Introduction to precip

This document outlines the procedures used for Quality Assurance (QA) of precipitation data at meteorological stations at H.J. Andrews Experimental Forest (HJA) which hopes to reduce or explain common sources of error and uncertainty. Collecting high quality precipitation data has many challenges. Measuring precipitation is more complicated than simply putting out a thermometer with something to record it. Systems must be created to funnel precip to a measuring mechanism that can log incoming volume and convert volume to depth; e.g. the tank depth increased by 0.2635 mm since the last measurement or metering funnel output directly in fixed amounts.

This becomes even more complex in winter when measurement mechanisms can freeze, or the precipitation must be melted to flow through the funnel, leading many, such as the Forest Service RAWS (Remote Automated Weather Stations), to stop collecting usable data altogether in winter months. It causes others, such as USGS/NRCS’s SNOTEL (Snow Telemetry) stations, to accept lower precision measurements. Heating requires large amounts of energy, whether electric or gas, and greatly increases maintenance, all of which is difficult in remote mountainous locations.

All of this leads to a complicated dataset that has many false signals that require careful quality assurance procedures. The quality assurance process used on rain gauges at HJA is outlined below. The document is split into three parts:

  1. Testing and Development of QA Methods- This part contains the development of each QA procedure. Each section of this part strats with a summary of the QA developed, and how and why it is applied. Anyone wishing for more explanation about how a procedure works, or why it is necessary can look at the rest of a section. They are publications of Jupyter Labs and show all the work done to arrive at a final procedure. The raw Jupyter Lab files can be downloaded and run from the /labs directory. These development labs sometimes use QA functions from the package below, but are not updated as the pacakge is updated which may lead labs to break when run.

  2. Probe QAQC Parameters- This part contains all of the QA procedures applied to each probe and their parameters, including manual flags. Since many different types of measurement devices are used, not all probes use all procedures and they may parameterize the same rule differently to achieve good results. See Testing and Development if you want to know why.

  3. Python Package- This part contains the source code for the actual rules.