Domain review edits

Addressing requested changes in #17 and #23

Required changes

02_visualize_massVflow.qmd

• In the `processing` chunk it says “Calculate average stream discharge” but it is median discharge
  • Renamed these objects and documentation to "median"
• Are there multiple mass measurements for the same stream comid streamflow values? Is this pseudoreplication? The data are spread across numerous sampling locations and periods, but then aggregated to a single value which is making me think twice about this.
  • The processing and analysis steps I followed here produces the same results as published in the manuscript. (Albeit I used the 95% percentile line instead of the 90% line. I used 95% line because it's more familiar of a threshold for most public audiences (e.g., associated with "significance").) The dots on the chart are the same as shown in the manuscript. I am not sure if they used pseudoreplication or not, but I am following the peer-reviewed, published methodology as stated and shown in the manuscript (Figure 2):

Suggested changes

README

• In the README add a version for sbtools rather than “the latest version”

All .qmds

• Move library(geomtextpath) to the top of the qmd scripts

02_visualize_massVflow.qmd

• Where does in/sciencebase/extras.csv come from? The comment says” These are from the Miller et al., 2018 data release” but it’s not downloaded in the 01_fetch_data.qmd. I am interpreting that this was done to avoid a massive download from ScienceBase. Do you know which of the child items contain these comids to be able to direct a user to it?
  • Correct, the science base item has many extremely large zip files that each contain thousands of CSVs, each for a different COMID. The zipped files are organized by Level 4 ecoregion, so that's how those individual comids could be tracked down. I added this information to the description of the data to help clarify how the zip files are organized.
• Site `Willow-108` has an NA length, but since it’s the only data point in that group it is throwing an error instead of filtering out with the na.rm=TRUE. Suggest filtering out NA Lengths first to avoid this warning, and a -INF that gets carried downstream
• I was a little confused about the ordering of steps starting on line 100 the site and fish data are joined before the fish data are filtered and MaxFishLength is calculated. I would probably join the data right before joining with the streamflow data. Regardless, it is inconsequential to the data processing and produces the same results.
  • Agree, this was a confusing set up. I reorganized this workflow slightly to make it easier to follow and this reduces the many-to-many warning that had been happening when data were merged before filtering.