Activities Related to the HEMP ______________________________ The purpose of the HET Echo Mapping Project was to develop and use the ``echo mapping'' technique to investigate the internal structures of active galactic nuclei (AGN). The proposal was submitted while the PI was at the University of Texas, then transfered to San Diego State University. This proposal was awarded $72,275 for 1 year, with multiple no-cost extensions approved. This project involved both extensive observations and modelling, and while the observational aspect was less than expected, overall the project has been very successful. We have so far produced 1 refereed publication, 2 Master's theses, 4 conference proceedings, and 3 AAS presentations with more publications in progress; this grant has also involved 4 graduate and 4 undergraduate students. The observations were made with the new Hobby-Eberly Telescope (HET), but as is well known in the astronomical community, the telescope had serious "birthing" problems. The problems were mainly related to image quality and glitches in the queue scheduling; details are not necessary here, though we do comment that most of the problems are now fixed and the HET will soon reach original design specifications (or better). Our observations were among the first taken by the HET, and they have several peculiarities and problems which make the data calibration tedious and very time-consuming. The data sets we obtained from the HET are not bad, nor are they good - they are "ugly". We had to solve many problems on an ad hoc basis before these data could be used. As examples, flat fields were found to be time-variable, wavelength calibration arc spectra were sometimes inadequate, and calibration spectra were missing from the archive. Nevertheless, we have made considerably progress. We obtained engineering-grade sets of data on the AGN Mrk 817 and Ark 564 in the 1999 season, and on Mrk 110 in the 2000 and 2001 seasons. The signal-to-noise ratios, while less than expected, are nevertheless very good (S/N ~ 100 per pixel) and significantly better than other AGN data sets. The Ark 564 data consist of 22 observations at 12 epochs, and was the first data set to be fully calibrated. For the main Mrk 110 dataset we obtained 82 spectra at 40 epochs spanning 6 months. These data are not equally sampled, and have a 4.4-day mean sampling, but this is easily enough to resolve the ~23 day Hbeta lag. The time-consuming data calibration issues have been the main cause of delay in this project, but mother Nature has not been particularly kind either: the first thorough investigation we embarked upon (Ark 564) turned out to give us a major surprise: using simultaneous RXTE, HST and ground-based data, we learned that despite very large and rapid variations in the X-rays, the UV and optical variations were extremely small. This lack of variability of course meant that we could not carry out echo mapping on this target. Nevertheless, the work resulted in an excellent Master's dissertation and presentations at conferences by Welsh's graduate student Luigi Gallo. Gallo received major support from this grant. An initial quick-look at the Mrk 817 dataset also showed a disappointing low amplitude variability. These observations had to be re-reduced several times because of calibration peculiarities, and near the end of the last iteration, the graduate student working with Welsh (Robert Reaves) took a leave of absence (due to a death in his family) and has not yet completed the reduction, though he does intend to finish the work. Reaves was supported with funds from this grant. HET data on Mrk 110 has resulted in several publications, but the second and larger set of data has not been reduced. Both data sets suffer from severe blending of the AGN and a nearby star, due to the poor HET image quality, and this limits to quality of the observations. The observations will be used, but currently the work is on hold because progress along the modelling aspect of the project are much more fruitful (details below). Along with the HET observations, CCD photometry was collected with the 36" telescope at McDonald Observatory and with the 40" telescope at Mt. Laguna Observatory. These data sets, part of a long-term ongoing QSO Monitoring program, will be used to find the most optically variabile high-redshift QSOs (z=2.3-2.8) for spectroscopic echo mapping investigation with the HET. In addition, when the HET was observing, these telescope observed the same targets and these quasi-simultantous photometric points will help fill in gaps in the spectrsocopic coverage. Several students have been involved in this part of the project: graduate students Luigi Gallo and Danielle Martino, and undergraduates Patrick Young and Margaret Moerchen (Univ. of Texas) and Crystal Hart and P. Travis Willingham (San Diego State). In addition to the observations, we have been working on improving the techniques to measure time lags in poorly sampled AGN light curves. A code called ``flicken'' has been written that does a Monte Carlo random walk interpolation to patch gaps in light curves. The resulting ensemble of CCFs that are generated via this method provide uncertainties on the lags, and more importantly, remove the sensitivity of the solution to specifics of the particular interpolation method (e.g. linear interpolation). The first version of the code is complete and have been used to verify past results on Mrk 110. The code is still under development, and Welsh's graduate student Jessica Castora, is currently working on improving flicken by adding aspects of the interpolation method developed by Gaskell and Peterson. Danniel Martino, Welsh's grad student, used flicken to analyze archival data on NGC 5548 (from Wanders and Peterson). This investigation led to her Master's thesis and MS degree (awared in the Summer of 2004) and a paper is in progress.