The People's Pluviograph!

[Greg Hall]

Record-breaking floods in the Northern Rivers regions of the state of New South Wales in Australia last year has set us looking for a way to report hourly, or more frequent, rainfall intensity, to better inform those downstream.

As the official gauge network is sparse and flaky, our present 'solution' is a few human Rain Sentinels' who live high up in the catchment. They agree to read their gauges hourly and get the readings out, by phone as long as that lasts (not very), then CB radio.

We haven't had much success with off-the-shelf weather stations, but maybe there are WeeWx gurus who would know how to help script such a reporting function.

I've bought a couple of 8" tipping bucket gauges, and am playing with Arduino code to keep daily, tips per minute files, on a local SD card, to be relayed on to a LORA-APRS system (o.n.o), and as last resort, on a local display.

Any guidance or assistance very welcome.

Greg

VK2GTH

-

[michael.k...@gmx.at]

Hi Greg,

I personally would try to get everything done with a solution that's as close to off-the-shelf as possible. WeeWX certainly can help you getting data from the off-the-shelf devices and relay them in any other form. What are the real issues in such events? The amount of rain, or the river stage? Yes, they are connected, but two years ago, a couple of villages just a couple of kilometers from my station, were heavily flooded, my station reported much rain, but just a few km away the amount of rainfall was more than triple of the amount that fell at my location. So even with a rather dense network of stations, there is a good chance to miss a huge part of what is happening. Wouldn't it be more significant to measure the levels of rivers and creeks upstream?

Are power and communication infrastructure outages, during such events, an issue?

[matthew wall]

On Monday, November 6, 2023 at 7:03:06 AM UTC-5 greg...@brushbox.info wrote:

Any guidance or assistance very welcome.

greg,

i know that your request is regarding the supply side of things - measuring rainfall.  but here are a few projects, vendors, and organizations i know of who are working on the other end - measuring the result of that rainfall.

there is a group called 'floodnet' in new york city using lorawan to measure local flooding, using maxbotix ultrasonic sensors and a loranwan network. they are pretty open about how they build their sensors and collect the data.

https://github.com/floodnet-nyc/flood-sensor

the aprs system has a mechanism for reporting water height, but it is rather anemic.  we're hoping to expand that...

http://aprs.org/flood-gauges.html

http://www.aprs.org/doc/APRS101.PDF

in september of this year, neracoos sponsored the first (us east coast dominated) water monitoring conference.  that was primarily for tide monitoring, but the usgs and some hardware vendors were present to discuss issues surrounding event-based monitoring.  the applications include hurricanes and the rainfall+stormsurge,  construction projects, environmental impact studies, aquaculture, coastal policy-making, etc.

there are a *lot* of people doing this kind of work, but here is a short list:

- noaa - united states national oceanic and atmospheric administration - https://www.noaa.gov/

- neracoos - https://www.neracoos.org/index.html

- gmri - gulf of maine research institute - https://www.gmri.org/

- usgs - united states geological services - https://www.usgs.gov/

- hohonu - they make water level sensors - https://www.hohonu.io/

- maxbotix - the go-to for ultrasonic sensors - https://maxbotix.com/

- rak wireless - loranwan hardware when you don't have time to build it yourself - https://www.rakwireless.com/en-us/technology/lorawan

- obscape - environment monitoring - https://obscape.com/site/

there are many more vendor sites, as well as companies who specialize in designing, building, and deploying these things.  it is not easy to deploy sensor networks that will survive weather, critters, vandalism, etc.

a few of us in mid-coast maine are working on community-oriented systems to measure tide and estuary levels. we have been focused on tide and rainfall monitors, located in areas subject to flooding and storm surges. but the intent is to add temperatures, salinity, ph, and perhaps a few other metrics that will help those working in coastal aquaculture.  we have been experimenting with commercial offerings including hohonu, divirod, onset, and obscape, and also diy systems using weewx, lorawan, arduino, maxbotix and some radar-based sensors, typically in partnerships with local schools/university, and the local town offices and public works.

https://belfasttide.org/

http://ferry.vinalhavenweather.com/

in particular, the vinalhaven tide monitoring project is replacing the human-with-a-stick system that has been in use for years.  during extreme high (and low) tide events, people go to their tide station to measure the water level.  but that is difficult/annoying when high tide is at 02:00, or in the middle of a massive storm with huge surges.  so we're trying to deploy low-cost, loranwan ultrasonics that might not be noaa-quality, but are good enough to track the flooding and surge that will improve forecasts and help first responders and public policy.

hope this helps,

m

[Graham Eddy]

i posted details of my implementation of lorawan river level sensors reporting via weewx on this forum about a year (or two?) ago. they are anxiously watched by community when we have severe rain events in the region (we flood fairly often - you learn to deal with weather extremes)

ditto re measuring the river level not the local rain - flooding can be caused by rain way up-river when there is no local rain at all

⊣GE⊢

[Ton vanN]

LoraWAN certainly is a candidate for an airborne link from a sensor to a local gateway_to_internet over better distance than with the usual meteo-sensors' communication lines.
SIGFox is a comparable solution for airborne communication over_between sensor and gateway-server.

Drawback for both is that they need internet for the next communicationsegment from the gateway to the world,
and in flooded areas that might be a practical handicap, because usually cables running from the gateway to the next service-contact.

Another option is to apply GSM, best 2G-version (if still available), because that has longest linking range.
With a compatible Mifi-router it is possible to link the sensor airborne over wifi to the mifi-router, with the mifi-router over 2G taking care of the next airborne step of communication.
Mifi-router is a small device usually running on battery, which battery is fed over USB-port, making it semi-independent of power-grid etc.

Probably not the most inovative solution, but might in practise be an airborne bridge over significant distance until arriving at the big internetbackbones.

Op maandag 6 november 2023 om 13:03:06 UTC+1 schreef Greg Hall: