© P. Le Dû et al. 2022

1 Introduction

Planetary nebulae (PNe) represent a brief but important 21000 ± 5000-yr phase (Jacob et al. 2013) in the late-stage stellar evolution of low- to intermediate-mass stars in the range ~1–8 M_⊙. They are at the most luminous evolutionary stage of low-mass stars where ~15% of the luminosity can be in a single emission line. This makes PNe visible out to large distances. Strong emission lines in the red optical region such as Hα and [NII] facilitate discovery in more obscured regions of the Galaxy as their identification is less affected by extinction than other tracers. The large number of PN emission lines permit the determination of abundances, the measurement of radial and expansion velocities, and the estimation of the exciting central star (CSPN) temperatures via photoionisation modelling. Their complex shapes provide clues to their formation, evolution, mass-loss processes, and the shaping role played by binary central stars (De Marco 2009; Miszalski 2009; Jones & Boffin 2017; Boffin & Jones 2019). These characteristics make PNe excellent objects with which to study late-stage stellar evolution and Galactic structure.

Over the last 20 yr the number of Galactic PNe discovered has more than doubled, with more than 2400 PNe added to the current total Galactic inventory of ~3800 since 2006, as recorded in the gold standard Hong-Kong/AAO/Strasbourg/Ha Planetary Nebulae (HASH) database (see e.g. Parker et al. 2016)¹. This significant increase was largely the result of discoveries made from narrow-band Hα wide-field imaging surveys of the northern and southern Galactic plane (Parker et al. 2005 and Drew et al. 2005). Careful examination of these narrow-band optical survey images (e.g. Parker et al. 2006; Miszalski et al. 2008; Sabin et al. 2011) has led to this remarkable increase.

Although professional astronomers and telescopes have provided most of these new Galactic PNe candidates and their spectroscopic and multi-wavelength confirmation (largely following the precepts laid down by Frew & Parker 2010), there has also been a rapidly emerging and increasingly important contribution from an extremely active group of amateur astronomers (e.g. Kronberger et al. 2012; Acker & Le Dû 2014; Le Dû 2018a). The production and dissemination online of publicly available wide-field digital surveys, particularly in the optical and infrared regime has facilitated large-scale coordinated amateur PN hunting programmes (see below). This effort has uncovered 209 spectroscopically confirmed True (122), Likely (51), and Possible (36) Galactic PNe, which represent ~5% of the PNe currently in HASH. A further 610 PNe candidates await follow-up. So far some 104 candidates have been shown to be a variety of PN mimics giving a current PN confirmation success rate of ~67%. All the discoveries and associated information have been compiled into an interactive online database by the amateur group². These discoveries demonstrate the power and value the amateur community can have in undertaking a coordinated, combined imaging and spectroscopic observing programme, as described in Sect. 2 below.

Fig. 1 PNe discovered from high-resolution images of the DECaPS DR1 colour survey by the amateur group. Orientation of all panels has north-east to the top left. Left: PN Pre 44, image size 3.34 × 3.27 arcmin; Centre: PN Pre 59 (blue CSPN visible), image size 1.34 × 1.32 arcmin. Right: candidate Pre 63 (blue CSPN also just visible at PN centre), image size 1.12 × 1.10 arcmin.

2 Amateur groups: Genesis and evolution

Planetary nebula candidates were uncovered by amateur team members from the careful and laborious scrutiny of wide-field survey imagery. This was initially done in the broad and narrow optical bands and in the mid-infrared (MIR) via imagery from the WISE space telescope by the Deep Sky Hunters (DSH) team led by Matthias Kronberger (e.g. Kronberger et al. 2006, 2012) and the French team led by Pascal Le Dû (e.g. Acker & Le Dû 2014). This paper focuses on the work carried out by the French group.

The French-led amateur group, which has grown to 80 members, has long been active in PNe hunting. The group is divided into three categories. There are ~50 discoverers, including some from the DSH group no longer independently active; the astrophotographers make deep narrow-band images of the discoveries; finally, there are about 15 members who take spectra of the PNe candidates. Some members belong to two or even all three categories. The current major initiative began in 2011 with discovery of LDû 1, the first of our amateur discoveries to result in a professional publication (Acker et al. 2012). During this time the processes to confirm and register a new PN candidate were set up. Amateur discoveries likely to be PNe were classified in two tables, the first with objects likely to be PNe based on indicative morphologies, often including a blue central star, and the second with objects of unknown nature. They were published for the first time in l’Astronomie (Acker & Le Dû 2014). A dedicated spectroscopic observing programme to determine the true nature of these PNe candidates was initiated with the help of professional astronomers. This was the beginning of a new era in professional-amateur collaborations. Amateurs were able to discover PN candidates and to determine their true natures. They then appeared in professional catalogues such as VizieR, but also in the PNe HASH database (Parker et al. 2016).

3 PNe discovery techniques

Planetary nebula candidates were uncovered via careful and laborious scrutiny of available online wide-field survey imagery.

3.1 Inspection of recent high-resolution surveys

We made use of high-resolution images of the DECaPS (Schlafly et al. 2018) and PanSTARRS (Chambers et al. 2016) surveys³. On these surveys’ composite multi-band images, we searched for candidate PNe. Any resolved candidates should have blue to green colours that stand out from the surrounding environment when contrast-enhanced using standard image processing techniques. Some objects reveal perfect PN morphology, for example Pre 44, Pre 59, and Pre 63 (see Fig. 1).

3.2 PNe candidates discovered from deep-sky imaging of other targets

Some of the first candidates were discovered serendipitously during narrow-band imaging of known large nebulae. The long exposure times revealed not just faint nebulosities and haloes around the target sources, but also discoveries of unknown objects hidden in the vicinity of such primary targets given the relatively large fields of view available (Acker et al. 2012). LDû 1 was discovered while imaging Sh2-124, and RaMul 2 while imaging the WR128 Ring nebula. Although interesting, this avenue for new discoveries is limited so a more dedicated approach was adopted.

3.3 Systematic PN candidate searches based on professional data or amateur surveys

The vast majority of our amateur PNe discoveries are the result of systematic searches of professional survey images available online (~89%). Search areas are not limited to the Galactic plane. The initial search can be performed from visual inspection of optical or non-optical imagery, but once an object is detected, it is systematically inspected at all available wavelengths. The Aladin tool⁴ from the Centre de Données astronomiques de Strasbourg (CDS)⁵ allows visualisation of an object on several image planes of different wavebands.

3.3.1 Research from optical imagery

To research our candidates we use images from the following surveys: DSS2 (Digitized Sky Survey - STScI/NASA) red, blue, colour⁶ (e.g. POSS-II images); SDSS9 (SLOAN Digitized Sky Survey) colour⁷; IPHAS – Isaac Newton telescope Photometric Hα survey (Drew et al. 2005) DR2 Hα⁸; SHS – SuperCOSMOS Hα survey (Parker et al. 2005)⁹; DECaPS (Schlafly et al. 2018) DR1 colour¹⁰; PanSTARRS (Chambers et al. 2016) DR1¹¹.

We are also researching objects on very wide-field images made by amateurs available online, such as the MDW Survey (David Mittelman, Dennis di Cicco, and Sean Walker)¹². We methodically scan these images, manually or automatically (see below), examining them carefully to spot objects with PN-like morphology or atypical colours. The selected objects must be nebulous and/or extended.

We use the selection criteria described in Jacoby et al. (2010): the object does not appear in the POSS-II IR image; the object is not already catalogued; a cross-check is performed of the candidate with other available imagery to guard against plate defects; a search for faint blue stars is made near the object’s centre as these could be CSPN; the Hα morphology is evaluated to see if it is PN-like.

3.3.2 Hα image inspection

Images from the two narrow-band Hα surveys of the southern and northern Galactic plane, the SHS (Parker et al. 2005) and IPHAS (Drew et al. 2005), respectively, allow the detection of isolated emission nebulosity via comparison to the accompanying broad-band ‘r’ red exposures. The DSH consortium and Laurence Sabin (e.g. Sabin 2008; Sabin et al. 2012) have already thoroughly inspected the images of the IPHAS survey while the third author, Quentin Parker, and his team did the same with the SHS survey, reporting their PN discoveries in Parker et al. (2006) and Miszalski et al. (2008). Inevitably, some very faint candidates were missed in these surveys and subsequently found by our group, given that these surveys are available online. To optimise this pursuit some of our team developed algorithms that automatically detect objects and object contours from the surface brightness (relative to the pixel) associated with the source catalogues. These are adjusted according to the survey used (see Sect. 3.4 below) and allow the detection of a large number of high-quality candidates, for example StDr 13 and StDr 20 (see Fig. 2 centre and right). To better determine the morphology of the Hα candidates, the sky background is removed by subtracting the associated broad-band ‘r’ image or, if necessary, from the DSS. In this way the object’s emission signal is highlighted.

3.3.3 The value of additional amateur imagery

To consolidate identification of our most interesting candidates, we often obtain images with our own telescopes for very deep multi-exposure [OIII] and Hα+[NII] narrow-band filter imagery. We have close ties with well-qualified amateur astrophotographers who can provide bespoke imagery. We even rent remote telescopes to produce very deep combined filter images (see e.g. Fig. 2) that are often superior to any previously available image of the object, including those from professional observatories.

3.4 Inspection at non-optical wavelengths

Emission from each object detected at optical wavelengths is also inspected in non-optical pass-bands such as the ultraviolet (UV), MIR, and radio regimes when available. The main surveys used are: GALEX (the Galaxy Evolution Explorer)¹³ in the ultraviolet; WISE (the Wide-field Infrared Survey Explorer Wright et al. 2010) in the MIR (W1, W2, W3, W4 and colour¹⁴); Spitzer, a higher resolution MIR space facility (Spitzer Space Telescope) colour¹⁵; NVSS (the NRAO VLA Sky Survey Condon et al. 1998)¹⁶ and MGPS2 in the radio (the second epoch Molonglo Galactic Plane Survey, Murphy et al. 2007)¹⁷.

In these images we check whether the object is detected and/or resolved and if it has a PN-type colour or multi-wavelength characteristics (e.g. Parker et al. 2012 for the MIR) or is atypical for its general environment. The WISE four MIR bands are systematically checked to find PNe masked by self-emitted or interstellar dust. Spitzer survey images allow us to detect extremely reddened PNe and have superior resolution allowing some additional morphological detail to be seen.

We also use specific MIR band selection criteria to find new candidates in the Infrared Astronomical Satellite (IRAS) or WISE point source catalogues (e.g. see Acker 2016¹⁸). Most of these MIR candidates do not show a decent optical counterpart and could be PNe suffering high extinction or just mimics. We only include in our main lists candidates confirmed as True, Likely, or Possible PNe. They are recognizable from the ‘IR’ tag in their names.

Fig. 2 Very deep combined narrow-band interference filter images ([OIII] and Hα+[NII]) of PNe by our group. Orientation of all figures has north-east to top left. Left: PN Fe 6 (IPHASX J015624.9+652830). The 7.75 × 7.67 arcmin image was taken in southwest Spain with twin 0.15 m refractors (f/7.9), twin cooled QSI6120 CCD cameras, and Astrodon Hα, [OIII], and LRGB filters. Exposure time: [OIII] 23 × 30 min binning 2 × 2; Hα 27 × 30 min binning 2 × 2; L 21 × 5 min binning 1 × 1; R 24 × 5 min binning 1 × 1; G 29 × 5 min binning 1 × 1; B 25 × 5 min binning 1 × 1. Exposure time total: 41.4 h (Peter Goodhew). Centre: PN StDr 13. The 41.37 × 40.92 arcmin image was taken in Germany at Dachsternwarte with a 0.28 m Schmidt-Cassegrain telescope with a Starizona Hyperstar 4 (f/1.9), a cooled ZWO ASI 1600MM CMOS camera, Baader Hα, [OIII], and Astronomik LRGB filters. Exposure time: [OIII] 40 × 10 min; Hα 40 × 10 min; LRGB 100 × 2 min per filter. Total exposure time: 27 h (Andreas Bringmann). Right: likely PN StDr 20. The 13.86 × 13.7 arcmin image was taken at Chilescope observatory with a 1 m Ritchey-Chretien telescope (f/8), a cooled FLI Proline 16803 CCD camera, forkmount, Hα, [OIII] filters. Exposure time: [OIII] 32 × 20 min; Hα 32 × 20 min; RGB 18 × 10 min per filter. Exposure time total: 24.4 h (Xavier Strottner, Marcel Drechsler).

3.5 Object reclassification and discrimination

Occasionally a known object gets re-classified as a result of our investigations. For example, object SH2-123 is classified as an HII region in Simbad, but its morphology resembles a bipolar PN in the DSS2 red image. We have also found a potential CSPN in the PanSTARSS DR1 colour image. Our spectrum, the first available for this object, taken in January 2020, shows strong [NII] lines relative to Hα that confirms the PN nature of the object; SH2-123 is now classified as a true PN in HASH.

Other than single epoch survey imagery (e.g. from the DSS, SHS, IPHAS, DEcaPS), multi-epoch photometry can also help discriminate between PNe and PN mimics such as symbiotic stars, proto-planetary nebulae (PPNe), nebulae associated with cataclysmic variables (CVs), or even young stellar objects (YSOs).

3.6 The search for CSPN

To help verify whether a PN candidate is a bona fide PN we systematically search for a central star (CSPN) in the PanSTARRS and DECaPS surveys or in the GAIA2 photometric catalogue. GAIA often provides additional information on these possible CSPN, such as distances or temperatures. A blue star in the centre of a weak nebulosity is potential evidence of a PN nature. Once our candidates have been incorporated into HASH all available imagery can be examined for traces of a CSPN including, where available, the GALEX UV imagery.

We also consult the American Association of Variable Star Observations (AAVSO) database¹⁹ and available white dwarf catalogues (e.g. McCook & Sion 1999; Gentile Fusillo et al. 2021). Variable stars can mimic a compact PN candidate by showing an excess in red photometry that can be mistakenly attributed to a Hα signal. We also use the VizieR²⁰ white dwarf candidates catalogues by filtering these objects by temperature. We look for stars above 30 000 K that are covered by the Virginia Tech Spectral-Line Survey (VTSS) Hα survey²¹.

4 Our discoveries

Our discoveries are shown in Tables 1 to 3 available at the CDS. These tables present the compiled results of our current work. At the end of 2021 our discoveries were divided into these three tables with a total of 923 objects. Table 1 contains True, Likely, and Possible PNe; Table 2 contains objects still awaiting classification decisions; and Table 3 contains objects clearly identified as non-PNe.

4.1 The contents of Tables 1 to 3

For each object the following information is given:

• the name given to the object at the time of its discovery;

• the PNG name based on Galactic coordinates (IAU Commission 5 recommendation);

• J2000 Coordinates in RA (HH:MM:SS.ss) and DEC (DD:MM:SS.ss);

• the dimension or apparent dimension in arc minutes, as estimated from the best available optical imagery;

• the publications where the object is designated (non-exhaustive list);

• if available, the origin of the spectra obtained for the object (non-exhaustive list) including the SAAO 1.9 m and Grantecan 10 m telescopes provided by our professional colleagues;

• if available, the origins of the images made with Hα and/or [OIII] filters;

• the HASH unique object identification number in the HASH database;

• the HASH status abbreviation of T (True), L (Likely), or P (Possible) PN. The publication references are as follows:

• A&A09 = Viironen et al. (2009);

• A14 = Acker & Le Dû (2014);

• A15 = Acker & Le Dû (2015);

• A16 = Acker (2016);

• A17 = Le Dû (2017);

• A18-19 = Le Dû (2018a);

• APN13 = Kronberger et al. (2014);

• PRC15 = Kronberger et al. (2016);

• Rev12 = Acker et al. (2012);

• A&A14 = Corradi et al. (2014);

• K2012 = Kronberger et al. (2012);

• O+T86 = Le Dû (2018b).

Table 1 includes 209 objects, of which 122 are True PNe, 51 are Likely PNe, and 36 are Possible PNe. Table 2 contains 610 objects, and Table 3 104 mimics of PNe (mainly emission-line stars, HII regions, and galaxies).

Fig. 3 Classification by the HASH team of 313 objects discovered by our group. On the x-axis are the types of objects (abbreviations from the HASH database); on the y-axis are the number of objects in each category.

4.2 Review of the discoveries of our PNe candidates

Of the 923 objects included in the Tables, 199 have a definitive classification and 114 have a temporary classification (see Fig. 3). Of these 313 classified objects ~67% are True (122), Likely (51), or Possible PNe (36) with the rest comprising mimics of various kinds (see Tables 1 and 3 and Sect. 4.2 below). The large majority we have discovered were unknown to professionals (94%) and are nearly all extremely faint and mostly compact (44.8% have a diameter smaller than 15 arcsec), while only 26.7% exceed 2 arcmin with the largest being 15 arcmin in diameter. The majority of these objects are located in the Galactic plane but nine (two Possible, three Likely, four True) or 4.3% are outside the northern Galactic plane IPHAS survey limits of −5° < b < +5° Galactic latitude.

These results demonstrate the high value of the work accomplished by the French-led amateur group in uncovering significant numbers of new Galactic PNe. These newly discovered objects are usually of very low surface brightness and may help probe the faint end of the PN luminosity function when distances can be determined. A major improvement in the success rate of our listed candidates could be achieved by deep Hα and [OIII] narrow-band filter imagery together with our amateur and professional spectroscopy. We provide the list of all of our candidates now, without waiting for complementary amateur data which may never come to fruition. We also inform our professional colleagues who may be better placed to obtain spectra on larger aperture telescopes, regardless of their final classification, so as not to confuse them in the future with real PNe.

Mimics. The imagery of some candidates can be confused with PNe (see e.g. Frew & Parker 2010). An amateur spectrum, even at low resolution, can often allow an indicative or even a definitive object classification. Common PN mimics include compact and isolated HII regions, Wolf-Rayet shells of high-mass stars, distant supernova remnants that may present as shells or ionised ISM around hot stars (see examples in Fig. 4).

Many objects of stellar appearance identified as candidates by their particular photometric properties actually turn out to be emission-line stars such as Be-type stars (see Fig. 4) or more often late-type stars whose strong molecular bands can give an apparent Hα excess compared to the broad-band r.

Some mimics are low-redshift emission-line galaxies with active nuclei easily recognizable by their redshifted emission lines that nevertheless still fall within the narrow-band Hα filter bandpass²².

An exceptional discovery of a very large object (Ou 4) radiating essentially in [OIII] and presenting a PN morphology is a perfect example of a mimic. Ou 4 has been the subject of a particular study and several publications (e.g. Corradi et al. 2014)²³. The origin of this object remains unexplained, but it seems related to the SH-129 nebula that surrounds it.

Fig. 4 PNe mimics. Left: 11.5 × 11.5 arcmin: Isolated HII region Fe 3 (SHS). Middle: 28.73 × 28.61 arcmin: Ionised ISM around hot star LDu 2. Image taken in France by Orange Observatory with a 0.4 m Newton telescope refractor (f/3.8), a cooled Moravian G4 CCD camera, a Paramount ME mount, and an Astrodon Hα filter. Exposure time: Hα 43 × 10 min. Exposure time total: 7.16 h (Nicolas Outters). Right: 3.22 × 3.21 arcmin: DeGaPe 33 emission-line star. Image taken in Chile by APO Team with 0.15 m refractor (f/7.3), a cooled Apogee ALTA U16M CCD camera, a mini-OHP MCMT mount, and Astrodon Hα, [OIII] and [SII] filters (APO Team).

4.3 Management of our discoveries

Candidate PNe retained during our research are first classified in Table 2 with a preliminary status (see Sect. 4.2). For some it is difficult to make a precise diagnosis with the data available on first discovery. Objects with the most robust PNe characteristics are transferred to Table 1 once their HASH classification has been determined. Others are transferred to Table 3. Objects awaiting a decision on their nature remain in Table 2.

Some candidates look like PNe but do not exhibit all the necessary characteristics for clear assignment. Some candidates reveal no similarity with PNe, but are considered interesting and worthy of further study in their own right (e.g. unusual nebulosities, nebulous haloes, jets near adjacent stars).

Our initial choice of object classification is also subjective and can be updated when the object’s true nature is later revealed via deeper imagery and/or spectroscopy. When a new object is found a name is assigned composed of the first two letters of the discoverer’s name. If the object is in Table 1, these initials are followed by a number. If the object is in Table 3, the word ‘Object’ is inserted between the initials of the discoverer and the number assigned. An update of the object name is possible after the discovery of its true nature and before it is moved to Table 1 or 3.

Some candidates are independently reported by different discoverers at similar times. By mutual agreement, the object name is composed of the initials of the different discoverers. Like the convention for our professional colleagues, this convention is used for amateur groups who have joined together to seek PNe candidates (for example StDr, Strottner-Drechsler) or who have a telescope held in common (for example DeGaPe). We perform a first-level triage by discarding objects considered false and also by classifying the particularly interesting objects in our tables. When new amateur discoverers contact us, we explain how to proceed. In this way, we present only relevant objects to our professional colleagues.

4.4 Database maintenance

Our database and accompanying tables are rigorously updated and curated. They are the basis of all communication within our community and of the spectroscopy and further study carried out. Updates are published on the planetarynebulae website²⁴ (see Sect. 4.4.2) partly created for this purpose.

4.4.1 Dissemination of our data

Once updated on the planetarynebulae website, the tables are sent to the HASH team at the Laboratory for Space Research (LSR²⁵) at the University of Hong Kong to be checked and migrated to the HASH PN database. Classification of the objects in our tables is independently assessed before inclusion in HASH. Our latest discoveries are also published in the French Science magazine L’astronomie every year, and then sent to the CDS for insertion into VizieR and Simbad.

4.4.2 The planetarynebulae website

The planetarynebulae website²⁴ was created to bring transparency and utility to the combined amateur work carried out and to make our work accessible to the PN research community.

The website allows users to visualise the candidates discovered and to follow the work of our collaborators. A data sheet is constructed for each object that provides detailed information on the candidate and selected images at different wavelengths centred on the object. Relevant amateur images and/or spectra may also be added. Published sources are indicated. Documentation and advice is also provided on how to search for PNe candidates, how to contribute to the analysis and how to communicate with the French-led amateur group. All spectra obtained by the group are available on a specific page of the website.

4.4.3 HASH and the integrity of our database

The HASH database²⁶ currently maintained by the LSR at the University of Hong Kong is the final arbiter of PN identification. It contains over 11,466 entries, as of March 2022, including over 3800 PNe in the Galaxy and 821 in the Magellanic Clouds. The remaining entries consist of the various mimics that have at one time or another been classified as PNe, emission objects found by the MASH and IPHAS teams but not PNe (spread across 40 different classifications), and a further 2802 candidates awaiting verification (see Parker et al. 2016; Bojičić et al. 2017).

A perfect complementarity has been established between the French-led amateur group and the professional team at the LSR that curates HASH and provides additional follow-up spectroscopy where possible on professional telescopes like the SAAO 1.9 m and Grantecan 10 m. This ensures that all objects discovered by amateurs are vetted, assessed, and sometimes spectroscopically confirmed independently by the HASH team before being incorporated into HASH and brought to the scientific community. It allows the HASH object identifier to be included by us, and allows us to check the past history of any common objects.

5 Spectral confirmation of our PN candidates

Planetary nebulae exhibit specific observational and physical characteristics in a large range of wavelengths from the UV to radio regimes. As seen above, high-quality multi-wavelength data sets and optical narrow-band images are of great help in confirming PNe candidates and in distinguishing them from their mimics (Frew & Parker 2010). An optical spectrum is essential in order to complete the diagnostic identification. Typical PN spectral signatures include the recombination lines of hydrogen and helium and the collisionally excited forbidden emission lines of heavier elements such as oxygen, nitrogen, neon, argon, and sulphur.

Spectroscopic confirmation of the first candidates observed by amateurs was conducted by professionals, and this is an ongoing process. Amateurs became aware of the importance of such observations and realised that at least for the higher surface brightness objects, they could undertake observations with their own telescopes and spectrographs. Amateurs started a dedicated spectroscopic observation programme in 2015 that led to the creation of the Planetary Nebulae Spectra Trackers (PNST²⁷) group in 2017 (Le Dû 2018a).

5.1 Amateur achievable objectives and equipment

In this section we describe the key objectives achievable with our available resources, including the specifications of the technical equipment.

Our newly discovered PN candidates are mostly low surface brightness objects and sometimes outside the Galactic latitude range of the narrow-band Hα surveys. The small apertures of our amateur telescopes is a major limiting factor. The main objectives of our amateur spectroscopy are the confirmation of our candidates via detection of typical PNe emission lines and the assessment of their relative intensities such as [O III] versus Hβ, Ha versus [N II], and Ha versus [S II]. This is in order to aid diagnostic identification (see e.g. Frew & Parker 2010).

To get better signal-to-noise spectra for our candidates, we use camera pixel binning and low-resolution spectroscopes. Kinematic information is effectively absent. The lack of telescope aperture is at least partially compensated for by very long exposure times (e.g. several hours on faint objects). This is something we amateurs can afford to do, unlike our professional counterparts.

Furthermore, we do not have to justify our need to carry out an observing campaign on specific objects and do not depend on the availability of professional observatories or on diverse programme priorities. Unfettered availability and responsiveness to take spectra on demand, coupled with the long time that can be spent to acquire data on a single object, makes this the key strength and advantage in pursuing this amateur programme.

We have shown that such objectives are achievable with our equipment for candidates detected on broad-band DSS2 red or blue wide-field survey data. This assumes they are emission line sources such as PNe, HII regions, supernova remnants, Wolf-Rayet shells, and emission-line galaxies. Other PN mimics like reflection nebulae or non-emission-line sources with a continuum may be undetected if they appear too faint on DSS2. Candidates only visible on IPHAS or SHS Ha surveys are those missed by the original professional searches. These tend to be very faint in Ha and not visible on the DSS2 red survey images. They are tough targets for amateurs, requiring excellent observing conditions to get low signal-to-noise spectra with detected emission lines. These sources are better pursued by our professional colleagues. The entire spectral acquisition process has been developed by core members of the Astronomical Ring for Access to Spectroscopy (ARAS)²⁸. ARAS is an informal group created in 2003, dedicated to the promotion of amateur astronomical spectroscopy and professional-amateur collaborations. This initiative is supported by professional astronomers based in France. Some of us have developed bespoke data reduction software and contributed to the design of efficient spectroscopes directly compatible with most telescopes and cameras available on the open market.

5.1.1 The equipment in our backyard observatories

Most spectroscopy (2018–2021) was performed from private backyard observatories (mainly Cornillon and Kermerrien in France). These provided ~60% of the data at the time of writing. Spectra are mainly acquired with small aperture Newtonian telescopes (0.2 m mirror and F/5) equipped with low-resolution spectrographs, for example the Alpy 600, with a 23 µm slit, cooled ATIK 414 EX CCD camera with dispersions of ~3.0 Å pixel⁻¹ at 6560 Å.

5.1.2 Missions to professional observatories

Every year since 2015 several amateur groups have observed at professional observatories to acquire data on faint objects beyond the reach of basic amateur equipment. By the end of 2021, 12 missions had been undertaken at three observatories. At the Observatoire de la Cote d’Azur (OCA), from 2017 to 2021, spectra were obtained with a 1 m F/6.5 Cassegrain telescope, the LISA spectrograph (R = 500), and ATIK 414 EX cooled camera with a dispersion of 2.5 Å pixel⁻¹ at ~6560 Å. At the Observatoire du Pic du midi, from 2017 to 2019, spectra were obtained with a 0.65 m F/3.5 Newtonian telescope, LISA spectrograph (R = 500), and ATIK 414 EX cooled camera, with a dispersion of 2.5 Å pixel⁻¹ at ~6560 Å. At the Observatoire de St Veran Astroqueyras, in 2015, 2016, 2018, 2019, and 2021, spectra were obtained with a 0.51 m telescope, LISA spectrograph (R = 500), and ATIK 414 EX, giving a dispersion of 2.5 Å pixel⁻¹ at ~6560 Å.

These missions, performed using almost identical equipment, provided ~22% of our data. Some amateur groups dedicate their observing time to confirming PN candidates with small aperture telescopes (0.2–0.36 m). These missions have provided ~9% of the data. Finally, the 2SPOT consortium operate a remotely controllable 0.3 m telescope in Chile, operational since the second half of 2021 (see Sect. 6). It provided ~6% of the data used in this paper but will play an increasing role in the future.

Supplementary to this are missions led by our professional colleagues to the SAAO 1.9 m in 2017, 2019, and 2020 and to the Grantecan 10 m telescope in 2016 and 2017 that provided confirmatory spectra for 34% of our candidates in Table 1.

Fig. 5 Sample of spectra taken directly from the HASH database for comparison and to show what the HASH interactive spectral plotting utility looks like. Individual emission lines can be identified and labelled. Various spectra can be overplotted and the cursor used to zoom in on interesting regions. In this example the green spectra are from the SAAO 1.9 m professional telescope and the red spectra are from our amateur 0.2 m telescope. Top left: Kn 42 True PN; top right: MPA J1827-1328 True PN; bottom left: PM 1-322, possible symbiotic system; bottom right: Kn 134, emission-line galaxy.

5.2 Amateur data reduction and follow-up

Our amateur spectroscopy is reduced using the Integrated Spectrographic Innovative Software (ISIS) software²⁹. All spectra are dark, bias, and flat-field corrected as standard. The instrumental response is derived from a reference star spectrum acquired under the same observing conditions as the target. Wavelength calibration is done via an argon-neon calibration lamp, and is checked against absorption lines of the reference star in the blue part of the spectrum. Fully processed 2D spectra are first produced after dark, bias, flat-field, instrumental correction, wavelength calibration, and the removal of atmospheric lines and cosmic rays. This often reveals the PN emission lines clearly in this case prior to extraction of the 1D spectrum. To confirm the reliability of the wavelength calibration and the instrumental response correction, we also check if the 1D spectrum of the reference star matches its theoretical spectrum.

Reduced data are sent to the LSR HASH team at Hong Kong University for inclusion and further assessment following Parker et al. (2006). The amateur team provides a detailed observation sheet for each PNe candidate³⁰. This includes the PN ID, discovery imagery, telescope used, observing conditions, exposure time, finding chart (with the slit position for the observation), wavelength verification, instrumental response correction, processed 2D and wavelength calibrated 1D spectra, other relevant images, relevant background and/or papers for the object, and a short analysis.

Fig. 6 Distribution of all known True, Likely, and Possible PNe on an all-sky Aitoff projection in Galactic coordinates (powered by Aladin: https://aladin.u-strasbg.fr/). The Galactic Centre is in the middle of the map. Blue dots: HASH PNe; small red circle: PNe from our list with diameter <2arcmin; large red circle: PNe from our list with diameter >2arcmin; small orange circle: PNe from DSH sample <2arcmin; large orange circle: PNe from DSH sample >2arcmin.

5.3 Comparison with professional data

Amateurs and professionals have sometimes independently obtained confirmatory spectra of the same candidates (in addition to the more dedicated professional missions specifically pursued to spectroscopically classify some of our amateur observed candidates). This makes it possible to compare and evaluate the reliability of the amateur spectroscopy. In Fig. 5 we present four examples of relatively bright, new candidate PNe observed both by the amateur group on a 0.2 m telescope and the SAAO 1.9 m telescope by our HASH colleagues led by QAP. This gives an indication of what our amateur spectroscopic facilities can provide. Typical amateur spectroscopic observations last from 1 to 3 h via sets of 20 min exposures, while the professional spectra vary from 5 to 30 min and are usually single. Our telescopes are in France (northern hemisphere), while the SAAO 1.9 m is in the south, so only candidates accessible from both sites can be compared. This places limits on the comparisons. Amateur results generally agree with professional spectra in terms of main emission line detections and resolution. These comparisons confirm that amateurs can provide decent, first estimation spectra for higher surface brightness candidates pending deeper professional study.

5.4 Overall results

By the end of 2021 the PNST group had acquired 283 spectra of PN candidates, of which 219 are considered True, Likely, or Possible PNe. Our first confirmatory spectra were taken in 2011, but more intensive observations started in 2015 with -20 spectra per year, significantly increasing after 2019. More than 200 spectra of candidates were obtained between 2019 and 2021, and 145 were also professionally confirmed as PNe by the HASH team. Observations are almost exclusively made from the northern hemisphere from France, which limits the sample of candidates that can be followed up directly by us. Apart from the spectral confirmations by our HASH colleagues, the southern hemisphere is now also covered by the 2SPOT consortium (see Sect. 6).

Our spectroscopy has not only been focused on our amateur discoveries; ~51% have been dedicated to previously discovered candidates in the astronomical literature missing optical spectra in HASH. This includes some objects in the IPHAS, IRAS, or DSH catalogues. In many cases our spectra provide the first PN spectroscopic confirmation.

Quite a few of our amateur discoveries actually postdated professional discoveries and even spectroscopic follow-up by many years. This is particularly the case with candidates found in the IPHAS (Drew et al. 2005) northern Galactic plane Hα survey. There are several reasons. Firstly, the registration of amateur discoveries is a quick process and that can go ‘live’ on our website almost immediately. Secondly, until recently, there has not been a decent PN professional-amateur collaboration for northern sources. Thirdly, professional astronomers tend to collect many candidates together and hold back announcing them until they get confirmatory spectra, and then publish their discoveries as a set in a professional journal (e.g. Parker et al. 2006; Miszalski et al. 2008; Sabin et al. 2012). Only for particularly interesting objects is a separate publication made. As the number of professional candidates requiring spectroscopic follow-up is large, this can lead to a long delay before discoveries are made public. At the same time the amateur community has access to the narrow-band survey materials online providing them with the opportunity to pursue their own candidate searches of these surveys and make any discoveries known immediately, even if they are already known to some members of the professional community. One of the main motivations for this current professional-amateur collaboration is to help resolve these issues equitably.

Table 4 presents the list of our spectral observations of PN candidates up to December 2021. Amateur spectroscopy of some of our PNe candidates and other PNe candidates from the HASH database have revealed a further 62 PN mimics. These mostly comprise HII regions, emission-line galaxies, supernova remnants, and emission-line stars. These objects are presented in Table 5 together with the preliminary identifications from their spectra to prevent future confusion. Figure 6 presents the distribution of True, Likely, and Possible PNe from our list, from the DSH list (only Kn and Pa objects), and from the HASH database (which includes the LMC PNe) on an all-sky Aitoff projection in Galactic coordinates (powered by Aladin³¹). The Galactic Centre is in the middle of the map.

6 Summary and future work

We have shown how a dedicated amateur group can combine searches for PNe candidates from wide-field multi-wavelength surveys with searches from their own facilities for deeper confirmatory, narrow-band imaging, and spectroscopy to provide an extremely powerful programme of Galactic PNe discovery. The French-led group have succeeded in creating a new dynamic around the ongoing search for PNe and their subsequent confirmation and analysis. Whether team members are involved in uncovering candidates from careful survey searches or in obtaining follow-up narrow-band imagery or spectroscopy, each has specific tools for the task. Regardless of the varied geographical disposition, everything is carefully integrated together.

Amateurs only rarely directly access large-diameter professional telescopes and the limit of their own equipment is quickly reached, which is why this professional–amateur collaboration is so valuable. The biggest challenge is perfecting and improving our tools and creating new ones that will assist future PN hunting activities. New spectrographs developed by Shelyak Instruments (UVEX in particular) will bring improvements to the quality of the spectra produced by amateurs. The planetarynebulae.net website will continue to grow and will offer improved interfaces for data synthesis and retrieval. The sustainability of the existing observatories is essential in order to continue making quality observations. Hardware will be kept up to date and operational. Finally, 2021 marked the arrival in Chile of the observatory of the 2SPOT association (created in 2019 by five French amateur astronomers). The infrastructure of the Deep Sky Chile (DSC) company, located just a few dozen kilometres from the Cerro Tololo observatory in Chile, was chosen so as to benefit from very high-quality sky. The 2SPOT equipment consists of a 12-inch Ritchey-Chretien telescope on an equatorial 10 micron GM 3000 mount. For spectroscopy, the very stable Alpy 600 spectrograph from Shelyak Instruments (resolution 600) was chosen.

Acknowledgements

Our pro-amateur group was forged with time and perseverance to build a team able to work together and bring the knowledge and competence of each to this research field with the discovery and analysis of PN candidates. The people involved are numerous and each one contributes, according to their degree of involvement and know-how to the relevance of the work and the reactivity that amateurs can show. Q.A.P. thanks the Hong Kong Research Grants Council for General Research Fund research support under grants 17326116 and 17300417. A.R. thanks H.K.U. for the provision of a postdoctoral fellowship. We would also like to thank Prof. Agnès Acker whose involvement in the amateur community greatly contributed to the birth of this pro-am collaboration. We also thank all the discoverers, photographers, and observers who contributed to this collaboration: F. Appert, G. Arlic, P. Bazart, K. Beaudoin, M. Behnke, F. Bellenfant, P. Bernhard, T. Bickle, S. Binnewies, M. Blauensteiner, T. Bohlsen, L. Bourgon, A. Brémond, A. Bringmann, C. Buil, J.P. Cales, J.C. Canonne, D. Chaplain, S. Charbonnel, S. Chareyre, T. Demange, G. Donatiello, J. Drudis, P. Dubreuil, K. El Kanbi, P. Erdmann, D. Erpelding, L. Ferrero, B. Foucher, J. Gallardo, R. Galli, J. Garcia, M. Germiniani, J. Gleason, S. Gloaguen, P. Goodhew, J. Guarro, B. Guégan, M. Hanson, D. Harmer, R. Hawley, T. Henne, R. Hess, R. Hoban, L. Huet, J. Illner, F. Jobard, M. Kronberger, V. Lecoq, T. Lemoult, M. Leveque, A. Lopez, F. Mathieu, U. Mishra, S. Mohan, E. Mol, K. Morefield, G. Murawski, J.P. Nougayrede, Z. Orbanic, D. Orriols, B. Paczkowski, M. Palenik, R. Pasturel, R. Pölzl, J. Pöpsel, C. Prost, K. Quin, C. Raïssi, S. Rasool, F. Romanov, T. Salomon, J. Shuder, J. Souchu, M. Stiles, C. Sullivan, G. Sun, H. Tan, M. Vanhuysse, D. Walliang, J. Zanon, M. Zauner, A. Zirke, S. Zoll. Part of this work was supported by the German Deutsche Forschungsgemeinschaft, DFG project number Ts 17/2–1.

References

All Figures

	Fig. 1 PNe discovered from high-resolution images of the DECaPS DR1 colour survey by the amateur group. Orientation of all panels has north-east to the top left. Left: PN Pre 44, image size 3.34 × 3.27 arcmin; Centre: PN Pre 59 (blue CSPN visible), image size 1.34 × 1.32 arcmin. Right: candidate Pre 63 (blue CSPN also just visible at PN centre), image size 1.12 × 1.10 arcmin.
In the text

Fig. 2 Very deep combined narrow-band interference filter images ([OIII] and Hα+[NII]) of PNe by our group. Orientation of all figures has north-east to top left. Left: PN Fe 6 (IPHASX J015624.9+652830). The 7.75 × 7.67 arcmin image was taken in southwest Spain with twin 0.15 m refractors (f/7.9), twin cooled QSI6120 CCD cameras, and Astrodon Hα, [OIII], and LRGB filters. Exposure time: [OIII] 23 × 30 min binning 2 × 2; Hα 27 × 30 min binning 2 × 2; L 21 × 5 min binning 1 × 1; R 24 × 5 min binning 1 × 1; G 29 × 5 min binning 1 × 1; B 25 × 5 min binning 1 × 1. Exposure time total: 41.4 h (Peter Goodhew). Centre: PN StDr 13. The 41.37 × 40.92 arcmin image was taken in Germany at Dachsternwarte with a 0.28 m Schmidt-Cassegrain telescope with a Starizona Hyperstar 4 (f/1.9), a cooled ZWO ASI 1600MM CMOS camera, Baader Hα, [OIII], and Astronomik LRGB filters. Exposure time: [OIII] 40 × 10 min; Hα 40 × 10 min; LRGB 100 × 2 min per filter. Total exposure time: 27 h (Andreas Bringmann). Right: likely PN StDr 20. The 13.86 × 13.7 arcmin image was taken at Chilescope observatory with a 1 m Ritchey-Chretien telescope (f/8), a cooled FLI Proline 16803 CCD camera, forkmount, Hα, [OIII] filters. Exposure time: [OIII] 32 × 20 min; Hα 32 × 20 min; RGB 18 × 10 min per filter. Exposure time total: 24.4 h (Xavier Strottner, Marcel Drechsler).

In the text

	Fig. 3 Classification by the HASH team of 313 objects discovered by our group. On the x-axis are the types of objects (abbreviations from the HASH database); on the y-axis are the number of objects in each category.
In the text

Fig. 4 PNe mimics. Left: 11.5 × 11.5 arcmin: Isolated HII region Fe 3 (SHS). Middle: 28.73 × 28.61 arcmin: Ionised ISM around hot star LDu 2. Image taken in France by Orange Observatory with a 0.4 m Newton telescope refractor (f/3.8), a cooled Moravian G4 CCD camera, a Paramount ME mount, and an Astrodon Hα filter. Exposure time: Hα 43 × 10 min. Exposure time total: 7.16 h (Nicolas Outters). Right: 3.22 × 3.21 arcmin: DeGaPe 33 emission-line star. Image taken in Chile by APO Team with 0.15 m refractor (f/7.3), a cooled Apogee ALTA U16M CCD camera, a mini-OHP MCMT mount, and Astrodon Hα, [OIII] and [SII] filters (APO Team).

In the text

Fig. 5 Sample of spectra taken directly from the HASH database for comparison and to show what the HASH interactive spectral plotting utility looks like. Individual emission lines can be identified and labelled. Various spectra can be overplotted and the cursor used to zoom in on interesting regions. In this example the green spectra are from the SAAO 1.9 m professional telescope and the red spectra are from our amateur 0.2 m telescope. Top left: Kn 42 True PN; top right: MPA J1827-1328 True PN; bottom left: PM 1-322, possible symbiotic system; bottom right: Kn 134, emission-line galaxy.

In the text

Fig. 6 Distribution of all known True, Likely, and Possible PNe on an all-sky Aitoff projection in Galactic coordinates (powered by Aladin: https://aladin.u-strasbg.fr/). The Galactic Centre is in the middle of the map. Blue dots: HASH PNe; small red circle: PNe from our list with diameter <2arcmin; large red circle: PNe from our list with diameter >2arcmin; small orange circle: PNe from DSH sample <2arcmin; large orange circle: PNe from DSH sample >2arcmin.

In the text