Two ECHs were observed with SOHO/SUMER on seven days during October and November of 1999. A total of 12 FUV lines were recorded during the passage of the holes across the disk in October and November of 1999 as part of SOHO Joint Observing Program (JOP) 115. Full-disk radio, H $\alpha$ and EUV observations were made synoptically during all of the SUMER observations. The difraction-limited resolution for the radio images is 10 $^{\prime\prime}$ for observations made within 2 hours of local noon. Outside this interval resolution decreases due to a decrease in projected array length. The effective resolution will also depend upon the image construction, which introduces artifacts (Nindos et al. 1999). The images collected from the Nobeyama web server appear to have a resolution of $\sim$ 18 $^{\prime\prime}$ , since there are no features in the images with smaller length scales. Images at visible and UV wavelengths include features with dimensions at their resolution limits of 1-4 $^{\prime\prime}$ , so it seems unlikely that 17 GHz features would be limited to length scales above 18 $^{\prime\prime}$ , unless their apparent size is limited by resolution. Radio images are provided in a 512 by 512 frame at 4.91 $^{\prime\prime}$ /pixel. H $\alpha$ images at line center are recorded daily at the Big Bear Solar Observatory (BBSO) and provided in an 1889 by 1889 frame at 1.07 $^{\prime\prime}$ per pixel and 1 $^{\prime\prime}$ to 2 $^{\prime\prime}$ resolution. Full-disk images in Fe XII 195 Å emission are recorded daily by EIT on board the SOHO spacecraft with a resolution of 2 $^{\prime\prime}$ and provided in a 1024 by 1024 frame at 2.6 $^{\prime\prime}$ /pixel. The Fe XII images are used to identify CHs and map their boundaries. Full-disk longitudinal magnetograms were recorded daily by the Michelson Doppler Imager (MDI) on board SOHO and provided in a 512 by 512 frame at 3.9 $^{\prime\prime}$ /pixel.

The FUV observations with SUMER in several chromospheric and transition region lines included portions of two ECHs and the surrounding areas (Wilhelm et al. 2001). The slit used was one arcsecond wide and either 120 $^{\prime\prime}$ or 300 $^{\prime\prime}$ long, and was scanned across the solar surface. A sub-frame image was constructed from the individual slit position measurements. Individual exposure times were 150 s and total exposure times were 4 to 6 hours. The regions observed varied from 100 to 300 $^{\prime\prime}$ in width, with a height of either 120 or 300 $^{\prime\prime}$ . All of the observations considered except one included area both inside and outside the ECH and all of the observations included regions of radio enhancement. The dates of the observations, ion species, wavelengths of the lines and their formation temperatures ( $T_{\rm f}$ ) are listed in Table 1. The ionization states range from neutral (H⁰, O⁰) to Ne⁷⁺, and formation temperatures from 10000 K to 630000 K. Since lines are formed over a range of temperature, $T_{\rm f}$ indicates the mean. The formation temperature limits for a given line are determined by the ionization potentials, and excitation energies. The table also notes the type of region contained in the field of view such as QS and CH, the difference in ground and excited state energies, $\chi$ , and a factor $\alpha$ , which will be discussed in Sect. 3.

Radio enhancements were present in five of the ECH observations. The outline of each CH was determined from the EIT Fe XII 195 Å image. There were small, well-defined ECH radio enhancements of 15 to 20% peak in the SUMER field of view ranging in size from 20 $^{\prime\prime}$ to 50 $^{\prime\prime}$ (FWHM) on October 24 and 29, 1999, as well as larger diffuse radio enhancements of 5 to 10% peak on three other observation days. The well-defined enhancements are found in regions of strong unipolar magnetic flux, consistent with previous observations (Gopalswamy et al. 1999a,b). Since the bright features are probably not resolved at the 18 $^{\prime\prime}$ to 20 $^{\prime\prime}$ resolution of the radioheliograph, the true enhancement is probably larger and therefore the temperature increase in the enhancement is at least 15 to 20% above the QS level.

Images of an ECH in 17 GHz radio, Fe XII and H $\alpha$ brightness and the magnetogram recorded on October 24, 1999 are shown in Figs. 1a-d.

$\begin{figure} \par\includegraphics[width=8.8cm]{fig1_2629.ps}\end{figure}$

Figure 1: Images of the October 24, 1999 ECH in a) 17 GHz radio, b) 195 Å (EIT Fe XII), c) H $\alpha$ emission, and d) the corresponding SOHO/MDI magnetogram. The CH boundary is outlined in white and the SUMER field of view is outlined in black.

The boundaries of the ECH and SUMER field of view (391 $^{\prime\prime}$ to 553 $^{\prime\prime}$ W, 0 $^{\prime\prime}$ to 300 $^{\prime\prime}$ N) are outlined in each image. Note the chain of bright regions within the hole boundary and SUMER field of view in the radio image. The particular Fe XII and BBSO H $\alpha$ images and MDI magnetograph considered for comparison were obtained closest in time to the UV observations. Ideally, comparisons should be made with simultaneous observations, but none were available. However, bright features present in the radio image at 03:06 UT on Oct. 24, during the UV observations, were clearly visible in the H $\alpha$ image obtained at 15:26 UT on Oct. 23, 12 hours earlier, and in the magnetogram obtained at 13:02 UT on Oct. 23, 14 hours earlier. In addition, the bright features were present in the radio images until the end of the observing day on Oct. 24, at 05:30 UT. Thus the features had a lifetime of at least 16.5 hours. These areas coincide generally with regions of strong unipolar flux and enhanced H $\alpha$ and Fe XII emissions. The magnetic fields measured in these regions with MDI had the largest strengths seen in the ECH enhancements, 150 G. These results are consistent with the previous analysis of radio and H $\alpha$ emission in CHs by Gopalswamy (2000), who first pointed out the association of microwave radio and H $\alpha$ enhancements in ECHs.

The October 23 and 29 observations both show that some of the radio enhancements are the bases of equatorial coronal plumes as recorded by EIT at 195 Å. This is consistent with observations of polar plumes, which showed that those plumes originate from strong unipolar flux concentration (DeForest et al. 1997). In addition to having the same magnetic field correlation, radio enhancements and plumes also have similar time variability, with both showing strong intensity fluctuations on time scales of ten minutes or less (for radio variability, see Gopalswamy et al. 1999, for plume variability, see DeForest et al. 1997). Thus, there seems to be a strong link between radio enhancements and equatorial coronal plumes. One significant difference between polar and equatorial plumes is the location of the radiant, or point of apparent origin of the plumes within the sun. Polar plumes appear to originate from a radiant which is approximately 0.6 $R_{\rm o}$ from the disk center, where $R_{\rm o}$ is the solar radius (Fisher & Guhathakurta 1995), while the equatorial plumes observed on October 23, 1999 appear to radiate from a point much closer to the surface, approximately 0.9 $R_{\rm o}$ . This effect may be partially caused by strong flux concentrations outside the ECH, such as the bipolar structure bordering the northeast boundary of the hole.

Sub-frames of the October 24 radio emission and H $\alpha$ brightness in the SUMER field of view are shown in Figs. 2a,b, the corresponding magnetogram is shown in Fig. 2c and the SUMER O I 948.7 Å , H I Ly 4 949.7 Å and He II 958.6 Å frames are shown in Figs. 2d-f.

$\begin{figure} \par\includegraphics[width=8.8cm]{fig2_2629.ps}\end{figure}$

Figure 2: SUMER field of view sub-frames of the October 24, 1999 observations. a) 17 GHz radio data, b) re-binned, smoothed H $\alpha$ image, c) magnetogram, and SUMER d) O I 948.7 Å, e) H I Ly 4 949.7 Å, and f) He II 958.6 Å images.

$\begin{figure} \par\includegraphics[width=8.8cm]{fig3_2629.ps}\end{figure}$

Figure 3: Plots of the October 24 radio observations (solid), H $\alpha$ (dotted), H I Ly 4 (dashed) and He II (dot-dash) brightness along pixel Col. 20, which is marked at the bottom by a vertical white dash in Figs. 2a-f. The ECH boundary is marked by a vertical dash on the abscissa.

The UV line emission is slightly lower in the ECH than in the quiet sun, but increases near the radio and H $\alpha$ peaks to quiet sun levels. Thus, the upper chromosphere and transition region in the radio enhancements appears to be hot relative to the ECH outside the enhancements, but not relative to the quiet sun. This suggests that the mechanism responsible for radio enhancements also affects the chromosphere and transition region. The difference in location between the radio, H $\alpha$ and UV emission peaks may be due in part to the spreading of field lines with height. In the case of the UV emission, observations have shown that the vertical scale height is increased inside polar coronal holes (Huber et al. 1974), which is likely true for ECHs as well. In addition, the radio, H $\alpha$ and UV observations were not made simultaneously, so the network structure could have changed slightly between measurements.

To summarize the primary results, we compared radio emission in ECH regions showing enhancements with FUV, H $\alpha$ and EUV Fe XII brightness. There was no increased SUMER UV line emission, as compared to quiet sun levels, in the radio enhanced regions, but local maxima were seen in the ECHs at the approximate locations of the enhancements. H $\alpha$ images showed increased brightness in the ECH correlated with compact radio enhancements and strong unipolar magnetic fields. EUV Fe XII observations showed the presence of equatorial coronal plumes attached to some of the compact radio bright features.

2 Observations