Testing DIY EC sensors in Papawai Stream

It is the first time I install and test the two EC probes at Papawai Stream. I am close to where the wastewater pipe is crossing the stream. I have decided to audio record myself talking along with my fieldwork for better documentation of my process and challenges in the field.

moving The kit from lab to field

I have put together a kit that is transportable by one person.

The crates make it easy to transport the probes and could also play in the role when installing the probes to keep the electronics in place and safely stored outside of the water during use. Additionally, I am carrying a backpack with my Thinkpad laptop, and a bag with additional equipment such as the TDS reader, batteries and USB cables and other necessities such as extra jars, paper towels, cable ties, pens, and a notebook. I also take a little camping stool and my smartphone with me. I transport the equipment by car to the Prince of Wales Park and further carry it from the carpark to the install site.

After a small hike up the council walkway, I reach the entrance to the site where I plan to install my work, just on the left before the bridge. The narrow walking path to access Papawai stream is covered in pine needles, lined by grass, bushes and pine trees.

Path to access Papawai Upstream. Brown narrow footpath in grass lined by bushes and pine trees.

I follow the stream upstream and reach the site that I have scouted earlier as a possible install location. The stream flows on the right side of the walkway. I walk one minute to reach the install location further upstream where the path meets the water. I have placed the probes in reused jars, but I have not waterproofed them yet, so I temporarily propped the jars up with bubble wrap to protect the electronics from humidity and insects while making them easily accessible during the testing.

Papawai test Install site showing crates with probes next to stream
Setting up

The Raspberry Pi is still running, and the Moturoa_Transmissions network is up. I start unrolling the cables and connect the probes one by one to their batteries. I make sure the crate is in a stable position, so it does not create a hazard once I launch the probes into the stream water.

When I attempt to launch the first probe, I notice it does not easily drown. The water is just a couple of centimetres deep.

Looks weird. Two electrical plugs in the water.

I am also testing the updated SD card node. I have updated the code so that a new CSV file is created every time the node is being reset, as opposed to overwriting any old data, which was good enough for earlier experiments, but now I need a more stable way of collecting and keeping data. I chose not to thoroughly test the SD card code in the lab because the main focus of this experiment was to evaluate the functionality of EC probes and I needed to get into the field as soon as possible to avoid having to work after dark.

I notice I haven’t brought a dedicated battery for the temperature probe, so I need to use the spare one from the equipment bag. The temperature probe is still in the plastic bottle casing from the previous install. It could work if I redesigned the cap to aid with accessibility to charge/change the batteries. The probe itself also still has a stone gaffer taped at the bottom of the wire to add weight to the probe for easier launching into less reachable parts of water. The previous install location was wildly overgrown, and I made this with what I had at hand in the field. I could add some pink tape over the black to make the submerged probe more visible as it is difficult to spot under water. The temperature probe got a quite long cable (partly seen on the photo below with red, yellow and black strands) and it already got entangled with a plant around here even though I have just carefully de-tangled the cable before leaving the lab. I am launching the probe into the water close to the already submersed EC probes, but carefully, not to short-circuit the prongs of the other probes.

2 EC probes and temperature probe submersed in stream water

That looks pretty good. I am also taking a panorama of the setup to capture the relationships of the probes with the stream.

Panorama of experiment setup showing a latop on a stool on the left side. Next to the stool are two wooden crates with glass jars containing electronics. Three probes are submersed in the stream visible on the right side. Cables can be seen on the ground.
accessing the sensor data

So now I got the probes in the water, I am moving on to the exciting part of seeing the data, and what values the two EC probes return. It’s quite sunny and dry today, so I thought it would be fine to take my development laptop with me into the field and use it to live-view the data of my devices.

Plants are already losing some bits and pieces on my keyboard.

I spend some time failing to connect to my Moturoa_Transmissions WiFI network  and need to restart my machine. This takes a while and gives me time to look at the probes again. I rearrange them because the cables are overlapping, but find it difficult to place them the way I like because the curl of the cable dictates the placement on the ground. I also notice that the probes disappear when underwater and I could consider adding some elements of color to set them visually apart from the surroundings. While trying to rearrange cables and probes, I notice that it would have helped to bring some gloves since I am not entirely sure if it is safe to touch the water. Additionally, some hand disinfectant as used during the SHMAK workshop would have been good.

My laptop has rebooted, and I finally see my own WiFI network, but I also notice another wireless network named AI-THINKER, which curiously seems to be another ESP8266 based Internet of Things network.

I open a terminal window on my laptop and use the mosquitto_sub command to subscribe to all topics to see all data floating around in my little PubSub network.

This is exciting.

I see EC returning a value of 267 and EC_rua on 276, which appears to be consistent with previous readings, where EC_rua returned a slightly higher value than EC. The readings seem to be remaining constant, which is good.

I don’t see the temperature in the feed, so maybe the battery I used is not working. It does not have an indicator light so I do not know whether it is working or not. I will swap the power supply to the spare solar battery I brought along. This one is clearly discharging according to the LED, so I should see a temperature value soon.

Oh yeah, watertemp here we go. 14.2

Now I will cross-check this result (in degrees Celsius) with the temperature my TDS reader returns. So I am stepping down towards the probes and submerge the prongs in the water, and it seems to be about 1 degree off. I might jot these values down in my notebook just in case.

EC is 265. EC_rua is 275. Watertemp sensor returns 14.2°C, the TDS 15.5°C. The TDS reader shows 158 ppm.

concluding the experiment

I would call this a successful experiment, I have collected 15 minutes of data so far. No one has walked past yet. It worked this way, but the gear obstructs the walking path a little, and I need to find a more stable and safe spot for longer installs that can also be left alone for some time. This was easier at the previous install where most devices were installed behind a fence and the others were safely suspended, with just the probes touching the ground without obstructing anyone’s path, sheltered by large Harakeke plants. The jars work well, so I should continue working with them as material and find a suitable solution to make them both outdoor- and longterm install-proof as well as provide easy access to the electronics for charging and possible hardware updates. The electrical cords work, but the way they are lying on the ground like this looks quite ugly. The white colour of the cable helps to set them apart from the dark ground, but I should consider some more thoughtful decoration or support for the cables. Suspending the electronics on one of the fallen trees, as previously planned, could work, but needs some careful consideration of the environment.

 

Observations and recommendations for next install:

Design development and install

  • Draft some designs, so the probes are stable and visible underwater while not disturbing the natural flow of the stream.
  • Design versions of outdoor-proof but easily accessible lids for glass jars.
  • Overall the electrical cords in this setup are visually not appealing. The white colour sets the cables visually apart from the background, which is good for safety but aesthetically unpleasant.
  • The installation of probes should be tested suspended from trees. This needs additional install tools such as hooks, cable ties and a variety of options should be tested in lab and field.
  • Replicate the probes and connect them within the network. Build three probe pairs temperature & EC and design an output that is accessible from the bridge.

Equipment and electronics

  • Bring all spare batteries just in case.
  • Use a lapel microphone attached to an audio recorder for hands-free recording.

Hygiene, health and safety:

  • Bring some (plastic) gloves to be able to safely touch the water when adjustment of probes in water is required.
  • Add paper tissues and hand sanitizer to the kit if getting in contact with water or dirt.
  • Consider bringing gumboots for installing probes in more inaccessible, boggy areas.
  • To stabilize the setup, I would need to bring something like pegs to fix the cables to the ground.
  • Add some colours to the bottom of the probes or consider using makers to make sensors and cables more visible when submerged in the stream water. Especially the black wire of the temperature probe disappears into the environment.

 

 

Exploring Papawai Stream

This blog post describes my first exploration up Papawai Stream, beyond the easily accessible footpath, on the search for where the water originates.

Unfortunately, the camera I took on the way had no GPS so I cannot accurately pinpoint images from this first walk to a location in retrospect.

It was an early Saturday afternoon, the weather was fair, early spring.
I cross the lower playing field to enter Prince of Wales Park, which is part of the Wellington Green Belt, an outdoor recreational area relatively close to the city centre.

On the sides of the field I can see the stream, surrounded by vegetation, almost hiding, but slowly gurgling away. To follow its course further I take a walking path uphill, entering the shaded, forested part of Prince of Wales Park. The stream is on the right side of the track, not easily accessible, but partly visible through the vegetation.


I arrive at a bridge that crosses the stream. The stream is accessible from here, I would need to duck to walk underneath the bridge, but I choose to walk back up and use the more accessible way on the other side of the bridge.

The bridge would lead me to a paved path uphill, leading up to Dorking Road, Brooklyn. I, however, head south-west, following the stream on an unpaved walkway that at times resembles a small gully.

The stream bed consists of stone and gravel, but I notice some old red bricks, adding colourful highlights to the otherwise merely brown environment.

Walking further upstream, I hear bird calls and creaking of trees in the wind. The stream is still visible and audible on the right side of the walking path, flowing a little lower than the walkway. Fallen trees are obstructing the track further up but can be passed to continue the walk.

The sun is coming through the trees at times and creates nice highlights of the landscape. The path is softly covered with pine needles and lined by green vegetation. I see very tall pine trees and beautiful ponga/tree ferns. The track appears to be getting more rough, narrower and city noises have entirely disappeared from the soundscape. I hear the gurgling of the stream beside me, accompanied by bird song and the sound of the wind brushing through the trees.

Walking further uphill, I notice the stream has suddenly disappeared. I head back down a few metres to reach the spot where the flow has died down and blends quietly into a brown boggy patch of forest ground.

By the looks of the area, it seems this part has higher water flows, perhaps during or after heavy rains. The further uphill I walk the more trash I encounter, mainly bottles and plastic packaging often buried halfway in the ground. It appears the garbage is being moved further down as part of a mudslide.

The territory is getting steeper, and accessibility starts getting to become an issue. I need to watch my steps as the ground is very boggy and slippery at times. I hold onto branches of trees for stability. The look of the territory suggests even stronger that a mudslide has brushed down the area recently, washing down earth and vegetation.

I appear to reach the top of the hill as I can see more of the sky again. The boggy ground is littered with brown bottles and red bricks. I need both my hands to hold onto bushes and trees as I carefully crawl uphill so I cannot use my camera to take more photos.

I see a pipe and what appears to be an old railroad rail, sticking out of the hill. I also see the bottom of a house, and I am not sure if I am on private land. There is no water coming from the pipe. I need to watch my steps, but I manage to reach the top of the hill.

I end up on a street that – as I would later find out – is called Connaught Terrace. I end up coming up and out of the gully just next to the garage of private property. I look back down into the area I just came out. I mostly see grass and trees.

I walk back to the street where I find a drain cover that must be part of the network of the pipe leading down to Papawai Stream. It is marked with a red dot. Leaves partly cover the grill. It appears this drain collects rainwater and runoff from Connaught Terrace, which would end up feeding into Papawai Stream.

I notice an official paved path that ends up approximately 10 metres next to where I exit the reserve and decide to take this way back down to Mount Cook. It appears this is the public path as marked on Google Maps, connecting the lower playing field of the Prince of Wales Park with Connaught Terrace.

The above photo shows the paved walkway downhill, which is very different from the path I chose for my uphill adventure. Papawai Stream must be somewhere in a parallel valley behind the hill on the left side. On the right side, I get a good view through the pine trees to what must be the suburb of Newtown.

I end up walking back down to the fields, following the stream around the bund and snap another photo of it entering the culvert next to the playing field changing sheds.

Further downstream it appears that there has been some significant event recently, causing washouts and uprooting trees.

The stream is flowing through a canyon that appears to be the result of a washout, possibly after heavy rain. The stream bed is shallow. Vegetation has suffered from slides.

I conclude my walk here having explored some of the more inaccessible parts of this stream. Connaught Terrace seems to be the highest point of the stream, but it is not entirely clear to me where the water is originating from. The area just about where the stream is losing flow has been particularly curious, as it appears that people are frequenting this area despite its inaccessibility.