Masers and Star Formation Activities in W51A

Concurrently with the maser flare observed in W51-North during the 3 month monitoring of H2O, NH3, and CH3OH maser variability from 2020 January to April using the Tianma 65 m Radio Telescope, we conducted Very Large Array mappings for these three maser species across the entirety of W51A region. After finding the ring-shaped H2O maser which might trace the disk surrounding the protostar residing in W51-North, the NH3 (9,6) maser delineated a jet which might be illuminated in the luminosity outburst possibly caused by the infalling streamer’s interaction with the protostar’s disk. An analysis of the comprehensive distribution of maser spots allowed us to affirm that W51N4 or ALMAmm31 serves as the primary source of the Lacy jet. Furthermore, we observed that class I methanol maser spots may extend beyond the locations of the H2O and NH3 (9,6) masers within the outflow. Additionally, emissions from other NH3 maser transitions coincided with specific 1.3 mm continuum sources. The arrangement of H2O maser spots in the vicinity of W51e2-E potentially indicates episodic accretions in this source. Combining the data from the Atacama Large Millimeter/submillimeter Array data archive for W51-North, W51e2, and W51e8, we have discovered that the H2O, NH3, and CH3OH masers, as well as the HC3N and SiO emissions are found to be good tools for tracing outflow in this work for W51A.


Introduction
The unclear nature of the formation process of high-mass stars compared to their low-mass counterparts has made it a crucial aspect of modern astrophysics for several decades.However, research on high-mass star formation (HMSF) has been relatively limited compared to low-mass star formation due to several obstacles (Zinnecker & Yorke 2007;Tan et al. 2014;Motte et al. 2018;Liu et al. 2021), including the significantly shorter evolution timescales of high-mass stars, the smaller number of high-mass stars in the Universe, the thick optical depth of dust and gas in their natal locations, the complex environment in which they form in clusters, and the intricate feedback effects during their birth.Nonetheless, astronomers can still study high-mass star formation using radio telescopes since the longer wavelengths can penetrate the interstellar environment.
Based on the pumping mechanism, the CH 3 OH maser is typically classified into two categories, Class I and II (Menten 1991b).Class I methanol masers, such as those emitting at 44 and 95 GHz (Morimoto et al. 1985;Ellingsen 2005;Chen et al. 2011Chen et al. , 2013)), are found to be associated with collisions (Cragg et al. 1992) in shocks and outflows (Sobolev et al. 2005;Voronkov et al. 2012Voronkov et al. , 2014)).On the contrary, Class II methanol masers, such as those emitting at 6.7 and 12.2 GHz (Batrla et al. 1987;Menten 1991a), are pumped in the radiation field (Cragg et al. 1992) close to ultracompact HII regions and massive young stellar objects (MYSOs; Walsh et al. 1998).Recently, flares from Class II methanol transitions have been observed in several high-mass episodic accretion sources, including G358.93-0.03-MM1(Breen et al. 2019;MacLeod et al. 2019;Sugiyama et al. 2019;Burns et al. 2020;Miao et al. 2022;Burns et al. 2023), NGC6334I-MM1 (Hunter et al. 2017(Hunter et al. , 2018;;MacLeod et al. 2018), and S255IR-NIRS3 (Fujisawa et al. 2015;Stecklum et al. 2016;Caratti o Garatti et al. 2017;Moscadelli et al. 2017), indicating variations of the radiation field in these MYSOs.In addition to these two typical maser species, NH 3 masers are also observed in interstellar space, although they are less common and weaker.So far, approximately 20 ammonia maser transitions have been detected in SFRs, such as Sgr B2, DR21, W51-IRS2, and NGC6334I (Henkel et al. 2013;Goddi et al. 2015;Mei et al. 2020;Zhang et al. 2022).Although there are some typical interstellar masers, a thorough understanding of their relationships has not yet been established.However, employing multiple maser tracers is valuable for determining the properties of isolated interstellar objects observed, as well as discerning unique sources that emit one or more maser transitions.Therefore, studying the relationships between masers is imperative.
The W51 giant molecular cloud (GMC) complex serves as a promising research site for investigating the relationship between masers.Within this cloud, three primary regions, namely W51A, W51B, and W51C, are located.However, only the W51A whose distance was determined to be 5.1 kpc (Xu et al. 2009) exhibits active star-forming phenomena, while W51B represents a large HII region and W51C is a supernova remnant (Ginsburg 2017).Numerous studies have detected H 2 O and CH 3 OH masers within W51A (e.g., Menten 1991a;Imai et al. 2002;Etoka et al. 2012).Particularly noteworthy is the observation of the most abundant ammonia maser emissions in W51-IRS2, a region within W51A (Henkel et al. 2013;Zhang et al. 2022).This region consists of two highly obscured YSOs named as W51-North and W51d2 (Sollins et al. 2004;Zapata et al. 2008), and an intriguing bipolar outflow called "Lacy jet" (Lacy et al. 2007;Ginsburg et al. 2016Ginsburg et al. , 2017;;Ginsburg 2017).The W51-North hot core is one peculiar source that hosts SiO maser, which is often detected to accompany AGBs (Hasegawa et al. 1986;Eisner et al. 2002).The W51d2 is a hypercompact HII region with molecular gas surrounding it (Ginsburg 2017).In regards to the Lacy jet, it was initially identified as a precessing blueshifted jet with a high velocity of approximately 110 km s −1 , emerging from the ionized molecular gas near W51-IRS2 into a HII region, as observed in the 12.8 μm [Ne II] and 10.5 μm [S IV] lines by Lacy et al. (2007).In Ginsburg et al. (2017), both blue-and redshifted outflows were observed in the CO J = 2-1 transition originating near the continuum source known as "ALMAmm31," which manifests that the driving source of the Lacy jet was found to be ALMAmm31.Beside the W51-IRS2 in the west of W51A, in the east is another region with active star formation.It was initially resolved into several cores, including W51e2-E, W51e2-W, and W51e2-NW (Shi et al. 2010a).Regarding the renowned starforming core W51e2-E, Shi et al. (2010b) identified a bipolar outflow and proposed that W51e2-E comprises either an O star or a cluster of B stars.Furthermore, Ginsburg & Goddi (2019) were the first to detect maser emissions from the CS J = 1-0 and J = 2-1 transitions in W51e2-E.Regarding another well-known core, W51e2-W, Shi et al. (2010b) initially classified it as an ultracompact HII region.For W51e2-NW, it was classified as a dusty core (Goddi et al. 2016(Goddi et al. , 2020)).
Recently, a flare event of H 2 O maser occurred in W51-IRS2 and was detected using the TianMa 65 m Radio Telescope (TMRT) in the K band from 2020 January to April (Zhang et al. 2022).In addition to this outburst of luminosity in the H 2 O maser emission, multiple NH 3 and CH 3 OH transitions, such as NH 3 (9,6), (7,5), (1,1), (2,2), and CH 3 OH 4 2 -4 1 , also exhibited luminosity outburst in their maser and nonmasing emissions.We hypothesized that a heatwave propagated outward before 2020 March 20, from the outburst origin, causing this maser flare phenomenon.Subsequently, the maser environment in W51-IRS2 was restructured in 2020 April due to the variations of environmental temperatures, as derived from a rotation temperature diagram analysis of the ammonia and methanol nonmasing emissions.To determine the origination of this luminosity outburst, we conducted Very Large Array (VLA) observations and analyzed the spatial distribution of H 2 O maser spots within W51-North in conjunction with the Atacama Large Millimeter/submillimeter Array (ALMA) archived data.It was found that an infalling streamer manifested by the SiO J=5-4 and HC 3 N J = 24-23 transitions unsteadily fed mass onto the disk indicated by the ring-shaped distribution of the H 2 O maser spots.Therefore, one possible origin of this flare is a collision between the infalling streamer and the disk surrounding the protostar.On or even before 2020 January, a ∼0.25 M e clump within the infalling streamer collided with the outer disk around W51-North, releasing its gravitational potential into the energy of the heatwave that was then propagated outward spherically from the collision region (Zhang et al. 2023).To check the scenario about the origin of this maser flare and the association between the SiO J = 5-4 and HC 3 N J = 24-23 emissions, we further analyzed the distributions of all the three maser species (H 2 O, NH 3 , and CH 3 OH) obtained from our VLA observation and the SiO J = 5-4 and HC 3 N J = 24-23 emissions from the ALMA archived data toward the entire region of W51A, including W51-IRS2 and W51e2/e8 regions in this work.
The rest of this paper is organized as follows.A detailed description of the observation and data reduction processes is introduced in Section 2. Subsequently, the results are presented in Section 3. Following that, an analysis of the results is provided in Section 4. Finally, the conclusion of this work is given in Section 5.

VLA Observation and Reduction
In Zhang et al. (2023), the details of our VLA observation and data reduction were introduced.Here, we give a short summary about it and append a bit information about the data reduction for this article: the K-band (18-26.5 GHz) VLA observation toward W51-IRS2 (J2000 position: R.A. =19 h 23 m 40 047, decl.=  ¢  14 31 05.530) was carried out on 2020 March 31, with the C-array configuration and the field of view expanding to include the W51e2/e8 region.For calibration, the gain/phase calibrator J1924+1540 and the bandpass/flux density calibrator 3C48 were observed.The configuration of the narrow-band spectral windows in the data processing terminal was designed to cover the 22.235 GHz water maser transition with a velocity resolution of ∼0.11 km s −1 (i.e., 7.812 kHz per channel), and the 15 NH 3 (1,1) transition, 23 ammonia transitions, and three methanol transitions with a velocity resolution of ∼0.06 km s −1 (i.e., 3.906 kHz per channel; see Section 3 for details).In addition, some wide-band windows were configured to record the continuum data.
In the data reduction, the flagging and calibration were made with the standard VLA calibration pipeline in the Common Astronomy Software Applications (CASA) package, and the clean process was conducted by using the MIRIAD package.Regarding the ammonia emission lines with hyperfine structure, we used the MOMENT task in MIRIAD to analyze the integrated flux density (moment 0) map for their main lines, of which the velocity ranges are typically 50-70 km s −1 .Regarding all the maser emission lines, we used the SAD command in the Astronomical Image Processing System (AIPS) to carry out two-dimensional Gaussian fitting to maser spots, channel by channel.Some useful information for the reduction is that the cell size of the image is ∼0 3, the detection limit (at 3σ RMS ) in a single channel is ∼0.01 Jy beam −1 , the positional uncertainty of the calibrator is less than 0 02, and positional uncertainties of the maser spots are smaller than 0 01.

ALMA Data Reduction
To help understand the structure of W51 and the results of the VLA observation, the SiO J = 5-4, HC 3 N J = 24-23 transition, and the 1.3 mm continuum data in the public ALMA data set from the Cycle 3 program 2015.1.01596.S were analyzed.As shown in previous involved works (Goddi et al. 2020;Tang et al. 2022;Zhang et al. 2023), this observation was carried out from 2015 October 27 to 31, with a high angular resolution being ∼0 02 at Band 6 (216-237 GHz), and two fields centering on W51-North (J2000 position: R.A. = 19 h 23 m 40 05, decl.=  ¢  14 31 05.50) and W51e2/e8 (J2000 position: R.A. = 19 h 23 m 43 91, decl.=  ¢  14 30 34. 50).The synthesized beam size is ∼0 031 × 0 025.For the SiO J = 5-4 and HC 3 N J = 24-23 lines, the spectral resolution is ∼1.13 MHz per channel, corresponding to a velocity resolution of ∼1.56 km s −1 , and the noise level σ RMS is ∼0.64 mJy beam −1 .For the continuum, the noise level is ∼0.02 mJy beam −1 .Checking the image quality of the relevant pipelinereduced image fits from the archive to be good, we directly analyzed them with the CASA package.

Results
In our VLA observation, maser or nonmasing emissions from the H 2 O, NH 3 , or CH 3 OH molecular species were detected mainly in the W51-North, W51d2, W51e2, and W51e8 subregions. 4 The maser detection results of different subregions are summarized in Table 1.Additionally, parameters of maser spots corresponding to the detected H 2 O, NH 3 (9,6), CH 3 OH 2 2 -2 1 /4 2 -4 1 and the rest of ammonia maser transitions (e.g., NH 3 (6,3), (7,5), (7,6), (7,7), and (8,6)) are listed in Tables 2, 3, 4 and 5, respectively.To provide a concise overview of the detection results, we presented the maser distribution in W51A by depicting the H 2 O, NH 3 (9,6), and CH 3 OH 4 2 -4 1 maser spots with different colors in the right panel of Figure 1.Furthermore, we included a figure from Ginsburg et al. (2017) on the left side as illustrative material for comparison.The W51-North and W51d2 subregions are the two main emission areas detected in W51-IRS2.To clearly present the maser detection results for each one, we have provided separate analysis for W51-North and W51d2 subregions.

Masers and Nonmasing Emissions in W51-IRS2 Region
For the W51-North subregion, there are six maser transitions that were clearly detected in total: 22.235 GHz H 2 O, 24.933 GHz CH 3 OH 4 2 -4 1 , 18.499 GHz NH 3 (9,6), and three other ammonia (i.e., NH 3 (7,7), (7,6), and (8,6)) maser emissions (see Tables 1-5 for details).The spectra and spatial distributions of maser spots are presented in Figure 2. Seen from the spectra in this figure, the H 2 O, CH 3 OH 4 2 -4 1 and NH 3 (9,6) transitions are pure masers with their spectral peaks at ∼60 km s −1 , while the ammonia 25.715 GHz (7,7),  3) are the maser species, transitions, and their corresponding rest frequency.Columns (4)-( 7) present the maser detection results.The marks "Y" and "N" represent that the corresponding maser transitions were detected and not detected, respectively.6) record the flux density at the peak and the total flux density of each two-dimensional Gaussian fits for the maser spots, respectively.All values in the brackets present the uncertainties of their corresponding parameters which are listed in columns (2), (3), (5), and (6).
(This table is available in its entirety in machine-readable form.)22.924 GHz (7,6),and 20.719 GHz (8,6) transitions exhibit a mixture of maser components with spectral peaks in the range of 40-55 km s −1 and nonmasing components with spectral peaks at ∼60 km s −1 .All the pure maser transitions feature at least two peaks in their spectra.Moreover, the flux densities at the peaks in the spectra of pure maser transitions are significantly more luminous than those in transitions combining maser and nonmasing components.
From the distribution of maser spots in Figure 2, it is observed that the H 2 O and NH 3 (9,6) maser transitions display a cluster of maser spots at a northwest location with respect to W51-North, along with a jet-like structure in the northwest-southeast direction.The H 2 O maser distribution alone reveals additional structures, including a ring-like structure and another cluster located at a southeast location.In addition, the spots of CH 3 OH 4 2 -4 1 maser transition occur in two congregate areas that seem to be outer regions of the jet-like structure, while the NH 3 (7,7), (7,6), and (8,6) maser transitions exhibit only one cluster of maser spots located at the western part of the central area of the W51-North.
In the case of the W51d2 subregion, four maser transitions have been detected: H 2 O, NH 3 (9,6), NH 3 (7,5) at 20.804 GHz, and NH 3 (6,3) at 19.757 GHz.The spectra and maser spots for these transitions are presented in Figure 3. Upon examining the spectra in this figure, it is evident that the H 2 O, NH 3 (9,6), and (7,5) transitions are pure maser emissions, indicated by their narrow line widths and high flux densities.The spectral peak for the H 2 O  6) and (7) record the flux density at the peak and the total flux density of each two-dimensional Gaussian fits for the maser spots, respectively.All values in the brackets present the uncertainties of their corresponding parameters which are listed in columns (3), ( 4), ( 6), and (7).
(This table is available in its entirety in machine-readable form.) Columns (1) and (2) are source names and ammonia maser transitions, respectively.The meaning of columns (3)-( 7) is the same as that in Table 4, but for ammonia maser transitions, excluding the NH 3 (9,6) one which is recorded in Table 3.
(This table is available in its entirety in machine-readable form.)maser transition is located at ∼60 km s −1 , while that of the ammonia maser transitions are found to be located at ∼55 km s −1 .
Regarding the distribution of maser spots in the W51d2 subregion, the H 2 O maser transition shows an elongated distribution along the east-west direction.Conversely, the NH 3 (9,6) maser spots are distributed at three scatter locations.In contrast, all the spots for the NH 3 (7,5) and (6,3) maser transitions are clustered around a single location, which is also the southest cluster of the NH 3 (9,6) maser spots (see Figure 3).
The emission area of each transition in W51-IRS2 is indicated with contours in blue in Figure 4, whose levels are 0.2 and 0.5 times the peak of the integrated flux density of these transitions.It can be seen that the contour level of 0.2 times the peak of the integrated flux density covers the emission area of each transition well, thus we regard the coverage of the contour level of 0.2 times the peak of the integrated flux density as the emission area of the 23 nonmasing transitions.In this way, the areas of emission for metastable ammonia transitions5 (2,2), (3,3), and (4,4) transitions have larger emission areas compared to those of the (5,5), (6,6), and (7,7) transitions. 6Regarding nonmetastable ammonia transitions, emission regions of the NH 3 (4,3), (3,2), and (2,1) transitions are larger than those of the (8,7), (7,6), and (6,5) transitions.Moreover, the latter group surpasses the emission areas of other detected ammonia transitions in W51-North, e.g., NH 3 (4,1), (5,3), and (6,1).Furthermore, emissions from the (2,2), (3,3), (5,3), (7,3), and (8,5) transitions are detected in both W51-North and W51d2 subregions, while the (9,7) transition is only detected in W51d2 subregion.Postulating that the ammonia emission areas depend on their E u /κ values (refer to Table 6), we derived several rules as follows: 1. 0 < E u /κ < 50 K: widely spread and their largest emissions have an offset from W51-North; 2. 50 < E u /κ < 250 K: widely spread and their largest emissions are located at the position of W51-North; 3. E u /κ > 250 K: the emission areas around W51-North decrease generally with their E u /κ increasing; some transitions, especially those that have E u /κ > 850 K, are only emitted in the W51d2 subregion.
3.1.3.SiO J = 5-4 and HC 3 N J=24-23 Emissions in W51-IRS2 In the analysis of the relevant ALMA data, we focus on two molecular transitions: the SiO J = 5-4 and HC 3 N J = 24-23.Figure 5 illustrates the blue-and redshifted emissions of these two transitions in blue and red, respectively, while the 1.3 mm continuum is shown in green.Examining this figure, it is found that while both the SiO J = 5-4 and HC 3 N J = 24-23 emissions trace the infalling streamer in W51-North (see Zhang et al. 2023 for details), the HC 3 N J = 24-23 transition emission poorly indicates the presence of the bipolar outflow in W51-North, appearing much fainter and weaker compared to the SiO J = 5-4 transition emission.Apart from W51-North, their emissions in two additional continuum sources also exhibit great differences.The SiO J = 5-4 emission accurately traces the point source to the east of W51-North, marked by white dashed circles in Figure 5, whereas no HC 3 N J=24-23 transition emission was detected in that region.For the source W51N4 situated to the west of W51-North, the SiO J = 5-4 emission appears significantly fainter compared to the HC 3 N J = 24-23 transition.Furthermore, the emission region of the HC 3 N J = 24-23 transition expands considerably in comparison to that of the SiO J = 5-4 transition.

Masers in W51e2 and W51e8 Subregions
The eastern region of W51A exhibits fewer detected maser emissions in comparison with the detection results in W51-IRS2.This region can be divided into two subregions centered on W51e2 and W51e8, respectively.The detailed findings are presented below.
Maser emissions of H 2 O, NH 3 (9,6), and CH 3 OH 2 2 -2 1 were detected in W51e2.The spectra and distributions of maser spots are presented in the upper panels of Figure 6.It can be seen that CH 3 OH 2 2 -2 1 and H 2 O transitions exhibit two and multiple velocity components in the spectra, respectively.Although the NH 3 (9,6) transition has the spectrum showing one emitting velocity component with its line width broader than the other two maser transitions, it is indeed a maser emission because the spectrum for NH 3 (9,6) transition detected toward W51e2 in Figure 1 of Pratap et al. (1991) had a flux density about 33 Jy, corresponding to an extremely high brightness temperature of 1.2 × 10 13 K for a Gaussian source model.Here, it is important to note that the absorption feature at 54 km s −1 in the NH 3 (9,6) spectrum originates from W51e2-W, an ultracompact HII region (Shi et al. 2010a(Shi et al. , 2010b)).The most prominent peak of the CH 3 OH 2 2 -2 1 maser transition appears at ∼64.5 km s −1 , while the other two maser species peak at ∼58 km s −1 .Regarding the distribution of maser spots, the NH 3 (9,6) and CH 3 OH 2 2 -2 1 transitions show a dense cluster of maser spots near the northwestern part of W51e2-E.However, the distribution of Figure 2. The spectra and spots of all the maser transitions detected toward the W51-North subregion in our VLA observation.For transitions of pure masers, the spectra are shown in colorful lines.The spectra of transitions mixed with maser and nonmasing components are shown in black, and their maser components are marked with gray bars.For maps of maser spots, the gray contours show the 1.3 mm continuum, with the start, step, and end levels to be 0.006, 0.012, and 0.023 Jy beam −1 .
H 2 O maser spots is complex, with two distinct maser clusters (W51e2-NW and W51e2-SE) and several potential symmetric bipolar outflows whose bases center approximately at W51e2-E being presented.
The H 2 O and NH 3 (9,6) maser emissions were also detected in the W51e8 subregion.The lower panels of Figure 6 show the spectra and maser spots for these two transitions.This figure illustrates the presence of at least two peaks in the spectra.The H 2 O maser transition exhibits its most luminous peak at ∼60 km s −1 , whereas this velocity component does not correspond to the most luminous peak in the NH 3 (9,6) maser transition spectrum (∼55.5 km s −1 ).Regarding the distribution of maser spots, the distribution of H 2 O maser spots is comparatively larger and more dispersed than that of the NH 3 (9,6) maser spots.Additionally, three small subclusters of H 2 O maser aligned in the east-west orientation exist, of which the subclusters at both ends with respect to the middle one may provide insight into the outflows or shocks originating from the protostar within the W51e8 subregion (see Section 4.3.3 for details).
3.2.2.SiO J = 5-4 and HC 3 N J = 24-23 Emissions in W51e2 and W51e8 Subregions In the upper panels of Figure 7, the emissions from the SiO J = 5-4, HC 3 N J = 24-23, and 1.3 mm continuum in W51e2 are shown.It can be observed in the figure that the SiO J = 5-4 and HC 3 N J = 24-23 transitions exhibit emissions solely in W51e2-E.They trace the same orientation of the bipolar outflow (Goddi et al. 2020).However, the emission region of the SiO J = 5-4 transition is narrower and brighter compared to that of the HC 3 N J = 24-23 transition, which exhibits a bicone morphology.
W51e8 was also found to exhibit the SiO J = 5-4 and HC 3 N J = 24-23 emissions.The moment 0 maps of these two molecular transitions are presented in the lower panels of Figure 7.The figure clearly shows that the emission from the SiO J = 5-4 transition effectively traces the outflow at redshifted velocities.However, the emission from the HC 3 N J = 24-23 transition depicts a symmetric bipolar outflow or shocks in the east-west orientation, resembling that observed in the SiO J = 5-4 emission.

The Lacy Jet and Its Driving Source in W51-IRS2
Both the blue-and redshifted outflows of the so-called "Lacy jet" were observed in the CO 2-1 transition originating near the continuum source known as "ALMAmm31" (Ginsburg et al. 2017), which is also referred to as "W51N4" in this article.Nevertheless, further observations of the Lacy jet are necessary to better understand its properties and explore the interaction between this bipolar outflow and the molecular cloud.As no maser observations have been conducted toward this structure so far, the question arises as to whether it induces maser emissions.
Figure 8   length of the Lacy jet measures ∼0.4-0.5 pc.The blueshifted outflow, as indicated by H77α, is seen in the projection on the sky to intersect with the redshifted CO or SO outflow emanating from W51-North (Ginsburg et al. 2017).The terminal section of the blueshifted outflow in the H77α RRL emission from the Lacy jet aligns with the position of W51d1.Since the ionized W51d1 is at a high-velocity range from −60 to −16 km s −1 (see Figure 9 in Ginsburg et al. 2016), it is a part of the Lacy jet, which means that the W51d1 region is neither a part of the molecular cloud, nor a separate HII region.Regarding the W51d region, its location corresponds to the termination point of the redshifted CO outflow originating from W51-North.Since it does not appear in the H77α RRL emission, it can be considered a region that interacts with or is projected onto the CO redshifted outflow from W51-North.These characteristics of the interstellar medium imply a complex environment surrounding W51-North and the Lacy jet.
W51N4, the driving source of the Lacy jet, resides at the central base of the CO red-and blueshifted outflows.It is detectable in H 2 O maser emissions, as illustrated in the middle panel of Figure 8.The distribution of maser spots situated in proximity to W51N4 exhibits both red-and blueshifted velocities, which trace a bipolar outflow within a brief segment of the Lacy jet.On the west side of W51N4, however, the blueshifted velocity of H 2 O maser spots corresponds to the redshifted CO emission.In addition, the H 2 O maser has some spots on the east side of W51N4 that are with redshifted velocity, of which their locations have CO emission with blueshifted velocity.There are three competitive possibilities to interpret this phenomenon.The first one is that these CO emission and H 2 O maser spots come from outflows that have different orientations.The second one is that they trace different parts of the outflow, and have different axis.The third one is that the H 2 O maser traces the recent change in the orientation of the outflow, while the CO emission traces the old outflow.Moreover, the distribution of additional H 2 O maser spots in close proximity to W51N4 reveals another nearly symmetric bipolar outflow (denoted as "mrk3" in Figure 8) with a distinct orientation compared to that of the Lacy jet.This suggests the presence of multiple outflows within this source.

The Maser Variability from 2020 January to April
We recently had applied a VLA observation of W51-IRS2 to resolve the spatial distributions of the H 2 O, CH 3 OH and NH 3 masers (Zhang et al. 2023), following the detection of a luminous outburst in the target using the TMRT (Zhang et al. 2022).In this paper, combining more masers detected in our VLA data may provide additional evidence to support our proposed scenario that a collision between a clump in the infalling streamer and the disk can be one cause of maser flares.
The spectra and distribution of H 2 O maser spots obtained using TMRT and VLA toward W51-IRS2 are displayed in panels (b), (c), and (d) in Figure 9. Detailed analysis of the variability from the TMRT monitoring is present in Appendix A. It can be seen from the H 2 O maser spectra detected by VLA in panel (b) that most of the velocity components belong to W51-North.However, the ∼58 km s −1 one primarily corresponds to the W51d2 subregion.Seen from the TMRT detected H 2 O maser spectra in panel (d) of Figure 9, the ∼58 km s −1 velocity component had a flux density >1000 Jy, which is much larger than that of the other three epoch's observations.Thus, the W51d2 subregion was influenced by the luminosity outburst before 2020 March 20.Furthermore, the "C" region in panel (a) of Figure 9 (see also in the lowerright panel of Figure 5 in Zhang et al. 2023) which aligns with the blueshifted outflow depicted by H 2 O maser spots within W51-North, also contains maser spots with the ∼58 km s −1 velocity component.Therefore, the heatwave must have propagated outward from an origin between the blueshifted outflow (the "C" region) in W51-North and the W51d2 subregion.In Zhang et al. (2023), this origin lies within the "A" region marked in panel (a) in Figure 9, characterized by the ∼59.0 and ∼63.0 km s −1 velocity components that show the largest variation.In other words, the analysis of Figure 9 provides support to the hypothesis in Zhang et al. (2023) that the luminosity outburst originated from the location of the redshifted outflow in H 2 O maser spots, where a potential collision between a clump within the infalling streamer and a disk could have occurred.Additionally, the newly appeared jet, oriented from the northwestern redshifted outflow to the southeastern blueshifted outflow, and the ring-shaped distribution of H 2 O maser spots shown in the panel (a) of Figure 9, could be illuminated in this luminosity outburst, thus provided a support for our hypothesis.
The spectra and distribution of maser spots for the NH 3 (9,6) maser transition are illustrated in Figure 10.The analysis of the left spectra obtained in the TMRT monitoring is presented in Appendix A. Seen in the upper-right panel, the most luminous velocity component at ∼54 km s −1 and a portion of flux density at ∼63 km s −1 originate from the W51d2 subregion, while all the other velocity components are from W51-North.In the lower-right panel of Figure 10, we observed a jet-like distribution of the NH 3 (9,6) maser spots within W51-North,  4) represent the upper-level energy, the product of the total torsion-rotational line strength and the square of the electric dipole moment, and the corresponding rest frequency for each transition depending on the data from the JPL catalog database.
which is similar to that observed in H 2 O maser spots.This suggests that the distribution of NH 3 (9,6) maser spots also traced this luminosity outburst, and the NH 3 (9,6) maser emission is a good indicator of the jet resulting from the collision between the infalling streamer and the disk (Zhang et al. 2023).Assuming the heatwave spherically propagates outward at a speed of c in diffuse interstellar space, it would take ∼3 months to propagate from the collision location to the position of the W51d2 subregion.In other words, the NH 3 (9,6) maser emission manifests that this maser flare happened even before 2020 January 8.In addition to our observations, another monitoring from 2020 January 1 to 2023 June 7 by using the 22 m radio telescope located in Katsively was carried out (Volvach et al. 2023).In that work, unusually powerful flare phenomenon of H 2 O maser toward W51-North7 was reported.In the future, maser monitoring lasting for long by using multiple telescopes and interferometers are required, so as to provide information, such as the due time of maser flares and the variations in the spectra, for understanding the dynamic star formation activities.

Distributions of Different Masers in W51A
As mentioned in Section 1, CH 3 OH, H 2 O and NH 3 maser transitions are detected in various SFRs, albeit in distinct sources.Notably, W51A is found to exhibit all three maser species, rendering it an ideal candidate for investigating maser interrelationships.Surprisingly, there has been limited research dedicated to exploring maser interrelationships, prompting us to conduct a thorough investigation of W51A.

Distributions of Masers in W51-North
Among the sources we observed with the VLA, W51-North exhibits the most abundant maser transitions, including the NH 3 (7,7) transition as discussed in Section 3.1.1.Figure 11 presents a scatter plot of maser spots color coded to represent their velocities, allowing for an investigation of maser relationships associated with this source.The left panel of this figure displays the distribution of maser spots corresponding to the most common maser transitions.Notably, our VLA observation reveals a ring-shaped distribution and a jet-like structure oriented from the northwest to the southeast of H 2 O maser spots surrounding W51-North.This observation suggests the disk and jet illuminated in a recent luminosity outburst in W51-North (Zhang et al. 2023).Furthermore, the distribution of H 2 O maser spots traces two maser clusters in the northwest and southeast directions relative to W51-North, which were already identified in the study by Imai et al. (2002).For the distribution of NH 3 (9,6) maser spots, a jet-like structure with a similar orientation to that of the H 2 O maser is observed, which supports NH 3 (9,6) maser as a tracer for collisions and confirms the presence of the jet illuminated by the luminosity outburst discussed in Zhang et al. (2022Zhang et al. ( , 2023) ) and Section 4.2.Intriguingly, the distribution of NH 3 (9,6) maser spots only traces the northwestern maser cluster, which may serve as substantial evidence supporting our hypothesis of collision(s) between the infalling streamer and the accretion disk within the "A" region of W51-North, and an outward heatwave propagation from this region (see the left panel of Figure 9 for details).In addition to the H 2 O and NH 3 (9,6) maser emissions, our observations also detected one typical class I methanol transition, CH 3 OH 4 2 -4 1 , at ∼25 GHz in proximity to W51-North.However, the distribution of maser spots for this methanol transition differs from that of H 2 O and NH 3 (9,6) masers.The southeastern CH 3 OH 4 2 -4 1 maser spots are situated further outward from the H 2 O maser emission jet, whereas the northwestern ones are positioned further away and slightly offset from the northwestern maser cluster associated with the H 2 O and NH 3 (9,6) maser emissions.This suggests that the class I methanol maser emissions can trace a more distant segment of jets or outflows, beyond the area accessible to H 2 O and NH 3 (9,6) masers within SFRs.
The distributions of additional maser transitions within W51-North, specifically NH 3 (7,7), (7,6), and (8,6), are illustrated in the right panel of Figure 11.It is worth noting that all three of these ammonia maser spots are concentrated along the eastern edge of the disk within W51-North.This concentration implies that the NH 3 (7,7), (7,6), and (8,6) maser emissions could Figure 6.Upper panels: the spectra and distribution of maser spots for H 2 O, NH 3 (9,6), and CH 3 OH 2 2 -2 1 transitions detected in the W51e2 region.The meaning of the colorful lines and gray bars in the spectra is similar to that in Figure 2.For maps of maser spots, the gray contours show the 1.3 mm continuum, with the start, step, and end levels to be 0.002, 0.002, and 0.018 Jy beam −1 .Lower panels: the spectra and distribution of maser spots for H 2 O and NH 3 (9,6) transitions detected in the W51e8 region.The gray contours in maps of maser spots show the 1.3 mm continuum, with the start, step, and end levels to be 0.002, 0.001, and 0.018 Jy beam −1 .The black dashed box in the lower-left panel marks the H 2 O maser spots belonging to W51e8.originate from a singular object or region situated on the disk.Interestingly, this location corresponds to the area labeled as "mrk1" in Figure 8, which contains an arc-shaped structure observed in the 1.3 mm continuum.Furthermore, this location significantly differs from the distribution of H 2 O, NH 3 (9,6), and CH 3 OH 4 2 -4 1 maser spots within W51-North.Consequently, it is plausible that the NH 3 (7,7), (7,6), and (8,6) maser emissions originate from a continuum source or a shocked region that projects onto the disk of W51-North.

Distributions of Masers in W51d2 Subregion
As shown in Section 3.1.1,the W51d2 subregion was also detected in H 2 O, NH 3 (9,6), and some other ammonia maser emissions.The spatial distributions of maser spots corresponding to these transitions are depicted in Figure 12.It can be seen from this figure that H 2 O and NH 3 transitions constitute the majority of the maser emissions in this subregion.Regarding the distribution of H 2 O maser spots, it is evident that they encompass nearly the entire W51d2 subregion.These maser spots are distributed from the western to the eastern extremities of this area and along the Lacy jet (see Section 4.1 for specifics).Furthermore, an arc-shaped distribution of H 2 O maser spots below the Lacy jet was identified, with its location marked as "mrk2" in Figure 8.This location coincides with a widely extending 1.3 mm continuum source.In the case of the distribution of NH 3 (9,6) maser emissions, it is noteworthy that the majority of NH 3 (9,6) maser spots are concentrated at three positions, with some scattering toward the eastern end of the overall region.Additionally, all these three positions coincide with the locations of continuum objects, which implies that these NH 3 (9,6) maser transitions are likely to be excited in   these continuum objects.In contrast to the widespread distribution of H 2 O maser spots, there are no NH 3 (9,6) maser spots observed at the location of W51N4 or within the arcshaped structure traced by H 2 O maser spots.Conversely, at the location of the southest NH 3 (9,6) maser spot cluster, no H 2 O maser spots are detected.
In addition to the NH 3 (9,6) maser emission, the spatial distributions of maser spots for the two other ammonia emissions exhibiting maser features in the W51d2 subregion, NH 3 (7,5) and (6,3), are also depicted in Figure 12.It is evident from this panel that neither the NH 3 (7,5) nor (6,3) transition exhibits maser spots in the region where H 2 O maser spots are found in the W51d2 subregion.Interestingly, both of these two ammonia emissions exhibit maser spots located at the position of the southest NH 3 (9,6) maser spot cluster.

Distributions of Masers in W51e2/e8 Region
W51-East, a large region different from W51-IRS2, consists of many continuum sources and HII regions.Within this region, two renowned star-forming subregions, W51e2 and W51e8, exhibit detections in H 2 O, NH 3 , and CH 3 OH maser transitions, as detailed in Section 3.2.1.In contrast to the detection outcomes in the W51-North and W51d2 subregions, only a few maser transitions have been observed in W51e2/e8.Nonetheless, it is still possible to conduct a comprehensive analysis  of the spatial distributions of maser spots across these three maser species in the W51e2/e8 Region.
For W51e2, we concentrate on discussing the spatial relationships among H 2 O, NH 3 , and CH 3 OH maser spots surrounding W51e2-E, which is the primary source that excites maser emissions.As observed in the left panel of Figure 13, there are indications of multiple sources, possibly two with distinct systemic velocities (Ginsburg & Goddi 2019), at the location where the base of the red-and blueshifted outflows is situated.Notably, none of these sources have NH 3 and CH 3 OH maser spots detected within them.Nevertheless, maser spots from two maser species, NH 3 (9,6) and CH 3 OH 2 2 -2 1 , are detected at the outermost extent of the blueshifted outflow, as indicated by the distribution of H 2 O maser spots.Additionally, they are not detected along the redshifted outflow.It is more likely to be a region associated with the outflow rather than a continuum source, since no 1.3 mm continuum counterparts were found to be associated with NH 3 (9,6) and CH 3 OH 2 2 -2 1 emissions in the comparison for pictures of W51e2-E in Figures 6, 13 and 14.
In the case of another source, W51e8, the irregular distribution of H 2 O maser spots, as observed in Figure 6, appears more complex than that in W51-North, W51d2, and W51e2 subregions, implying a convoluted maser environment within W51e8.Notably, the region depicted in Figure 6 and the right panel of Figure 13 encompasses not just W51e8, but also W51e4.To differentiate between the two, we have outlined a dashed box in Figure 13 to delineate the region containing W51e8.Within this region, two clusters of H 2 O maser spots, positioned along an east-west orientation, exhibit symmetrical distributions in relation to the central maser cluster within W51e8.This pattern indicates the detection of outflows and a possible disk associated with W51e8 in the H 2 O maser transition.Regarding the somewhat fragmented bipolar outflow, its overall orientation, as indicated by the distribution of H 2 O maser spots, aligns with that of the SiO and HC 3 N outflows shown in Figure 7.In addition to the H 2 O maser transition, NH 3 (9,6) maser spots have been detected in W51e8, primarily concentrated at two locations.Both of the locations align with one end of the two perpendicular outflows observed  in the H 2 O maser transition, yet no NH 3 (9,6) maser emissions are detected at the other ends (see Figure 13).

A Comprehensive Understanding of Relationships among Masers
The H 2 O, NH 3 , and CH 3 OH molecules have maser transitions that can be used to trace outflows.For H 2 O, its maser spots evidently form or present in the NW-SE oriented jet in W51-North and W51e2-E, the Lacy jet in the region containing W51d2, and the presence of multiple outflows in W51e8.For NH 3 , its (9,6) maser has spots that form the jet in the NW-SE orientation within W51-North, as depicted in Figures 2 and 11.In the case of W51e2 and W51e8, the NH 3 (9,6) transition has maser spots that are situated at the outer section of one branch of the bipolar outflows, as revealed by the distribution of H 2 O maser spots.For CH 3 OH, the 4 2 -4 1 maser spots in W51-North are scattered along the outer reaches of the bipolar outflow, as indicated by the distribution of H 2 O and NH 3 (9,6) maser spots.The CH 3 OH 2 2 -2 1 transition in W51e2 has maser spots congregate at the outer extremity of the outflow(s) identified by H 2 O maser spots, where NH 3 (9,6) maser spots are also observed.Therefore, the class I CH 3 OH maser spots are primarily concentrated in the outer regions of outflows.Compared with the CH 3 OH and H 2 O maser distribution in the sequential evolution hypothesis of the hot cores in Figure 10 of Miyawaki et al. (2021), it can be derived that W51-North and W51e2-E are both in the hollow hot core stage.
Beside tracing outflows, the H 2 O maser may also trace accretion disks.Theoretical studies have shown that H 2 O maser can also be influenced by radiation (Goldreich & Kwan 1974;Gray et al. 2022), which is a property somewhat similar to the pumping mechanism of the 6.7 GHz methanol maser transition.In our VLA detection, the presence of a ring-shaped structure in H 2 O maser spots in W51-North provides compelling evidence supporting the notion of radiation-induced pumping for H 2 O maser spots.
In summary, H 2 O and NH 3 (9,6) maser spots can serve as excellent tracers of jets during a luminosity outburst.Furthermore, H 2 O maser can also serve as a tracer for accretion disks.In the absence of luminosity outbursts, H 2 O, NH 3 (9,6), CH 3 OH 4 2 -4 1 , and 2 2 -2 1 maser spots are still valuable tracers of outflows.As demonstrated in Sections 3.1.3and 3.2.2,emissions from both the SiO J = 5-4 and HC 3 N J = 24-23 transitions were observed in W51-IRS2 and W51e2/e8 regions in the ALMA archival data.A brief analysis of different subregions can be summarized as follows: 1.The continuum source at the eastern end of the W51-IRS2 region that has detected with SiO J = 5-4 and without HC 3 N J = 24-23 emissions (see Section 3.1.3and Figure 5 for detail) is unique and needs to be studied in detail in the future.2. In W51-North, the HC 3 N J = 24-23 emission outlines the edge of the infalling streamer (Zhang et al. 2023), whereas SiO J = 5-4 emission fills its inner portion.Moreover, the HC 3 N J = 24-23 emission does not apparently trace the bipolar outflow traced by the SiO J = 5-4 emission (see Figure 5).3. W51N4, the driving source for the Lacy jet, has an HC 3 N J = 24-23 emission likely to trace an outflow or infalling streamer that extends nearly to the westernmost continuum source in the W51-IRS2 region.However, the SiO J = 5-4 emission extends over a much smaller area, which could have the most shocks in the flow (see Figure 5).4. In W51e2-E, the red-and blueshifted emission of the SiO J = 5-4 transition is concentrated in a shorter section of the bipolar outflow, indicating that it traces the inner content or environment of the outflows.Conversely, the HC 3 N J = 24-23 transition emission primarily outlines the profile of the bipolar outflow edges (refer to the upper panels in Figure 7). 5.The W51e2-W is not detected to have SiO J = 5-4 and HC 3 N J = 24-23 emissions, which suggests that SiO and HC 3 N molecules cannot exist in this ultracompact HII region (refer to Figure 7).Additionally, this source is also detected with no HCN transitions and significant accretions (Shi et al. 2010a).6.In W51e8, the redshifted SiO outflow is prominently visible, while its blueshifted counterpart is faintly detected and concentrated at the base of the bipolar outflow.Because a substantial continuum object is situated near the SiO blueshifted outflow, we suspect that this continuum object obstructs the blueshifted lobe of the SiO outflow.However, the HC 3 N J = 24-23 emission delineates a symmetrical bipolar outflow or shock along the east-west direction (refer to the lower panels of Figure 7).
Regardless of how the SiO J = 5-4 and HC 3 N J = 24-23 transitions manifest themselves in different sources, the clear correlation observed between SiO J = 5-4 and HC 3 N J = 24-23 emissions in outflows in our ALMA maps suggests that HC 3 N may also be excited in shocked environments.This finding aligns with the results reported in  Eisner et al. (2002), Imai et al. (2002), and Goddi et al. (2020), respectively.Nevertheless, to date, no studies have explored the distinctions among these three types of emissions regarding their capacity to trace shocks and outflows.Consequently, we examined their disparities in terms of their ability to delineate outflow orientations in W51-North, W51e2, and W51e8.
In the left panel of Figure 14, we observed an offset between the orientation of the bipolar outflow in W51-North, as indicated by the distribution of H 2 O maser spots, and that indicated by the SiO J = 5-4 emission.Furthermore, the orientation of the H 2 O maser jet illuminated by this luminosity outburst also exhibited an offset from that of the SiO bipolar outflow (refer to Zhang et al. (2023), Figure 1 for details).Indeed, the offset between the outflows indicated by the distribution of H 2 O maser spots and SiO υ = 2, J = 1 → 0 maser emissions, can be also derived from Figure 2 of Imai et al. (2002) or Figures 3 and 4 of Eisner et al. (2002).Besides, the distinction between H 2 O maser and SiO emissions in tracing the outflow was observed in W51e2-E.Examining the middle panel of Figure 14, we observe that the bipolar outflow in W51e2-E, as revealed by the H 2 O maser, exhibits an offset of several degrees compared to that depicted by the emissions from the SiO J = 5-4 transition.Because SiO maser emissions were not detected in W51e2, we cannot ascertain whether the outflow orientations derived from the SiO nonmasing transitions align with those from SiO masers in W51e2-E.
In the right panel of Figure 14, however, H 2 O maser spots trace not only the SiO redshifted outflow but also its blueshifted counterpart in W51e8, following the same orientation.The extent of the outflow from W51e8, as indicated by the distribution of H 2 O maser spots, closely matches the scale of the outflow revealed by the HC 3 N emission, as depicted in the lower-right panel of Figure 7. Furthermore, H 2 O maser spots induced by W51e4, a source located approximately ∼1 north of W51e8, were also detected.Additionally, it appears that the possible southeast-northwest oriented outflow from W51e4 merged with the east-west oriented outflow of W51e8, enhancing the H 2 O maser emission located at the position of the eastern SiO redshifted outflow from W51e8.In Shi et al. (2010a), four cores named W51e2-E, W51e2-W, W51e2-N, and W51e2-NW are resolved within the SFR W51e2.Among these cores, W51e2-E is currently experiencing active star formation.
As depicted in the upper-left panel of Figure 15, H 2 O maser emissions reveal a bipolar outflow within W51e2-E, extending in the southeast-northwest direction.For the H 2 O maser spots in the southeastern lobe of the outflow, they have redshifted velocities, while the CO J = 3-2 emission shows blueshifted velocities (Shi et al. 2010b).Correspondingly, the H 2 O maser spots and CO J = 3-2 emission (Shi et al. 2010b) in the northwestern lobe of the outflow have blue-and redshifted velocities, respectively.This contrast between H 2 O maser and CO emission is much clearer than that in the Lacy jet as mentioned in Section 4.1, thus the three possibilities came up with to interpret the different shifted velocities between these two emissions for the Lacy jet work for the outflows in W51e2-E.Furthermore, subfigures in the upper panels of Figure 6 reveal the presence of two maser clusters symmetrically positioned relative to W51e2-E.Their blue-and redshifted velocities are consistent with those observed in CO J = 3-2 emission (Shi et al. 2010b) and SiO J = 5-4 emission (Ginsburg & Goddi 2019).In comparison to the positions of W51e2-E, W51e2-W, W51e2-N, and W51e2-NW marked in Figure 1 of Shi et al. (2010b), it is evident that the northwestern H 2 O maser cluster corresponds to W51e2-NW, while the southeastern cluster represents a novel source that is already referred to as "W51e2-SE" in Figure 6.
In order to explain the symmetry of the W51e2-SE and W51e2-NW maser clusters with respect to W51e2-E, and the inverse velocity shifts between H 2 O maser and CO molecular transitions, we initially hypothesized that the accretion and outflow activity in this source occurs episodically.This led us to consider the W51e2-SE and W51e2-NW maser clusters as "knots" or remnants within the ejected material resulting from a prior accretion episode.From this perspective, W51e2-NW is regarded as a "knot" associated with a past accretion event.However, Shi et al. (2010a) identified this source in the SMA 0.85-mm continuum data.Then, Goddi et al. (2016) detected weak ammonia emissions toward this continuum source in the VLA data, and Ginsburg et al. (2017) detected more molecular emissions toward it in the ALMA data.Therefore, the W51e2-NW is a hot core associated with the jet "knot" from W51e2-E.Consequently, the whole system of W51e2 depicted in Figures 6 and 7 is likely composed of outflows, knots, and continuum sources, as also evidenced by their detection in the HCN J = 4-3 transition presented in Figure 1 of Shi et al. (2010b).
As for the hourglass or bicone morphology of HC 3 N emission in tracing shocks in W51e2-E (see the lower-left panel of Figure 15), we found it can be interpreted with the turbulent entrainment scenario for protostellar outflows (Li et al. 2013; see the lower middle panel of Figure 15 for a quick overview), which was tested in explaining the hourglass shape of the CO outflows in G240.31+0.07(Qiu et al. 2009;Li et al. 2013).Indeed, this morphology was also found in the CO outflows in G201.52-11.08, G203.21-11.20, G208.68-19.20, G209.55-19.68, G210.49-19.79, G210.82-19.47, G211.16-19.33, G212.84-19.45 (Dutta et al. 2024).When compared with the schematic scenario presented in the lower-right panel of Figure 15 or Figure 12 of López-Vázquez et al. (2024), which is for multiple shells within the outflow driven by episodic disk winds blew by the protostar HH 30 located in the dark cloud L1551 (Zucker et al. 2019), we found the SiO and HC 3 N emissions traced multiple shells in the outflow, while the "knots" within the distribution of H 2 O maser spots traced a previous episodic accretion event which might previously happen in W51e2-E, though it is a high-mass star-forming core (Shi et al. 2010b).
Investigating the core of W51e2-E is also of utmost importance.Genzel et al. (1981) identified four H 2 O maser clusters within W51e2-E, comprising three high-velocity clusters located in the northern, middle, and southern regions.
The remaining cluster was labeled as "double knots".Subsequently, Leppänen et al. (1998) identified a rotating protostar cocoon within the "double knots".They determined its inner and outer radii to be 5 and 66 au, assuming a distance of 7 kpc to W51e2-E (Genzel et al. 1981).If we consider the more recent trigonometric parallax distance of 5.4 kpc (Sato et al. 2010), the radii of these two layers would be 3.9 and 50.9 au.Afterward, Imai et al. (2002) determined that these four maser clusters had remained stable over a period of two decades by analyzing their observational data and the historical data from Genzel et al. (1981) and Leppänen et al. (1998).However, we cannot recognize these four maser clusters in the upper-right panel of Figure 15 based on the relative distances among them defined in Figure 6 of Imai et al. (2002), thus these maser clusters perhaps have undergone significant changes within the centroid of W51e2-E since 2002.This discrepancy complicates the task of accurately aligning them and studying their motions over the past 20 yr.Nevertheless, the detection of maser emissions from the CS J = 1-0 transition at ∼64 km s −1 and J = 2-1 transition at ∼34.5 km s −1 in two locations (distance: ∼0 035) near the centroid of W51e2-E by Ginsburg & Goddi (2019) suggests that the environment of at least two H 2 O maser clusters in W51e2-E have the potential to excite CS masers.Recently, CS masers have been found to be collision excited (van der Walt et al. 2021), of which the pumping mechanism is the same as the H 2 O maser (de Jong 1973).

Conclusion and Summary
Along with the detection of H 2 O and NH 3 (9,6) maser flares in W51-North by using TMRT, the following-up VLA observation was carried out to find the origin of this luminosity outburst.Some other ammonia and methanol maser or nonmasing emissions were also detected from the VLA observation.This work aims to find the relationships among the distributions for masers of different kinds of molecular transitions.In the final, fruitful results and findings were found as the followings: 1. Maser emissions from the 22.235 GHz H 2 O transition, NH 3 (9,6), and other five ammonia transitions, NH 3 (6,3), (7,5), (7,6), (7,7), and (8,6), and CH 3 OH 4 2 -4 1 and 2 2 -2 1 transitions were detected mainly in W51-IRS2, W51e2 and W51e8 from our VLA observations.In addition, 23 nonmasing emission from NH 3 (22 in total), 15 NH 3 (only one), and CH 3 OH (two in total) transitions were also detected.2. The NH 3 (9,6) maser is excited in continuum sources and outflows, while other ammonia masers excite only in continuum sources.In addition, the CH 3 OH maser spots are mainly distributed at the outer region of outflows.3. The emission of the HC 3 N J = 24-23 transition outlines only the profile about the edges of the bipolar outflow or infalling streamer, while the SiO J = 5-4 emission fills their inner portions.4. The interaction between the infalling streamer and the disk of W51-North as the origin of the luminosity outburst was confirmed by the distribution of NH 3 (9,6) maser spots in a larger spatial scale toward the W51-IRS2 region.5. W51N4 as the driving source of the Lacy jet was confirmed through the distribution of H 2 O maser spots in W51d2 subregion.6. Complicated outflow(s) with "knots" traced by the H 2 O maser were detected in the W51e2-E region, suggesting a previously potential episodic accretion event in this region.7. The orientation of the outflow indicated by H 2 O maser spots sometimes is offset from that indicated by the SiO maser or nonmasing emissions.
As discussed in the content, more astrochemical works, highresolution observations or surveys by using single dish and interferometers are required in the future to check the universality of the relationships among the maser emission from H 2 O, NH 3 , CH 3 OH transitions, and nonmasing emission from SiO and HC 3 N transitions.Moreover, further observation can be helpful for studying Lacy jet and W51e2-E region.

Figure 1 .
Figure 1.Left panel: a picture of the W51A region extracted from Ginsburg et al. (2017).Its background is made of millimeter emission of lines and continuum from ALMA observations: HC 3 N 24-23 in purple, C 18 O 2-1 in blue, CH 3 OH 4 2 -3 1 in orange, 1.3 mm continuum in green.Right panel: the dots in blue, red, and green represent the overall distributions of the H 2 O, NH 3 (9,6), and CH 3 OH 4 2 -4 1 masers in W51A, overlaid on the gray K-band continuum background detected in our VLA observation.

Figure 3 .
Figure3.The spectra and spots of all the maser transitions detected in the W51d2 subregion.The maser emissions are shown with colorful lines in the spectra of the H 2 O and NH 3 (9,6) transitions, and marked with gray bars in the spectra of the NH 3 (7,5) and (6,3) transitions.The gray contours in maps of maser spots show the 1.3 mm continuum, of which the start, step, and end levels are the same as that in Figure2.
presents a comprehensive depiction of the Lacy jet within W51-IRS2.The upper-left subfigure, sourced from Ginsburg et al. (2017), illustrates the Lacy jet through H77α radio recombination line (RRL) emission with blue contours, overlaid on the CO 2-1 emission depicted in red and blue.The driving source, W51N4, is delineated by white contours in the ALMA 1.3 mm continuum.Observing this subfigure, the entire

Figure 4 .
Figure 4.The integrated intensity (moment 0) maps of all the nonmasing transitions detected in W51-IRS2.Contour levels in red for the integrated flux density of different transitions are 0.05, 0.1, 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 Jy beam −1 km s −1 , while those in blue are 0.2 and 0.5 times the peak of the integrated flux density for each transition.

Figure 5 .
Figure 5. Upper panel: the SiO J = 5-4 emission toward W51-IRS2 is shown in red and blue, which correspond to the velocity ranges of 67-97 and −17 to 56 km s −1 , while the 1.3 mm continuum emission is presented in green.Lower panel: the HC 3 N J = 24-23 emission toward W51-IRS2 shown in red and blue corresponds to the velocity ranges of 66-90 and 39-52 km s −1 .The white dashed circles mark a compact source to the east of W51-North (see Sections 3.1.3and 4.4.1).

Figure 7 .
Figure 7. Upper-left panel: the emission of the SiO J = 5-4 transition from the W51e2 region shown in red and blue corresponding to velocity ranges of 78-159 km s −1 and −54-54 km s −1 , respectively.Upper-right panel: the emission of the HC 3 N J = 24-23 transition from the W51e2 region shown in red and blue corresponding to velocity ranges of 70-80 km s −1 and 28-49 km s −1 , respectively.Lower-left panel: the emission of the SiO J = 5-4 transition from the W51e8 region shown in red and blue corresponding to velocity ranges of 70-82 km s −1 and 4-56 km s −1 , respectively.Lower-right panel: the emission of the HC 3 N J = 24-23 transition from the W51e8 region shown in red and blue correspond to velocity ranges of 74-84 km s −1 and 29-52 km s −1 , respectively.In all these four subfigures, the 1.3 mm continuum is shown in green.

Figure 8 .
Figure 8. Upper-left panel: a figure showing the W51-IRS2 regions taken from Ginsburg et al. (2017).The emission shown in green is the NACO K-band continuum (Barbosa et al. 2008; Figuerêdo et al. 2008), and these in blue and red are the blue-and redshifted CO 2-1 emission (Ginsburg et al. 2017).The blue contours are the H77α RRL transition detected in Ginsburg et al. (2016), and the white contours are the ALMA 1.3 mm continuum (Ginsburg et al. 2017).Upper-right panel: a zoomed version of the region toward a red dashed box in the upper-left panel.The background is the ALMA 1.3 mm continuum from the project 2015.1.01596.S.The contour levels in white for our VLA K-band continuum are 0.05, 0.10, 0.15, 0.20, 0.40, 0.60, and 0.80 Jy beam −1 .The red solid ellipse denotes the W51d2 subregion.Lower panel: a zoomed version of the region toward a red dashed box in the upper-right panel.The background is the ALMA 1.3 mm continuum, with gray contours marking the most luminous objects in this region.The "+" symbols mark the location of the H 2 O maser spots.The labels "mrk[number]" are temporary names given by us for studying these objects or subregions that do not even have a name before.Among them, "mrk1", "mrk2" and "mrk3" are studied in Sections 4.3.1,4.3.2 and 4.1, respectively.

Figure 9 .
Figure 9. Panel (a): a replotted figure of the right panel of Figure 1 in Zhang et al. (2023) to show the maser distribution in W51-North.The light-colored circles and dark-colored squares represent the H 2 O maser spots detected in our VLA observation and that in Imai et al. (2002), respectively.The reference position is R.A. (J2000) = 19 h 23 m 40 047, decl.(J2000) = +14°31¢05 530.The 1.3 mm continuum detected in the archival ALMA data is shown with dark blue contours, with the start, end, and step levels to be 0.0025, 0.0100, and 0.0025 Jy beam −1 , respectively.The "A," "B," and "C" labels mark the regions that have shown variations, as shown in Figure 5 in Zhang et al. (2023).Panel (b): the spectra of H 2 O maser transition extracted from W51-IRS2, W51-North, and W51d2 regions detected with the VLA.The red dashed lines at ∼58 km s −1 in panels (b) and (d) mark the strongest velocity component in the W51d2 subregion.Panel (c): the spatial distribution of H 2 O maser spots in the whole region of W51-IRS2.Panel (d): the spectra of H 2 O maser transition detected toward W51-IRS2 using TMRT from 2020 January 8 to April 7, which was taken from Zhang et al. (2022).

Figure 10 .
Figure 10.Left panel: the spectra of NH 3 (9,6) maser transition detected toward W51-IRS2 using TMRT from 2020 January 8 to April 7, which was taken from Zhang et al. (2022).Upper-right panel: the spectra of NH 3 (9,6) maser transition extracted from W51-IRS2, W51-North, and W51d2 regions detected with the VLA.The red dashed lines at ∼54.2 km s −1 in the left and upper-right panels mark the strongest velocity component in the W51d2 subregion.Lower-right panel: the spatial distribution of H 2 O maser spots in the whole region of W51-IRS2.

Figure 11 .
Figure 11.Left panel: distributions of maser spots of the typical kinds of maser transitions in W51-North, including H 2 O, NH 3 (9,6), and CH 3 OH 4 2 -4 1 ones.Right panel: distributions of maser spots for the NH 3 (7,7), (7,6), and (8,6) transitions in W51-North, overlaid on that of H 2 O maser for comparison.The white contours in both panels show the 1.3 mm continuum emission detected from the ALMA archived data.

Figure 14 .
Figure 14.Left panel: a figure taken from Zhang et al. (2023), of which white circles and orange squares represent the H 2 O maser spots detected in W51-North from our VLA observation and from Imai et al. (2002).The meaning of the colors in the background is the same as that in the upper panel of Figure 5. Middle panel: white circles represent the H 2 O maser spots detected in W51e2-E from our VLA observation, and the meaning of colors in the background is the same as that in Figure 7. Right panel: white circles filling in gray represent the H 2 O maser spots detected in W51e8 from our VLA observation, and their size indicates the flux intensity of the H2O maser spots.Since the flux intensities in W51e8 are much lower than that of other sources, they are amplified about 100 times for clear comparison.The meaning of the colors in the background is the same as that in Figure7.Notably, the source "W51e4" is beyond the range of this subfigure, although we pointed out the relative direction of the location of this source to the "W51e8" subregion.

Figure 15 .
Figure 15.Upper-left panel: same as the detection result of the distribution of H 2 O maser spots in W51e2 in Figure 6, with names of different regions being marked.Upper-right panel: a zoomed map of the distribution of H 2 O maser spots, corresponding to the region marked with a gray dashed box in the upper-left panel.Lowerleft panel: the 1.3 mm continuum, SiO J = 5-4, and HC 3 N J = 24-23 emissions in the W51e2 region.For orange contours, the start, step, and end levels are 0.06, 0.01, and 0.10 Jy beam −1 , respectively.For cyan contours, the start, step, and end levels are 0.10, 0.10, and 0.50 Jy beam −1 , respectively.Lower middle panel: a cartoon taken from Li et al. (2013), which is used to indicate the turbulent entrainment origin of protostellar outflows.In this figure, the p wind and p envelope represent the ram pressure of the wind and the envelope.Lower-right panel: a cartoon taken from López-Vázquez et al. (2024), which is used to indicate the multilayer outflows driven by disk winds of source HH 30.

Table 2
Details of the Detected H 2 O Maser Spots

Table 5
Parameters of the Rest Detected NH 3 Maser Spots

Table 6
The Physical Parameters of All the Detected NH 3 Nonmasing Transitions Note.Column (1) shows all the observed NH 3 transitions.Columns (2)−( Wang et al. (2022)dWang et al. (2022).4.4.2.Outflow Orientations Indicated by H 2 O Maser and SiO J = 5-4 Emissions H 2 O, SiO masers, and SiO nonmasing emissions have been employed in previous studies to investigate outflows, as demonstrated by