4 Study of the host galaxy environment

At present it is unknown if GRB host galaxies preferentially are located in dense environments, or if there is any correlation between the local density of galaxies and the presence of a GRB. So far the two z=2.04 bursts, GRB 000301C and GRB 000926, are the only ones for which the environment of the host galaxy has been studied (Fynbo et al. 2002). In both of these fields a number of galaxies at the same redshift were detected, but the lack of blank fields studied to similar depth prevented those authors to conclude if the GRB host fields were overdense. The photometric redshifts of the galaxies in the GRB 000210 field provide the opportunity to look for other galaxies in its environment. The same calibration and photometry software used to obtain the host galaxy magnitude was applied to the rest of the objects in the field.

Figure 2: The evolution of the fitted SED $\chi ^{2}/$dof as a function of the photometric redshift. The dotted vertical line indicates the spectroscopic redshift proposed by Piro et al. (2002). As shown the minimum of $\chi ^{2}/$dof (at z=0.842) is consistent with the spectroscopic redshift.

Figure 3: The plot shows the redshift distribution of galaxies in a area of $6^{\prime } \times 6^{\prime }$ around the GRB 000210 host. The sample is made up of the 169 galaxies whose SEDs were fitted with $\chi ^2/dof < 2$. The vertical dashed line shows the spectroscopic redshift. As shown there is no special concentration of galaxies with redshifts similar to the host galaxy. For the construction of the histogram we assumed a MiSc79 IMF and the extinction law given by Calzetti et al. (2000). Other IMFs and extinction laws yield similar results.

We consider that an object is suitable for redshift determination when it is detected at least in four bands. Objects detected in less than four filters were rejected due to the large uncertainty in the determination of their redshifts. The considered region covers a $6^{\prime } \times 6^{\prime }$ area around the host galaxy. At the redshift of the host galaxy (z=0.8463) this corresponds to $\sim$3 Mpc $\times\ 3$ Mpc. The SEDs used to determine the photometric redshifts of the field objects consist of 8 synthetic templates (Stb, E, S0, Sa, Sb, Sc, Sd, Im) based on a MiSc79 IMF and the extinction law given by Calzetti et al. (2000). As in the case of the host SED, four additional observed spectra from CWW were considered.

Among the 169 galaxies of the field with acceptable fits ($\chi ^{2}/$dof < 2) we considered the ones with photometric redshifts compatible (within $1\sigma$) with a $\Delta z = \pm 0.1$ redshift range around the host galaxy spectroscopic redshift. In Fig. 1 a deep V-band image around the host galaxy is displayed. The FOV covered by the image is $5^{\prime } \times 5^{\prime }$. The circles represent the galaxies having photometric redshifts consistent with 0.7463 < z < 0.9463. The radius of each circle is proportional to the inverse of the absolute B-band magnitude (1/|M_B|) of the galaxy contained inside. From the distribution of the circles on the image it is clear that there is no obvious concentration of galaxies around the host (indicated by the tick marks). The lack of clustering around the host galaxy can also be visualised in Fig. 3, where the redshift distribution of the galaxies in the field are plotted. Figure 3 shows that there is no spike of galaxies at the redshift of the host. The same study was performed considering the 200 galaxies with SEDs fitted having $\chi ^2$/dof < 3, again yielding no apparent concentration of objects around the host. The exercise was repeated using several extinction laws and IMFs, giving similar results.

Figure 4: The points show the measured flux in the UBVRIZJs HKs bands for the GRB 000210 host, once the Galactic dereddening is introduced (Schlegel et al. 1998). The solid curve represent the best SED fitted to the photometric points ($\chi ^2$/dof = 1.096), corresponding to a starburst synthetic template at a redshift of z = 0.842 generated with a Sa55 IMF. The derived value of the starburst age corresponds to 0.181 Gyr. The fit is consistent with a very low extinction, ( $A_{\rm V} \sim 0$). The extinction law used to construct the plot is given by Calzetti et al. (2000). The SED shows a prominent $\sim 4000 \times (1+z)$ Å break, bracketed between the R and I-band photometric points, typical of early galaxy types (Stb, E, S0) with stellar population ages >108 yr.

We have calculated the neighbour detectability threshold of our images; in other words, the minimum luminosity of a neighbour galaxy for which our data allows a photometric redshift determination. With this purpose the GRB 000210 host galaxy SED has been dimmed until detecting it only in seven bands (UBVRJsHKs) above the limiting magnitudes given in Table 1. We consider that the minimum number of bands to have an acceptable redshift determination is four. So, if the host galaxy SED was dimmed in all filters by 1.16 magnitudes (see the limiting magnitudes of Table 1 and the host galaxy magnitudes of Table 2), it would have been still detected in seven bands and a secure photometric redshift determination would have been still possible. If the SED is dimmed by more than $\sim$1.29 mag then the host would have been detected only in RJ (and may be marginally in HK), so no redshift determination would have been possible. The absolute B-band magnitude of a host galaxy-like SED at z=0.8463, 1.16 mag shallower, is M_B=-19.0. So, a photometric redshift of a neighbour galaxy (with a SED similar to the host) with M_B > -19.0 would have been indeterminable. This magnitude corresponds to a Luminosity of $L=0.23 \ L^{\star}$ (considering $M^{\star}_{B}=-20.6$, following Schechter 1976). The value deduced for L based on Lilly et al. (1995) ( $M^{\star}_B=21.33$, discussed in Sect. 5.3) corresponds to $L=0.12 \ L^{\star}$. Thus M_B=-19.0 implies a luminosity ranging from 0.12 to $0.23\ L^{\star}$ depending on the adopted $M^{\star}_{B}$ value. Therefore we consider $L=0.18 \pm 0.06\ L^{\star}$ as an indication of the limiting luminosity of our host environment study. This procedure assumes that the galaxies in the host environment have similar SEDs, thus their SEDs can be reproduced by dimming the GRB 000210 host galaxy SED by the same factor in all bands.