Mobile Sensor Platform?

@ensign has done some work with a moving/drifting sensor stations: https://www.fondriest.com/news/hydrosphere-drifter-brings-lagrangian-sampling-freshwater.htm

The way Monitor My Watershed’s data portal is set up right now is very “site-centric” – the web pages are just not set up to accept moving sensor data like you would generate. The database model which the website is built on can manage trajectory data, but the portal isn’t exposing that portion of the database. To put the data in right now you could pretend it’s all from a single station and then call latitude and longitude two of the your measured variables, but that’s definitely a cludgey way to do it.

For how the trajectory results can work in ODM2, see: https://github.com/ODM2/ODM2/blob/master/doc/ODM2Docs/ext_results_measurementframeworks.md and https://github.com/ODM2/ODM2/blob/master/doc/ODM2Docs/ext_results_trajectory.md

Scott Ensign, formerly of Planktos Instruments, is now at Stroud. (He’s @ensign.)

I don’t know how “unusual” trajectory data is, but, at least in long term stream monitoring, it’s definitely far less common than fixed station monitoring. That type of data is also harder to manage, visualize, or conceptualize. I’d guess if others don’t support trajectory data it’s for those same reasons. Even including just site-specific data our feature wish list is far longer than we have the resources to ever complete.

What were you hoping to accomplish by uploading your data to MonitorMW? If your goal is to have your data easily sharable and searchable, I’d recommend using HydroShare (https://www.hydroshare.org/landingPage/). Hydroshare supports uploading ODM2 databases and netCDF multi-dimensional data. HydroShare is also a DOI provider for permanently tagging specific datasets. If you want to manage your own data, you might look into CZ Manager (https://github.com/ODM2/CZ-Manager), which is based on ODM2 (like MonitorMW) but gives much, much deeper database control. You could also look into the tools provided by groups like NCAR (https://ncar.ucar.edu/) – I don’t have any personal familiarity with any of them though. I don’t actually know of any spiffy online data viewers, to send you to, but I haven’t ever looked either.

I hope one of those is more helpful! I’m sorry MonitorMW isn’t ready for your data yet!

Hi Oscamo. Indeed, I was formerly at Planktos Instruments but now at the Stroud Center. In responding to Sara’s comment about trajectory data being hard to visualize, check out this video showing dissolved oxygen data along 22 miles of the Neuse River in North Carolina.

We’ve been collecting trajectory data with drifter technology here at the Stroud Center; this post doesn’t show data but the video is fun.

Can you share more about the data you are trying to collect and what parameters?

@ensign – yup, it’s a nice video. But… creating a video like that automatically for every set of uploaded trajectory data is just a little bit more difficult than making a simple plot with time on the x axis.

The WinRiverII software for the moving acoustic Doppler current profilers that we use to take discharge measurements has some nice visualizations for the boat tracks and the velocity profile. I think the “stick ship track” plot is a really nice way of cramming multi-dimensional data into a very simple single plot. But even though it’s “trajectory” data, the path of the ADCP is much simpler than what I’m guessing @osca is up to and MonitorMW still couldn’t handle it.

Hm. I’m not quite sure I follow you.

It sounds like you have a sensor package that you deploy at one spot for a while, then you stop it, pack it up, and move it somewhere else. After some time, you move the sensor package back to the original site. Is there ever data being collected while it’s moving, or is the GPS data only being collected to help the technician remember where they were on any given day? If the data is only being collected in discrete chunks while the sensors are stationary, you could create multiple stations for each stopping point and then manually chop your data up by location and upload the chunks. It’s kind-of a pain in the bottom, but if the stations are unrelated except for the equipment you’re using to make the measurements, I think that is the appropriate way to handle the data.

It sounds like what you’re really after is a good LIMS system to help your technicians ties measurements to the correct sites and then visualize them all together on a map. The MontitorMW portal can only really help with the visualization part.