\cite{Shore_2005}
We found in Section \ref{sec:veloc-disp} that the radial velocity of NGC 1980 is indistinguishable, or has a difference of the order of a few km/s at best, from the radial velocity of the embedded Trapezium population. This surprising result implies that the radial velocity of NGC 1980 is essentially the same as the velocity of the gas in the Orion A cloud, since the ONC population has the same radial velocity as the cloud \citep{2009ApJ...697.1103T}. This strongly suggests that NGC 1980 is somehow connected to the Orion A cloud, or better, that the cloud that formed NGC 1980 was physically related to the current Orion A cloud. One would not expect the distance to NGC 1980 to be substantially different than the current distance estimate to the ONC, and a fitting of the ZAMS on the optical data presented in this paper is indeed consistent with a distance of 400 pc.
Despite its relatively older age, lack of obvious H\(_{\rm{II}}\) region, and lack of measurable dust extinction, NGC 1980 moves away from Earth at the same velocity as the large Orion A cloud on which it is seen in projection. Because of their likely proximity, one wonders about the effects of the ionizing stars from NGC 1980 on the Orion A cloud, or what the cloud was about 4-5 Myr ago, in particular about a possible acceleration and compression of the cloud by the UV radiation from these stars. How significant was/is this process in this region? Could the formation of the ONC have been triggered by its older sibling, as suggested in \citet{Bally2008}? At first glance our results seem to argue that the impact would have been minimal, NGC 1980 has essentially the same radial velocity as the Orion A cloud, but the work of \citet{1954BAN....12..177O}, \citet{1954BAN....12..187K}, and \citet{Oort:1955cn} suggests that final speeds between the ionizing star and the cloud would be of the order of a few km/s, which cannot be ruled out by the current accuracy of the data. While our results do not give final evidence in support of the tantalizing suggestion that the formation of the ONC could have been triggered by NGC 1980, they are not inconsistent with it either. A new dedicated radial velocity survey of the region, together with a sensitive proper motion survey, are needed to understand the interplay between these two massive clusters. This configuration (an embedded cluster in the vicinity of an \(\sim 5\) Myr cluster) is unlikely to be unique in massive star-forming clouds, but it will best addressed in the nearest example.
We have shown above that there is a rich and distinct foreground population of stars, likely associated with the young (\(\sim 5\) Myr) poorly studied but massive NGC 1980 cluster, which is not directly associated with the ongoing star formation in the ONC. This finding raises concerns on the contamination of currently available observables for this important region, and future studies should take this foreground population into account. But how strong is this contamination? There are two well-known ONC catalogs used in the literature, namely \citet{Hillenbrand1997} and the catalogs of \citet{2009ApJS..183..261D}, \citet{DaRio:2010cz}, and \citet{2012ApJ...748...14D} which cover a roughly square area of about \(0.5^\circ\times0.5^\circ\) (\(\sim 3.5 \times 3.5\) pc) centered on the Trapezium cluster. The \citet{2012ApJ...748...14D} catalog supersedes all previous catalogs, but it is the most recent, hence least used in the community. On the other hand, the \citet{Hillenbrand1997} catalog has been used extensively in the literature and has spawned a large number of the star formation studies on the star formation properties of the ONC region. We estimate here the likely foreground contamination fraction for the \citet{Hillenbrand1997} catalog because it is the most used one, but also because it is likely to be the least contaminated since the Da Rio catalogs cover a slightly larger area of the sky toward NGC 1980.
To estimate the probable contamination fraction of \citet{Hillenbrand1997}, we matched the foreground population with this catalog for stars falling within the A\(_V \geq 5\) mag region where the foreground was selected (see Figure \ref{fig:density}) and where I-band \(<\) 16 mag. The last constraint accounts for the fact that the \citet{Hillenbrand1997} sample is not uniformly deep (it reaches about 2 magnitudes deeper around the Trapezium cluster), and that the selection of foreground stars, made at g-band, seems complete to about I-band \(\sim\) 16 mag (after transformation of the SDSS photometry into Johnson’s \citep{2007AJ....134..973I}). We find that 11% of the sources in the \citet{Hillenbrand1997} catalog have a match in the foreground sample (8% if we remove the constraint on the I-band brightness). If one sees the Trapezium cluster as a component of the ONC, and not as the only component, and excludes it from consideration, the fraction of foreground contaminants in the ONC rises to 32%. For this estimate the area on the sky covered by the Trapezium cluster is taken as an ellipse with a\(=7.5^\prime\) and b\(=3.8^\prime\), with a position angle of \(-10^\circ\), similar to the definition in \citet{Hillenbrand1998}. One can also estimate the possible contamination to the entire \citet{Hillenbrand1997} catalog by applying equation \eqref{eq:1} to it, which yields a contamination fraction of 20%, or 63% when the Trapezium is excluded from consideration. Note that all these estimates assume that the fraction of ONC stars without measurable extinction is negligible, which is likely given the distribution of foreground stars in Figure \ref{fig:density}, but will need to be investigated further in future work.
Even including the Trapezium cluster in the consideration, contamination fractions of about 10–20% are significant and will necessarily lead to systematic errors in the basic derived physical quantities for this star formation benchmark. Still, these are necessarily lower limits to the true contamination fraction of the ONC sample for at least two reasons: 1) our g-band Megacam survey is not as sensitive in regions of high nebular brightness, especially around the Trapezium, and 2) we are not sensitive, by design, to background sources. While it is normally argued that the high background extinction behind the Trapezium blocks most background stars, this is only valid for the inner regions of the Trapezium cluster (\(\sim25^{\prime \,2}\)), but not valid for the entire \(\sim700^{\prime \,2}\) Orion nebula, \citep[e.g.][]{1999ApJ...510L..49J,2012MNRAS.422..521B}. So background contamination is variable across the ONC and expected for any optical or infrared survey of this region. With regard to the contamination of the ONC region by NGC 1980, it is a function of the position in the nebula, which is at a minimum at the center of the Trapezium where the Trapezium cluster stellar density is highest, gradually increasing toward the south as one approaches the core of NGC 1980 (see Figure \ref{fig:density}).
\label{sec:unrel-galact-field}
Most of the identified foreground population (624 sources) is likely to belong to NGC 1980, as seen from the symmetric and peaked spatial distribution in Figure \ref{fig:density}, but some fraction of these must consist of the Galactic field population between Earth and the Orion A cloud. A first and simple estimate of the size of this population can be made by correlating the foreground population with the sample of \citet{2009ApJ...697.1103T} for which good radial velocity measurements exist (see Figure \ref{fig:contamination}). The distribution of radial velocities of the 188 sources in common to both populations reveals a Gaussian like distribution centered at \(\sim 26.1\) km/s, with a gap roughly between \(-20\)-\(0\) km/s and \(40\)-\(60\) km/s without any source, and three sources below \(-20\) km/s and nine above 60 km/s (i.e., 12 potential outliers). Not surprisingly, if we sigma-clip the entire distribution at 3 \(\sigma\), we find 11 outliers. If we remove the 12 potential outliers from the distribution described above and then sigma-clip the rest of the distribution, we find five more outliers, but this time at the wings of the distribution. This suggests that about 6–9% of the foreground sources identified in this work are likely field sources unrelated to NGC 1980. This estimate of the Galactic field foreground contamination for Orion A roughly agrees with what would be expected from the Besaçon stellar population model \citep{Robin03} for the depth of our Megacam survey.