Return to Daily Summaries page.

Introduction

The daily summaries provide an opportunity to visualize a large quantity of engineering and science quality data in a single view. The idea is to be able to see at a glance how the previous daytime ops went and what sort of environment occurred during the night. This will allow us to pull out correlations between various parameters and to see relationships between events that happened during observing. The summaries will be produced at 7:15 AM civil time every morning via automated software. This is always after noon UT. The timespan of the summaries runs from noon to noon UT. This is generally sunrise to sunrise at McDonald. Thus the first half of the plot covers the daytime and the second half covers the night time.

Upper Chart

The top plot covers events that happened during the time period. Particular events include shutter open/close, stacks and science observations on the sky. In addition, the Dimm seeing data will be plotted. The horizontal scale is in hours. The vertical scale is in arcseconds and the Dimm data is plotted as FWHM at HET zenith. Shutter open/close are given as green lines close to the top of the plot. Stacks are yellow? lines that begin at the stack start time and the post FWHM of the previous stack on Mars. They end at the pre-FWHM when the stack is finished. Thus you can immediately see how long and how good the stacks were. Hrs, Mrs, and Lrs observations on the sky are plotted as straight lines from start time to end time. The vertical lines at the start and end have the open ends at the Dimm FWHM and the closed end at the instrument FWHM.

Dimm data is collected from the Dimm archive through a web interface program. The shutter open/close times and stack data are obtained from the TO night report. Instrument observations and seeing data are obtained from the RA night report. Additional events that could be added are shutter/louver open/close during the day, air conditioning status, etc.

Middle Chart

The middle plot covers various environmental data. In particular we include the wind speed and direction, relative humidity and dust counts. The horizontal scale is in hours to match the upper plot. The left hand vertical scale is in degrees of azimuth in a north based frame. This scale should be used for wind direction. The right hand vertical scale is nominally 0-50 mph and is used for wind speed. Note that 50 mph is the closure criteria for McDonald. The relative humidity scale is 0-100 and is not marked but note that it is simply twice the wind speed scale. The closure criteria for humidity is 90% so the upper end of the plot is again the closure criteria. The dust scale is 0-100000 particles per cubic foot. Again this is not marked but just like the relative humidity the upper limit represents the closure criteria and the values are easily scales from the wind speed scale.

Wind speed, direction, relative humidity, and dust levels are obtained from the Mt. Fowlkes weather archives. Additional data that could be included would be telescope and dome azimuths, or wind bearing on the structure, that is relative to the telescope rather than true north.

Lower Chart

The lower plot shows all temperature data. Included are the outside air temperature, the outside dew point, the average truss temperature, the interior dewpoint, enclosure air temperature, the stratification index and the zero stratification line. The horizontal scale is in hours to match the previous two plots. The vertical scale is adjusted to cover the minimum to maximum temperatures recorded over the last 24 hours. Therefore this plot will look similar from plot to plot. Care must be taken when comparing different summaries.

Outside air temperature is obtained from the Mt Fowlkes weather archive, the outside dew point is obtained from the Mt. Locke weather archive. Average truss temperatures and the enclosure air temperatures are taken from the PMC temperatures archives. The enclosure air temperature is taken within the truss structure. The stratification index is defined as the difference between the top and bottom sensors in the ladder temp system. These data and the zero stratification are scaled to the mean enclosure air temperature so that they fit on the plot and can be easily compared to the truss and air temps. Additional data that could be included included the Spectrograph room temperature as well as instrument room temperatures.

Problem, Bugs, and Future Plans

The Daily Summary is still in the very early stages of development and there are a number of problems as well bugs. Many of the data sets are still missing and additional data sets are being thought of faster than the software can be written.

Bugs

• The code that collects data from the weather server does not handle missing data. Large data values are generated if the software reads +-----E.-- in the data stream. Need a method of generating multiple curves if large chunks of data are missing as well as a method for interpolating around small gaps.
• The Dust Monitor at HET has communications problems and is not currently running. This needs to be fixed and the dust value compared to the outside dust value.
• Update example summary when we have a plot with additional data and events.
• Need to define generic graph lists and event lists. All data collection software will write the generic lists; the plotting softwae will read generic lists and generate specific plot commands for the plotting software. This allows us to isolate the plotting routines from the data collection routines.
• Need to define generic event logger to collect all events. The file can then be parsed to collect interesting events for plotting. Consider some variant of syslog for this software.

Additional Data Sets

• Add Dimm data. Talk with Chris Wilkenson about this.
• Add software to parse RA/TO report.
• Add sunrise/sunset times to event chart.
• Add Air Conditioning events; on/off, setpoint changes, etc.
• Add Dome open/close, vents open/close? louvers open/close.
• Other events accessable with event logger??

Example Summary

Return to Daily Summaries page.