Helmet Orchid Pollinator Study

Tell us about your projects. Update us regularly.
parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Helmet Orchid Pollinator Study

Post by parkview » Sun Aug 17, 2014 9:28 am

Apparently no one really knows what pollinates the native Australian Helmet orchid, Corybas species. July-August is the main flowering season for the plain Helmet orchid - Corybas recurvus, that lives around Busselton.

So a Perth guy and myself have set out on a multi year journey to see if we can figure this out. Here is a photo of the target orchid:

Helmet-orchids.jpg
Helmet-orchids.jpg (26.07 KiB) Viewed 3897 times


The above flower is around 10mm wide and the leaf can range from 8 to 20mm wide, so it is kind of small. Therefore, the pollinator must be small as well. It also has to be able crawl down inside the flower, as the anther that holds the pollen is down inside at the very bottom of the base of the flower. Orchid Pollen is not free flowing, and is stored in sticky, waxy sacks called pollinia. The insects brushes up against a sack, and it will be transferred and stuck onto the insect.

[insert Corybas cut-away drawing here]

It's suspected that a Fungus Gnat (family: Mycetophilidae - http://en.wikipedia.org/wiki/Fungus_gnat) is the main pollinator of the orchid.
Last edited by parkview on Sun Aug 17, 2014 10:49 am, edited 1 time in total.

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study -II

Post by parkview » Sun Aug 17, 2014 10:26 am

This years plan was to trial a few kinds of different traps, therefore seeing what might work best. They where placed at two locations, some 400m apart. At each location, we placed out traps at various location to see what might work out best. We put them out on a Sat. afternoon and checked the traps on Sunday morning and again on Sunday afternoon, recording what each trap had found. We left them onsite for the week, and I collected them the following Sunday. We placed three kinds of insect traps over Helmet orchids, Mosquito orchids and some controls sites (over non-flowering leaves or dead leaves):

1) clear plastic drink bottle with top inverted and bottom cut off:
botom-less trap.jpg
Plastic drink bottle - bottomless trap
botom-less trap.jpg (77.88 KiB) Viewed 3895 times

I didn't think this one through very well, as how was I going to collect the insects. If they died, they would fall onto the copious leaf litter. I had based this trap on my RPi based orchid photography rig. Surprisingly, out of four traps, I did trap one non-gnat insect. No pollinia found

2) clear plastic drink bottle with top inverted, raised up over site:
rasied-trap.jpg
drink bottle - raised trap
rasied-trap.jpg (110.41 KiB) Viewed 3895 times

This did work better, but out of 6 traps, we caught 5 insects, mostly gnats. No pollinia found

3) Tanglefoot [http://www.entosupplies.com.au/equipment/field/traps/tanglefoot-_-tangle_trap-insect-trap-coating] painted onto clear plastic sheets and set on wire like a miniature flag
tanglefoot-trap1.jpg
Tanglefoot trap
tanglefoot-trap1.jpg (122.69 KiB) Viewed 3895 times

Most traps where placed vertically, but I did put three out horizontally above the flowering orchids. These traps where very successful and caught 331 insects, two spiders and 2 tick looking things, 4 small 1mm long beetles. Importantly, it caught 4 fungus gnats that had pollinia sacks on their backs!

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study - III

Post by parkview » Sun Aug 17, 2014 10:45 am

So, here are some not so great photos of our trophy catches. The free version of phpBB only allows 3 photos:

Site One:
only one found and it was found on the first Sunday visit, so we where pretty stocked.
polinator1_sml.jpg
polinator1_sml.jpg (40.86 KiB) Viewed 3895 times


Site Two:
the rest where found here. One trap had 50 gnats on it, two of which had polinia sacks!
Polinator6.jpg
Polinator6.jpg (28.36 KiB) Viewed 3895 times

Polinator8_sml.jpg
Polinator8_sml.jpg (22.87 KiB) Viewed 3895 times


So, how could I fit a RPi into this kind of project? Next year I would like to put a battery powered RPi out there to record light levels, air and ground temperature, rain and try and time lapse photograph them growing. It would have to be a pretty water proof rig though.

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study - 2015

Post by parkview » Wed Jul 15, 2015 10:05 pm

Well, this year season is ramping up already. I am no where near ready for it as I should be.

Last year I wrapped the Tanglefoot insect traps in glad wrap, so they could be transported to Perth for further analysis. That lead to a sticky mess when we tried to unwrap them. This year I have quickly designed up a 3D printed box to hold the 59mm x 82mm clear plastic sheets. This version can only hold 13 sheets, but it could be modified to hold what ever size sheet we want to use. Such are the ways of 3D printing :D

Here is the design:
Insect trap holder - design.jpg
Insect trap holder - design.jpg (43.2 KiB) Viewed 3888 times


and the finished print out. I have some sample white paper sheets in there to simulate the clear plastic sheets. There is one unused plastic sheet in there as well:
Insect trap holder.jpg
Insect trap holder.jpg (34.63 KiB) Viewed 3888 times


The Helmet Orchids are already flowering, so I hope to test these out over the next few weekends.

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Sun Aug 16, 2015 11:17 pm

It's hard photographing small Helmet Orchids in winter. July through August is their peak flowering time. When you have your head stuck in a sedge bush, under some Peppi trees, there usually isn't much light around for a good photo. My elcheapo Canon camera's flash only starts working out at 50cm from the target. I want to get right in there, so i am only 1-2cm from the flower, so I need more light.

I had an epiphany a few weeks ago, and designed up a small 6cm dia. ring that will hold two rounds of 12V waterproof flexible LED string. The first iteration was correct, but in practice too difficult to get the LED string wrapped around. Yesterday I was out there again - late in the evening, which prompted me to get on with Mark II:

3D Design LED ring.jpg
3D Design LED ring.jpg (29.63 KiB) Viewed 3885 times


and here it is in real life:

3D Printed LED ring.jpg
3D Printed LED ring.jpg (31.51 KiB) Viewed 3885 times


Simplify has auto added in the supports, some of which has merged and protruded through to the inside of the ring. I can live with that for the moment.

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Sun Aug 16, 2015 11:23 pm

Oh, so how well does this work?

yes, it's ok:

LED ring.1.jpg
LED ring.1.jpg (50.46 KiB) Viewed 3885 times


well that doesn't look that bright does it? Lets just pull the camera back a bit:

LED ring.2.jpg
LED ring.2.jpg (31.62 KiB) Viewed 3885 times


Yep, that's bright enough :)

I picked up some ring LED's in Shenzhen. This one eventually will have a 3D printed mount that will fit onto my camera. I wish I had picked up the other larger rings as well.

LED Ring.3.jpg
LED Ring.3.jpg (36.58 KiB) Viewed 3885 times

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Sat Aug 29, 2015 2:02 pm

Nothing like a nice winters day to find time to blog!

I went out last Saturday and tested out the two LED camera lights. Here is the end result after it was printed and LED fitted to the adaptor.

printed adaptor.jpg
printed adaptor.jpg (18.3 KiB) Viewed 3871 times


Here is above camera adaptor photo of a Sandhill Orchid - Corybas despectans:
example_orchid.1.jpg
example_orchid.1.jpg (63.09 KiB) Viewed 3871 times


This works really well and I am very happy with it. I can photograph objects much closer than I can with a flash, and it doesn't have the sharp edges that a flash does. I wonder how long it will be before cameras are sold with them build in.

Here is a photo of the ring LED in use:
example_orchid.2.jpg
example_orchid.2.jpg (94.9 KiB) Viewed 3871 times


This ring could have a much wider aperture. maybe something twice as wide might work better? It can see a specific use in that it would allow me to light up a flower 360 degrees.

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Mon Aug 31, 2015 9:58 pm

Of course I now wish I had picked up a greater range of LED's from Shenzhen. Here is a sample of the different shapes and sizes that one (out of hundreds of shop in one building) sold:

LED Display.jpg
LED Display.jpg (71.5 KiB) Viewed 3870 times

User avatar
Jubbp
Master Maker
Posts: 209
Joined: Sun Jun 22, 2014 8:15 pm
Location: Bunbury WA
Contact:

Re: Helmet Orchid Pollinator Study

Post by Jubbp » Thu Sep 03, 2015 10:32 am

It's great to see you using your printer for more than making toys. One day I hope mine might be used for something useful. ;)

Now all you need to do is manage the power usage and reduce the battery size so you can make into a self contained unit.. :)

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Mon Apr 11, 2016 9:39 pm

A goal for the 2016 Orchid Pollinator study is to have a go at photographing a helmet orchid flower for 24 hours. Since this will be in July, it means the RPi rig will have to be insect and waterproof. I have started on building my first go at such a housing.

Here is a laser cut acrylic waterproof housing for the RPi IR camera:
laser cut acrylic RPi waterproof case.2.jpg
laser cut acrylic RPi waterproof case.2.jpg (227.72 KiB) Viewed 3861 times


Acrylic seems to have a bit more kerf than MDF, so cut parts tend to have a bit of slop to them. I still am waiting on the 'microscope slip' that I am going to use to cover over the lens hole.

I don't have any black acrylic yet, but I see that Bunnings carries a small range of expensive-ish acrylic sheets. maybe I might have a go at spray painting the box black, but I think it does look pretty cool housing as it is.

User avatar
BeJay
Maker
Posts: 139
Joined: Mon Jun 23, 2014 6:31 pm
Location: Perth
Contact:

Re: Helmet Orchid Pollinator Study

Post by BeJay » Fri Apr 22, 2016 8:59 pm

Wow! Paul you have been busy. Can't wait to see how this goes. Did you end up getting a laser cutter?

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Fri Apr 22, 2016 9:33 pm

Yes, I picked up an eBay 60W Chinese laser, with a cutting bed size of; 500mm x 700mm. I will post up a review of it one day, but having lots of fun with it it!

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Sun Jul 10, 2016 11:13 am

Time to get a move on with this new build, as the Helmet orchids finish up in August.

Last month I settled on breadboarding everything, but the breadboard couldn't handle the current needed to power the RPi, so I ended up transferring it all onto a Vero board:

breadboard-sml.jpg
breadboard-sml.jpg (79.7 KiB) Viewed 3877 times


it now all looks much neater:
vero-board.jpg
vero-board.jpg (38.68 KiB) Viewed 3877 times


I laser cut a 3mm MDF board to hold everything in place so that it fits into a Bunnings 13L plastic tub:

photography-rig1.jpg
photography-rig1.jpg (227.45 KiB) Viewed 3877 times


I also designed up a sun/rain shade for the camera aperture. I am not sure how this is going to go, as I suspect that spiders will love the overhang to build a web on:
camera-sunshade.jpg
camera-sunshade.jpg (61.83 KiB) Viewed 3877 times


I hope to run a test trial tomorrow...

User avatar
Jubbp
Master Maker
Posts: 209
Joined: Sun Jun 22, 2014 8:15 pm
Location: Bunbury WA
Contact:

Re: Helmet Orchid Pollinator Study

Post by Jubbp » Tue Jul 12, 2016 2:18 pm

looking good Paul.

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Sun Jul 17, 2016 10:21 pm

Today was Pollinator-Pi's first real day of action out in the field. I have been visiting and recording flower stats at a local site for three years, usually once a week. There are eight spots ranging over 150m that I record each week from early July to late August and this time period covers the flowering of two species of Helmet orchids.

During the week I quickly made up a board with some low voltage drop Schottky diodes. This allows me to plug one or two 12V SLA 7.2Ahr batteries in, and swap them out in the field if needed. It also has some jacks for power future items like a USB hub, and my HDMI-Pi 9" display out in the field:
battery distribution board_sml.jpg
battery distribution board_sml.jpg (56.16 KiB) Viewed 3871 times


to prevent the Vero board copper strips from shorting out on some thing in the plastic tub (I haven't cut a mounting site on the spacer board), I laser cut a matching piece of flexible plastic from a yoghurt tub lid. This worked very well and was fast to design and cut out.

Here is the field setup:
getting ready to go_sml.jpg
getting ready to go_sml.jpg (105.6 KiB) Viewed 3871 times


And from the other side:
camera in action.2_sml.jpg
camera in action.2_sml.jpg (63.81 KiB) Viewed 3871 times


The camera was maybe 10mm to far away from the flower, but due to the good depth of field of the RPi IR camera, it's kind of ok. Ideally the camera should be 70mm-80mm from the flower. Note: this is a IR camera, so normal colours are all washed out, and in fact, when the IR flood light works, the photo is even more washed out, into just black/white:
2016-07-17_114746.034146_sml.jpg
2016-07-17_114746.034146_sml.jpg (66.9 KiB) Viewed 3871 times


Graphs:

When the system powers up, a once a minute crontab job kicks in and a python script records the battery voltage, current, watts as well as, air and CPU temperatures, into a csv file. Here are a few quick Excel graphs for the 38 minutes of run time:

I started on one slightly used battery, then at the 8 minute mark, I plugged in the higher charged HDMI-Pi battery:
Battery_voltage.jpg
Battery_voltage.jpg (14.13 KiB) Viewed 3871 times


The DS18B20 temperature probe was balanced on a few twigs, about 30mm from the ground. I could make up a special stand for it if I wanted too.
Air_Temp.jpg
Air_Temp.jpg (15.4 KiB) Viewed 3871 times


You can see the internal box temperature slowing rising, while the photos are being taken, the temperature drops. At least it's winter.
cpu_temp.jpg
cpu_temp.jpg (12.87 KiB) Viewed 3871 times


Results:
  • I only captured 1238 photos. The file system filled up! I had a quick look out in the field, but I think it was adding the 32GB USB flash drive used space onto the RPi SD card space and deciding that it was full, when in reality there was lots of file space available. I will have a look at that issue tomorrow night.
  • Looking at the first lot of photos, initially I thought this was going to be boring, but if I quickly flip through the photos, there is a LOT of action going on. Ants, spiders, mites and other bugs all scurrying around. Most fascinating and exciting. Unfortunately, anything not in the immediate vicinity of the Helmet orchid, is out of focus, but I am hoping the software will be able to pick up the movement in the series of photos. At least now
  • I have some photos, so I can start on the next phase of building a bug detection system. I haven't give this much thought yet, as I was focusing on the fun hardware side of the project. I will look at OpenCV, Processing and Numpy to see what is easiest/best/fastest.

Things to do next time:
  • take a ruler, or a specially made/cut depth of field measuring device (camera to flower)
  • take some electrical tape. The IR flood light didn't light up due to the built in LDR and I wasn't going to sit there holding my finger over the LDR.
  • take some mosquito repellent
  • do more home testing. I knew I should have done this first, but the day was getting away from me, and I really wanted some real live action shots, not back yard shots looking at a boring plant.
  • run some more tests with the IR flood light, to work out the optimal distance from an object.
  • build a stand for the IR flood light. I will pick up another cheep dx.com one, but that will take 4-6 weeks to arrive.
  • If I really get into this, I could add another waterproof power jack and HDMI socket onto the wall of the plastic tub, then I wouldn't have to open the box to plug everything in

User avatar
Jubbp
Master Maker
Posts: 209
Joined: Sun Jun 22, 2014 8:15 pm
Location: Bunbury WA
Contact:

Re: Helmet Orchid Pollinator Study

Post by Jubbp » Tue Jul 19, 2016 7:01 pm

Great write up Paul, it looks like you're coming along nicely

Might need to work on some shade for a summer deployment.

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Wed Jul 20, 2016 9:40 pm

Yes, that could be an issue, although the main purpose was to study these winter flowers. I will eventually use to monitor other things too, but that's why I thought I would record the CPU temperature. It's a nice easy free one to record.

User avatar
BeJay
Maker
Posts: 139
Joined: Mon Jun 23, 2014 6:31 pm
Location: Perth
Contact:

Re: Helmet Orchid Pollinator Study

Post by BeJay » Thu Jul 21, 2016 10:19 pm

Nice work Paul :)

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Sat Jul 23, 2016 8:51 am

It's been a productive week. I fixed the 32GB USB Flash drive mounting issue, but after running some tests, I have realised that at full 5MP resolution, it's only going to be able to store around 13K photos, or only 3.6 hours worth. I have ordered a 64GB flash drive and lowered the resolution down to 1920 x 1080 pixels.

Thanks to Jaycar, I have a new HDMI-Pi power cable using a 4mm x 1.7mm DC power solder-able connector.

parkview
Master Maker
Posts: 445
Joined: Tue Jun 24, 2014 8:25 pm
Location: Busselton
Contact:

Re: Helmet Orchid Pollinator Study

Post by parkview » Sat Jul 23, 2016 11:01 pm

After a few morning tests, I thought it might be handy to add in a HDMI and USB socket pass-through on the plastic tub wall. This way I don't need to open the tub up at all when I am out in the bush. The USB socket is IP67 waterproof rated. The HDMI socket is really just taken off a wall plate mount I found in Shenzhen. It comes with a plastic plug, so as long as I am not using it in a deluge, it should be ok. The tub it's self will be wrapped in a green plastic wrapping, to help keep prying eyes from finding it, so it will help with the waterproofing.

HDMI Socket pass-through (inside tub view):
HDMI_passthrough.jpg
HDMI_passthrough.jpg (21.08 KiB) Viewed 3853 times


USB waterproof socket on the LH side:
USB waterproof socket.jpg
USB waterproof socket.jpg (52.05 KiB) Viewed 3853 times


100% dark room test with IR flood light:
camera caught in action.jpg
camera caught in action.jpg (31.21 KiB) Viewed 3853 times


The IR flood light tripod popped in one of my RSS feeds yesterday, so I printed it out this morning: http://www.thingiverse.com/thing:1460181 easy to print and right what I needed:

IR Flood light tripod.jpg
IR Flood light tripod.jpg (61.98 KiB) Viewed 3853 times


And the action shot:
2016-07-23_223302.857842.jpg
2016-07-23_223302.857842.jpg (40.83 KiB) Viewed 3853 times


The camera is set to around 70mm focus depth, so this evening I cut out a 75mm MDF ruler

Post Reply