2.906313	0.026884	11.73	S	INT
J1400+4251	J14005783+4251203	210.240969	42.85566	0.032741	11.87	S	INT	OT2_cxu_2	27.27	234.3
	J14005879+4250427	210.244986	42.845198	0.033522	12.18	S	INT
J1405+6542	J14055079+6542598	211.461632	65.716632	0.03063	12.8	S	JUS	OT2_cxu_2	23.27	39.6
	J14055334+6542277	211.472286	65.707721	0.030762	12.0	E	JUS
J1406+5043	J14062157+5043303	211.589913	50.725106	0.006456	9.7	S	JUS	OT2_cxu_2	25.21	240.6
	J14064127+5043239	211.671969	50.723327	0.007258	9.56	E	JUS
J1407−0234	J14070703−0234513	211.779296	−2.580923	0.058555	12.44	S	INT	OT2_cxu_2	14.43	300.0
	J14070720−0234402	211.780016	−2.577849	0.057555	12.96	E	INT
J1423+3400	J14234238+3400324	215.926602	34.009016	0.013553	11.48	S	JUS	OT2_cxu_2	16.14	294.0
	J14234632+3401012	215.943004	34.017002	0.012573	11.04	S	JUS
J1424−0304	J14245831−0303597	216.242996	−3.066606	0.052517	12.37	S	JUS	OT2_cxu_2	13.91	300.0
	J14245913−0304012	216.246381	−3.067027	0.053517	11.9	S	JUS
J1425+0313	J14250552+0313590	216.272925	3.233119	0.037083	11.98	E	INT	OT2_cxu_2	22.37	114.0
	J14250739+0313560	216.280712	3.232055	0.037463	12.97	S	INT
J1429+3534	J14294766+3534275	217.448585	35.574313	0.028996	10.93	S	JUS	KPGT_soliver_1	22.07	180.3
	J14295031+3534122	217.459663	35.570065	0.029597	11.9	S	JUS
J1433+4004	J14334683+4004512	218.445139	40.080911	0.026047	10.78	S	INT	OT2_cxu_2	28.01	114.0
	J14334840+4005392	218.451475	40.094155	0.026427	11.17	S	INT
J1444+1207	J14442055+1207429	221.085649	12.128593	0.030445	11.51	S	MER	OT2_cxu_2	8.26	299.4
	J14442079+1207552	221.086655	12.132	0.031443	10.72	S	MER
J1500+4317	J15002374+4316559	225.098935	43.282197	0.031088	11.34	E	JUS	OT2_cxu_2	14.58	147.0
	J15002500+4317131	225.10417	43.286997	0.031578	11.73	S	JUS
J1505+3427	J15053137+3427534	226.380709	34.464851	0.074528	12.98	S	INT	OT2_cxu_2	9.55	300.0
	J15053183+3427526	226.38266	34.464635	0.073528	12.36	E	INT
J1506+0346	J15064391+0346364	226.682879	3.776844	0.036278	11.48	S	JUS	OT2_cxu_2	25.24	315.0
	J15064579+0346214	226.690883	3.772574	0.035228	11.6	S	JUS
J1510+5810	J15101587+5810425	227.566059	58.178518	0.030343	11.77	S	JUS	OT2_cxu_2	10.54	402.0
	J15101776+5810375	227.574164	58.176979	0.031683	12.35	S	JUS
J1514+0403	J15144544+0403587	228.689368	4.06633	0.0386	11.89	S	INT	OT2_cxu_2	18.81	186.0
	J15144697+0403576	228.695709	4.066007	0.03922	12.06	S	INT
J1523+3748	J15233768+3749030	230.907038	37.817522	0.023365	12.55	S	INT	OT2_cxu_2	20.09	78.3
	J15233899+3748254	230.912489	37.807073	0.023626	12.43	E	INT
J1526+5915	J15264774+5915464	231.698953	59.262907	0.044712	12.45	S	INT	OT2_cxu_2	8.77	224.1
	J15264892+5915478	231.703856	59.263304	0.045459	12.13	E	INT
J1528+4255	J15281276+4255474	232.053241	42.929924	0.018839	10.02	S	INT	OT2_cxu_2	26.53	243.0
	J15281667+4256384	232.069607	42.944107	0.018029	10.59	S	INT
J1552+4620	J15523258+4620180	238.135769	46.338356	0.059385	12.07	E	INT	OT2_cxu_2	19.46	489.0
	J15523393+4620237	238.141398	46.339933	0.061015	12.75	S	INT
J1556+4757	J15562191+4757172	239.091225	47.954829	0.019103	12.1	S	JUS	OT2_cxu_2	22.96	240.0
	J15562738+4757302	239.114264	47.958429	0.019903	12.16	E	JUS
J1558+3227	J15583749+3227379	239.656243	32.460535	0.049368	13.16	S	INT	OT2_cxu_2	10.8	261.0
	J15583784+3227471	239.657667	32.463088	0.048498	12.29	S	INT
J1602+4111	J16024254+4111499	240.677392	41.197267	0.033536	11.69	S	JUS	OT2_cxu_2	18.66	69.0
	J16024475+4111589	240.686475	41.199706	0.033306	12.5	S	JUS
J1608+2529	J16080559+2529091	242.023328	25.485866	0.041547	11.94	S	MER	OT2_cxu_2	10.96	224.1
	J16080648+2529066	242.02704	25.485182	0.042294	11.32	S	MER
J1608+2328	J16082261+2328459	242.094219	23.479425	0.04092	13.17	S	JUS	OT2_cxu_2	22.25	31.8
	J16082354+2328240	242.098125	23.473347	0.040814	12.44	S	JUS
J1614+3711	J16145418+3711064	243.72577	37.185123	0.058169	12.13	S	INT	OT2_cxu_2	9.13	0.0
	J16145421+3711136	243.725893	37.187131	0.058169	12.04	E	INT
J1628+4109	J16282497+4110064	247.104069	41.168463	0.033017	11.47	S	JUS	OT2_cxu_2	27.9	370.8
	J16282756+4109395	247.114849	41.161	0.031781	11.52	S	JUS
J1635+2630	J16354293+2630494	248.928908	26.513727	0.070061	12.29	S	INT	OT2_cxu_2	14.99	378.0
	J16354366+2630505	248.931925	26.514045	0.071321	12.22	E	INT
J1637+4650	J16372583+4650161	249.357631	46.837824	0.057817	11.8	S	JUS	OT2_cxu_2	25.82	299.7
	J16372754+4650054	249.364787	46.834858	0.056818	12.44	S	JUS
J1702+1859	J17020320+1900006	255.513366	19.000181	0.057322	12.39	E	INT	OT2_cxu_2	17.14	462.6
	J17020378+1859495	255.515763	18.997088	0.05578	13.18	S	INT
J1704+3448	J17045089+3448530	256.212001	34.814721	0.057213	13.06	S	INT	OT2_cxu_2	11.74	270.0
	J17045097+3449020	256.212288	34.817342	0.056313	12.4	S	INT
J2047+0019	J20471908+0019150	311.829434	0.320801	0.012971	9.08	S	JUS	OT2_cxu_2	28.98	414.0
	J20472428+0018030	311.851263	0.300826	0.011591	9.74	E	JUS

Note. Descriptions of columns: (1) Pair ID. The designations are "KPAIR J0020+0049," etc. (2) Galaxy ID, taken from 2MASS. (3) R.A. (deg, J2000). (4) decl. (deg, J2000). (5) redshift z taken from SDSS or other telescopes. (6) ${K}_{\mathrm{rms}}$ (${K}_{{\rm{20}}}$) magnitude taken from 2MASS. (7) Morphological type (S: Spirals, E: Ellipticals). (8) Interacting morphological type (INT: with interaction signs; MER: mergers; JUS: without interacting signs, "just" pairs). (9) Herschel proposal ID. (10) Projected separation (s(p)) between two galaxies in pairs, in kpc. (11) The difference in radial velocity (δ(Vz)) between two galaxies in pairs, in km s⁻¹.

A machine-readable versions of the table is available.

Download table as:  DataTypeset images: 1 2 3

3.2. PACS and SPIRE Photometry

The dust mass M_dust and integrated infrared luminosity L_IR (8–1000 μm) of individual galaxies are estimated via FIR-submillimeter spectral energy distribution (SED) fitting using the dust emission model of Draine & Li (2007; hereafter DL07). The DL07 model includes emissions from polycyclic aromatic hydrocarbon (PAH) molecules and graphite and silicate grains, covering the entire IR wavelength range from the mid-infrared (MIR) through the submillimeter. It is consistent with observations of a variety of infrared continuum and PAH features in local galaxies (e.g., Draine et al. 2007; see also Dale et al. 2012 for more detailed descriptions). In the SED fittings, the Milky way extinction curve (MW3.1) is assumed and the maximum interstellar radiation field intensity is fixed at $\mathrm{Umax}={10}^{6}$. The best-fit values of parameters M_dust, gamma, qPAH (between 0.5 and 4.5 with the step of 0.1), and Umin (among values of 0.10, 0.15, 0.20, 0.30, 0.40, 0.50, 0.70, 0.80, 1.00, 1.20, 1.50, 2.00, 2.50, 3.00, 4.00, 5.00, 7.00, 8.00, 12.0, 15.0, 20.0, and 25.0) are then found via a simple χ² minimization. The errors of M_dust include two terms: (1) error associated with model fitting, estimated using Bayesian analysis (da Cunha et al. 2008); and (2) error caused by observational uncertainties, estimated using a Monte-Carlo method: it equals to the standard deviation of the values of M_dust resulting from fittings of 300 simulated IR-submillimeter SEDs. In each SED, the flux in a given Herschel band is generated randomly, assuming a Gaussian probability distribution with the mean equal to the observed flux and the σ equal to the photometric error. The mean errors of M_dust for spirals in H-KPAIRs are estimated to be: 0.086 ± 0.020 dex (log₁₀(M_⊙)), 0.101 ± 0.044 dex, and 0.213 ± 0.191 dex for those with all 6-band Herschel detections, 5-band Herschel detections, and 4-band Herschel detections, respectively.

We carried out a test to check whether the lack of MIR broadband fluxes in the SED fittings introduces any bias, using ∼25 H-KPAIR spirals observed by Spitzer (Xu et al. 2010). As shown in Figure 2, the M_dust obtained using SED fittings with and without MIR fluxes (IRAC 8 and MIPS 24 μm) agrees very well with each other. The scatter is quite small and much lower than other uncertainties. We also found good agreement between L_IR and L_TIR (estimated from 24, 70, 160 μm luminosities; Dale et al. 2005). Thus, both the M_dust and L_IR estimated using the DL07model SED fittings from FIR-submillimeter bands are reliable even though the MIR data points are missing. On the other hand, in trials that used the two-graybody model (2GB; Dunne & Eales 2001) in the SED fittings of H-KPAIR galaxies, we found that the resulting M_dust and L_IR are both systematically lower than those obtained using the DL07model (Figure 3). This is mainly due to the fact that the 2 GB model does not include the MIR emission at $\lambda \lt 20\;\mu {\rm{m}}$; therefore it underestimates M_dust and ${L}_{{\rm{IR}}}$ (Dale et al. 2012).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

It should be noted that there might be a systematic underestimation (up to 20%) of the L_IR for galaxies with very warm MIR-to-FIR colors (f(25 μm)/f(60 μm) > 0.2). These sources are generally associated with bright active galactic nuclei (AGNs; Surace et al. 1998). As discussed in Section 8.1, very few galaxies in our samples may be associated with bright AGNs. Therefore our results shall not be affected significantly by this possible bias. Indeed, Figure 4 shows the dependence of L_IR estimations using DL07 SED fittings without or with MIR (Spitzer IRAC and MIPS 24 μm) data points on Spitzer MIR-to-FIR colors: f(24 μm)/f(70 μm), for 30 spirals in H-KPAIRs that were also included in Xu et al. (2010). There is only one spiral in pairs (J20471908+0019150) with warm MIR-to-FIR color (f(24 μm)/f(70 μm) > 0.2), and there are no obvious underestimations of L_IR without the use of MIR points.

Zoom In Zoom Out Reset image size

For H-KPAIR and control sample galaxies (see Section 5) with a number of Herschel band of detections less than four, we used an empirical relation between M_dust and ${L}_{250\mu {\rm{m}}}$ (the monochromatic luminosity $\nu {L}_{\nu }$ at 250 $\mu {\rm{m}}$) to estimate the dust mass (or its upper limit) from the 250 μm flux (or the upper limit). The adopted relation is $\mathrm{log}({M}_{{\rm{dust}}}/{M}_{\odot })=-0.13\ +0.83\;\times \;\mathrm{log}({L}_{250\mu {\rm{m}}}/{L}_{\odot })$. The L_IR (or its upper-limit) of these galaxies was estimated from the the 100 μm flux (or the upper-limit) according to another empirical relation: $\mathrm{log}({L}_{{\rm{IR}}}/{L}_{\odot })\ =0.84+0.95\;\times \;\mathrm{log}({L}_{100\mu {\rm{m}}}/{L}_{\odot })$. Both relations were derived using H-KPAIR spirals with all 6-band or 5-band detections. Comparing results obtained from SED fittings to those calculated using these relations, the rms errors of the two empirical relations are 0.27 dex (for M_dust) and 0.11 dex (for L_IR), respectively.

In Tables 2 and 3 we listed the Herschel fluxes, stellar mass M_star, the SFR, and total gas masses M_gas (${M}_{{\rm{gas}}}=100\times {M}_{{\rm{dust}}}$) of paired spiral and elliptical galaxies, respectively. The SFR is derived from L_IR using the formula of Kennicutt (1998), with an additional correction factor of ${10}^{-0.20}$ for the conversion from the Salpeter IMF to the Kroupa IMF (Calzetti 2013). It is worth noting that this formalism misses the contribution from unobscured UV radiation, which is on the order of 20% for KPAIR galaxies (Yuan et al. 2012). It is also contaminated by the dust emission powered by the radiation of old stars (Buat & Xu 1996). However, since the same formalism is applied to both H-KPAIR galaxies and control galaxies, these possible biases shall not affect the results on the star formation enhancement of paired galaxies. M_star is estimated using 2MASS Ks band luminosity: M_star/L_K = 1.32 ${M}_{\odot }$/L_⊙ (Xu et al. 2004; Domingue et al. 2009), with an additional correction factor of ${10}^{-0.39}$ (Xu et al. 2012b) in order to match those given in SDSS value-added catalogs (Kauffmann et al. 2003; with Kroupa IMF).

Chang et al. (2015) recently published stellar masses and SFRs for 1 M galaxies from SDSS+WISE, estimated based on 0.4 to 22 μm SEDs using the MAGPHYS (da Cunha et al. 2008) energy balance SED fitting technique. Chang et al.'s work also includes 85 spirals in pairs and all 132 spirals in controls in our samples. Comparisons of stellar masses and SFRs estimated in our work and in Chang et al. (2015) are shown in Figures 5 and 6, respectively.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

From the comparisons, for most of the spirals with higher stellar masses, we found that using 2MASS Ks band luminosities (our work) and SED model fitting (Chang et al. 2015) will result in very similar M_star values. While this only applies for spirals with the lowest stellar masses, Chang et al. (2015) estimated lower M_star (∼0.3 dex) than ours. For comparison, for SFRs Chang et al.'s (2015) values are systematically lower than ours, especially for spirals in the control sample. This may be caused by: (1) in our work, larger beams of Herschel FIR-to-submillimeter bands resulting in more background contaminations; (2) in Chang et al. (2015), underestimation of dust obscurations for optical-to-MIR bands in MAGPHYS SED fittings; (3) the fact that the sizes of our local paired and control galaxies are generally much larger than the PSFs of SDSS and WISE, therefore, using MODELFLUX for SDSS 5-bands and PSF model flux for WISE 4-bands (although applied a correction using R_e) may still cause some flux loss in Chang et al. (2015), especially for the estimations of SFRs using WISE bands.

The control sample was selected among 2MASS galaxies of redshift z < 0.1 (from SDSS), found in four Level-5 HerMES fields: Bootes HerMES, EGS HerMES, ELAIS N1 HerMES, and Lockman SWIRE. Details of the Herschel PACS and SPIRE observations in these fields can be found in Oliver et al. (2012). The morphologies of these galaxies (S or E) were obtained from visual classifications by Galaxy Zoo (data release v1, Lintott et al. 2011). For those galaxies labeled  "UNCERTAIN" in the Galaxy Zoo catalog, two of us (C. C. and C. K. X) visually re-classified them into S or E categories using SDSS optical images. For those galaxies with controversial classifications between us, we used an automatic classification algorithm described in Xu et al. (2010) (u−r color and R₅₀/R₉₀ ratio) as the third party. Then, close pairs (have companions with projected distance <70 kpc and dM_star <0.4 dex), peculiars (visually selected by C. C. and C. K. X.), and objects at the edge of PACS or SPIRE images (with low coverage) were rejected from the sample. The resulting parent control sample was then 1-to-1 matched with H-KPAIRs. The matched control galaxies have: (1) the same morphology (S or E); (2) similar stellar mass: $\delta {M}_{{\rm{star}}}\lt 0.1$ dex except for the match of J13082737+0422125, which is $\delta {M}_{{\rm{star}}}=0.15;$ (3) with the closest z to that of the corresponding H-KPAIR galaxy.

The Herschel PACS images at 100 and 160 μm (the 70 μm band was not included in the HerMES survey) and SPIRE images at 250, 350, 500 μm were taken from the HerMES data release (v2) for the 176 control sample galaxies. PACS and SPIRE aperture photometries were made using IDL/phot and circular apertures. Fixed circular aperture sizes were used for most galaxies, with aperture radii of 12'', 18'' for PACS 100 and 160 μm, and 24'', 3333, and 48'' for SPIRE 250, 350, and 500 μm, respectively. Larger apertures were used for a few extended galaxies. The background and error estimates are similar to those in the aperture photometry of H-KPAIR galaxies (Section 3). The total dust masses, SFRs and stellar masses of control sample galaxies were calculated using the same method as for the H-KPAIR galaxies (Section 4). In Table 4 we list the galaxy ID (name), R.A. and decl., redshifts, the Herschel fluxes, stellar mass M_star, the SFR, and total gas masses M_gas (${M}_{{\rm{gas}}}=100\times {M}_{{\rm{dust}}}$) of spiral galaxies in the control sample.

It should be noted in the following comparisonal analysis with the control sample, each galaxy in the pairs of our H-KPAIR sample was included independently, rather than the pair as whole, which would introduce a bias.

Figure 7 is a scatterplot for the relation between M_star and the sSFR for spirals in H-KPAIR and in the control sample. The maximum likelihood Kaplan–Meier (K–M) estimator (Kaplan & Meier 1958, see also Xu et al. 1998), which exploits information in the upper limits, was used to calculate the means of the sSFR of different populations. For spirals in H-KPAIR the mean is $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.66\pm 0.06$, and for spirals in the control sample it is nearly identical: $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.65\pm 0.04$. However, detailed inspections of Figure 7 reveal a mismatch between the H-KPAIR and the control sample: while none of the control galaxies have a value or upper limit of $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})$ below −11.3, many H-KPAIR spirals have their $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})$ values/upper limits below this threshold. This is due to the fact that control galaxies are more distant (median z = 0.056) than H-KPAIR galaxies (median z = 0.040), and their Herschel observations in the HerMES Level-5 survey (Oliver et al. 2012) are shallower than our H-KPAIR observations. Therefore their upper limits are in general significantly higher than those of H-KPAIR galaxies, and none of them were detected below $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-11.3$. This mismatch can bias the mean of the control sample toward higher values even though the K–M estimator is used, for the algorithm requires that the detections cover the entire data range and the detections and upper limits are well mixed (Feigelson & Nelson 1985). In order to confirm that this is indeed the case, the following test was carried out: a hybrid sample was constructed by replacing those galaxies in the control sample that matched to the low sSFR H-KPAIR galaxies ($\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})$ values/upper limits below −11.3) with the H-KPAIR galaxies themselves. Again using the K–M estimator, we found a mean $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.94\pm 0.05$ for such a sample, significantly lower than the result for the control sample. The implicit assumption in this test is that below $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-11.3$ the control and H-KPAIR samples have nearly identical sSFR distributions, which is reasonable if the red spirals with low sSFRs are similar to early-type galaxies that show no interaction-induced star formation enhancement (Sulentic 1989).

Zoom In Zoom Out Reset image size

In Figure 7 we plotted the lines representing the blue sequence and the red sequence, taken from Figure 7 of Schiminovich et al. (2007), where it is also shown that galaxies with $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})\lt -11.3$ belong to the red sequence.

Zoom In Zoom Out Reset image size

In order to minimize the bias due to the mismatch, we confined the star formation enhancement analysis to H-KPAIR galaxies with values/upper limits of $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})\geqslant -11.3$ and the control galaxies matched to them. Among these 101 paired SFG galaxies, the sSFR of 5 galaxies are upper limits; for their matches in the control sample, 19 have the sSFR in upper limits. It is possible that the true values of some of the upper limits are below the $\mathrm{log}(\mathrm{sSFR})$ cutoff. However, they are only small fractions of both samples. Furthermore, because in the K–S estimator the probability distributions of the true values of the upper limits are estimated using data points included in the analysis (none of them is below the cutoff), the possible contamination by low sSFR red spirals should not significantly affect the mean sSFR of SFGs in both the H-KPAIR and control samples.

We found a mean of $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.34\pm 0.05$ for SFGs in the H-KPAIR, and $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.67\pm 0.04$ for their matches in the control sample. The corresponding sSFR enhancement (Xu et al. 2010) is $\mathrm{enh}=\langle \mathrm{log}(\mathrm{sSFR}){\rangle }_{\mathrm{pair}}$ $-\langle \mathrm{log}(\mathrm{sSFR}){\rangle }_{\mathrm{cont}}=0.33\pm 0.06$. This indicates that the sSFR of paired SFGs is enhanced on average by a factor of ∼2 compared to control galaxies, and the enhancement is significant at the $\gt 5\sigma$ level. When paired SFGs are divided into subsamples of galaxies in S+S pairs and in S+E pairs, we found a mean of $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.21\pm 0.06$ for the S+S subsample and $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.64\pm 0.07$ for the S+E subsample. This shows that SFGs in S+S pairs have strong sSFR enhancement (with $\mathrm{enh}=0.46\pm 0.08$) and those in S+E have no significant sSFR enhancement. The result is in very good agreement with that of (Xu et al. 2010) based on Spitzer observations of a smaller subsample of KPAIR. We further divided the S+S and S+E subsamples by the interaction morphology (INT/MER vs JUS), and found that the mean values for SFGs in S+S pairs of INT/MER types and of JUS type are $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.14\pm 0.07$ and $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})=-10.31\pm 0.08$, respectively. This suggests that SFGs in S+S pairs of INT and MER types are slightly more enhanced (by 0.17 dex) than those in JUS pairs, though the significance of the difference is only at the 1.5σ level. No significant sSFR enhancement is found for SFGs in S+E pairs of either the INT/MER types (mean $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})\ =-10.58\pm 0.11$) or the JUS type (mean $\mathrm{log}(\mathrm{sSFR}/{\mathrm{yr}}^{-1})\ =-10.71\pm 0.11$).

In Figure 8 we compared the sSFR of SFGs in S+S pairs and S+E pairs with the controls in four ${M}_{{\rm{star}}}$ bins (log(M_star/M_⊙) <10.4, 10.4 $\leqslant$ log(M_star/M_⊙) <10.7, and 10.7 $\leqslant$ log(M_star/M_⊙). For SFGs in S+S pairs we found a similar amount of sSFR enhancement in different M_dust bins. For SFGs in S+E pairs no significant sSFR enhancement is found in any M_dust bin. It is worth noting that Xu et al. (2010) found that SFGs in S+S pairs in their lowest M_star bin (corresponding to $9.3\lt \mathrm{log}({M}_{\mathrm{star}}/{M}_{\odot })\lt 9.8$ for the M_star calibration used in this paper) do not show any significant sSFR enhancement. These low-mass galaxies are not included in H-KPAIR because of the exclusion of pairs of $v\lt 2000\;\mathrm{km}\;{{\rm{s}}}^{-1}$.

According to James et al. (2002) the total gas mass (M_gas) can be estimated from M_dust : ${M}_{{\rm{gas}}}/{M}_{{\rm{dust}}}=1/(Z\times e\times f)\ =114/Z$, where Z is the metallicity in solar units, e = 0.456 is the fraction of metals incorporated in dust in the ISM, and f = 0.019 is the metal mass fraction in gas of solar metallicity. Draine et al. (2007) found a very similar gas-to-dust ratio (M_gas/M_dust ∼100) for SINGS galaxies, including interacting galaxies. Studies on gas-to-dust mass ratios in local galaxies (e.g., Rémy-Ruyer et al. 2014) show that for metal-rich galaxies the ratios are uniformally distributed, and only very metal-poor galaxies have significantly higher M_gas/M_dust ratios. In the local universe, only low-mass galaxies have such low metallicities (Tremonti et al. 2004). All SFGs in H-KPAIR have ${M}_{{\rm{star}}}\gt {10}^{9.8}{M}_{\odot }$, and the control galaxies are 1-to-1 matched to paired galaxies according to M_star. Therefore we adopted a fixed gas-to-dust ratio of 100 for the conversion from M_dust to M_gas. Using the K–M estimator, we found that the means of $\mathrm{log}({M}_{{\rm{gas}}}/{M}_{{\rm{star}}})$ of SFGs in S+S pairs, in S+E pairs, and in the control sample are −0.93 ± 0.05, −1.07 ± 0.07, and −1.00 ± 0.04, respectively. For both SFGs in S+S pairs and in S+E pairs, the mean $\mathrm{log}({M}_{{\rm{gas}}}/{M}_{{\rm{star}}})$ differs from that of control galaxies only at a $\sim 1\sigma$ level. In Figure 9 the means of $\mathrm{log}({M}_{{\rm{gas}}}/{M}_{{\rm{star}}})$ of different samples in different mass bins are compared. No clear trend is found in the plot.

Zoom In Zoom Out Reset image size

The SFE is defined as the ratio between the SFR and total gas mass: $\mathrm{SFE}=\mathrm{SFR}/{M}_{{\rm{gas}}}$. The SFE analysis was confined to SFGs with M_gas detections. The means of $\mathrm{log}(\mathrm{SFE}/{\mathrm{yr}}^{-1})$ of SFGs in S+S pairs, in S+E pairs, and in the control sample are found to be −9.26 ± 0.05, −9.56 ± 0.05, and −9.67 ± 0.05, respectively. Figure 10 shows that the means of $\mathrm{log}(\mathrm{SFE})$ of SFGs in S+S pairs are consistently above that of control galaxies at a ∼0.4 dex level in all mass bins. On the other hand, the means of $\mathrm{log}(\mathrm{SFE})$ of SFGs in S+E pairs are seen both above and below that of control galaxies, consistent with no significant SFE enhancement.

Zoom In Zoom Out Reset image size

In Figure 11 we plotted the mean $\mathrm{log}(\mathrm{SFE})$ against the gas fraction ${f}_{{\rm{gas}}}={M}_{{\rm{gas}}}/({M}_{{\rm{star}}}+{M}_{{\rm{gas}}})$ for the S+S pair and control samples. Within the range of f_gas covered by SFGs in our samples, which is [0, 0.25], there is no significant dependence of the SFE enhancement on f_gas. Taken at the face value, this does not confirm the results of theoretical simulations showing weaker merger-induced star formation enhancement for high f_gas (Hopkins et al. 2009; Perret et al. 2014; Scudder et al. 2015). However, it appears that the effect of f_gas on star formation enhancement is significant only when ${f}_{{\rm{gas}}}\gtrsim 0.3$ (Scudder et al. 2015), which is beyond the range covered by our samples.

Zoom In Zoom Out Reset image size

8.1. AGNs in H-KPAIR

8.2. Why is SFR Enhancement Absent in S+E Pairs?

The significant sSFR enhancements of SFGs in S+S pairs found in Xu et al. (2010) and in this work are due to a significant SFE enhancement within them, while they have the same mean gas content as the control galaxies. On the other hand, the lack of sSFR enhancement in SFGs in S+E pairs cannot be attributed to a deficiency of gas content, as speculated by Xu et al. (2010), but to a significantly lower SFE than that of SFGS in S+S pairs (Table 5). It is difficult to understand the SFE difference between SFGs in S+S pairs and those in S+E pairs, if the enhancement is triggered by tidal effects (the tidal effects caused by a spiral companion and by a elliptical companion should be similar). It has been suggested that cloud-cloud collisions (Scoville et al. 1986; Tan et al. 2006) or cloud-squeezing by shock-heated diffuse gas in two colliding disks (Jog & Solomon 1992; Saitoh et al. 2009; Hayward et al. 2014; Renaud et al. 2015) can trigger starbursts. Our results indicate that these mechanisms, associated with the collision between the ISM in two gas-rich galaxies (which is absent in S+E pairs), may indeed be the major mode for the SFR enhancement in paired SFGs (mostly early-stage mergers). Hwang et al. (2011) also found that in the GOODS-Herschel fields the SFR and sSFR increase as a late-type galaxy approaches a late-type neighbor (S+S pairs), while the SFR and sSFR decrease or do not change much as the galaxy approaches an early-type neighbor (S+E pairs). They argued for X-rays from the ellipticals as the reason for the suppression of star formations in the spirals in S+E pairs.

Table 5.  sSFR and SFE Enhancement

(1)	(2)	(3)	4
Type	sSFR Enhancement	SFE Enhancement	$\delta \mathrm{log}({M}_{{\rm{gas}}}/{M}_{{\rm{star}}})$
	(dex)	(dex)	(dex)
S in S+S	0.46 ± 0.08	0.41 ± 0.07	0.07 ± 0.06
S in S+E	0.03 ± 0.08	0.11 ± 0.07	−0.07 ± 0.08

Note. Mean values of sSFR and SFE enhancements (dex) and of the difference in $\mathrm{log}({M}_{{\rm{gas}}}/{M}_{{\rm{star}}})$, for SFGs in S+S and S+E pairs.

Download table as:  ASCIITypeset image

8.3. Separation and SFR Enhancement