Breeding with Simulated Solar Cycles

Breeder's Lab
#1

Has anyone tried doing this before?

One potential benefit I can see is that you can easily determine when plants will finish outside at your latitude. It’s hard to know exactly when some plants will start to flower – it’s not exactly at 12/12. Especially with “semi-autoflowering” plants, you might find varieties that will start flowering earlier in the season, like when daylight is 14 hours. It would be hard to determine when plants actually start flowering without a proper simulated # hours per day, and hard to identify the so called semi-autoflowering trait.

Using a real light cycle targeted for a specific latitude, you would be tracking sun hours in your indoor environment. Therefore, you would know pretty accurately which plants will likely finish by first frost.

One downside is that you don’t know exactly when your plants will start flowering. So you might end up vegging excessively if you don’t start from a reasonable date.

10 Likes
#2

I do this for my freshwater planted fish tanks :rofl:

Over there it’s real easy to get a light or a controller for a light that will do sunrise/sunset cycles. Even increase/dim the light per time of day, and even dim for simulated cloud cover and whatnot too. Makes a big difference on the plants and algae growth in the tank.

I would think the big reason for not seeing it as prevalent in herb growing is, we’re always trying to KISS and min/max the grow. If you’re getting 8 or less hours of light in there, you WILL sacrifice yield to do it. We see a significant difference with Auto’s grown in 12/12 compared to 24/0. Now maybe this will give us a better quality herb since the plant is in a much more natural environment, maybe we’re getting 90% of the plants potential as-is but doing this will squeeze out that last 10%.

Overall though, we don’t care if the plant is happy or not, we care if the plant is productive, it being happy is secondary to our need for it to be productive. Just because it doesn’t look like it has any issues, doesn’t mean it’s growing to the absolute best of it’s ability.

So I guess… TLDR; There doesn’t seem to be much benefit to the indoor grower to do this. You lose known flower time, possibly sacrifice yield, for potentially slightly(<10% ?) better quality? Maybe?

I’m always down for a good science experiment :sweat_smile:

7 Likes
#3

But what if photo plants start flowering at 14/10? Perhaps if you don’t know what light cycle will cause your plant to flower, you’re leaving light hours on the table? Just like autos, maybe you could flower some varieties at 14/10 and get some benefit as well.

Also if you’re doing a hybrid breeding scheme of indoors and outdoors, you could do a better job of selection during your indoor grows if you had a progressive light cycle. You could use the finishing date as a selection parameter as well.

I know it’s not convenient, but it seems to me that breeding isn’t convenient in the first place. And maybe the biggest problem is that you aren’t fully characterizing your plants either, so you aren’t informed about any photoperiod except 12/12.

8 Likes
#4

Yep all great points.

And definitely for a hybrid(indoor/outdoor) grower it could be a huge boon.

And would be useful for indoors to know if you’re potentially leaving hours on the table regardless :thinking:

4 Likes
#5

Might come in handy. Check this out:

This source calculates all of the sun parameters based on date and location. Feed the intensity information to your dimmer(s) to simulate the solar cycle. With some spectral control (not currently in the code) you could get pretty close to simulating the effect of far reds in the mornings and evenings.

Combine with this:

… and you can calculate the amount of solar irradiation for at any point in time for any location in the world or integrate over the day to obtain the DLI.

The source doesn’t have anything to simulate weather, though. Which, by the way, there is some research I’ve seen out there that details experiments on the effect of varying sunlight intensities. I can’t recall the conclusions.

Sample using the code to generate a DLI map for the world throughout the year:

image
image1440×1756 213 KB

11 Likes
#6

That’s great stuff. Have you been incorporating it into your grows? I don’t have dimmer control (yet), just a programmable way to control the light cycles via node red. But I was thinking adding dimming would be the next step. I think there are some compelling reasons to add it to a breeding methodology.

If you’ve been setting your lights this way, I’d love to hear your thoughts. Like do you notice many plants starting into flower at different times (ie different simulated dates)? When you grow, have you been tried to match strains to their natural light cycle?

5 Likes
#7

Well, I have some hugely complicated plans around a control system to provide for the flexibility to run scenarios similar to what you’ve described. And, is the reason the code exists.

But, I have not run an actual grow simulating a specific location at this point for a natural light cycle. I do control spectrum and dimming but to the general accepted practice. I figure someone would want to give it a shot before I ever get around to it :slight_smile: So, I do not have an explicit answer or than the assumptions based on what we know about short day flowering (which we also assume the timing can vary depending on the genetics).

There is value there to try these things. And, perhaps there is some unique insight to be gained particularly if combined with other instrumentation. For instance, Pr / Pfr conversation is a key piece we generally overlook or ignore because the grow still works in most cases even when we explicitly ignore far red and instead rely on time to do the work (dark). Though we know there are reasons to include it. The natural light cycle has a relative increase of far red at the end of the day, for instance. There is still a lot of stuff to explore with on the ground experiments such as what you’re considering.

8 Likes
#8

I’ve started thinking about it recently because I want to get back to Lebanese on my next grow. And Lebanese is one that I know can have a very early flowering onset — or what people have called semi-autoflower.

This is a desirable trait for for outdoor growing at my latitude. I’ve been hybridizing a mix of different Lebanese genetics, and now I want to start selecting for the ones that onset flowering the soonest. But without at least a diminishing light schedule, I don’t think I’d be able to determine when plants onset.

I think I’ll work on getting something basic setup. I can use either javascript or python in node red, so I might start trying at least just a simple on/off control that sets daylight hours and work from there. Dimming and red supplementation would probably help with getting more accurate dates for the end of flower.

5 Likes
#9

I will have read a long, wanted to do this with HID light and do a sun rise - mid day - sunset

to be more like outside

a great topic and thanks

Dequilo

3 Likes
#10

What comes to my mind is that if you could dial in a shorter dark cycle for growing indoors that you could reduce stretch and extend veg

3 Likes
#11

You know I have been manipulating my light cycles a bit :slight_smile: @HolyAngel if part of the cycle is 16/8, 15/9, 14/10, 13/11, 12/12 do we actually give up yield? I dunno about that. I can see where going shorter than 12/12 would cut yield …but maybe not if we drop slowly.

And then the OP has valid points about being able to tell which plants, the thing he did with simulating east/west and the stretch lack on the one side was intriguing.

2 Likes
#12

:sweat_smile: er…uhh…yeah.

:thinking: wasn’t exactly sure what ‘simulated’ is intended to mean. Because loosely speaking, any indoor growing is… right?

Dawn/dusk… far-red nitey-night… gradual/natural photoperiod changes vs. 18 -> 12 :clap:, etc.

:nerd_face:

:evergreen_tree:

4 Likes
#13

Well it could mean anything that makes lighting more “natural”, but I was especially thinking of gradual photoperiod changes.

Supposing I want to breed a photoperiod plant to flower under longer light cycles, like 14 or 16 hours, then I think the best way to do it would be to select the ones that flower earliest under a gradual photoperiod reduction. But also, it would help me to identify if my shiny new hybrid will finish outside at my latitude.

6 Likes
#14

My Bluefish LED controller has a feature that sort of let’s you do this. I can select a location and it will simulate that photoperiod and weather.

6 Likes
#15

Beta Script

image
image1285×1249 140 KB 

import datetime
import ephem
# set these at the beginning of flower
# do not change after!
latitude = ## lat goes here
longitude = ## lon goes here
elevation_m = ## elevation in meters
init_time = datetime.datetime(year=2021, month=1, day=5) # set at first day of flower
cycle_start = datetime.datetime(year=2021, month=8, day=1) # simulated start date of flower

# calculate current simulated date based in inits
offset = cycle_start - init_time
current_dt = datetime.datetime.utcnow()
simulated_dt = current_dt + offset

# in
obs = ephem.Observer()
obs.lat = '%.2f'%latitude
obs.lon = '%.2f'%longitude
obs.elev = elevation_m
obs.date = simulated_dt

node.warn('Simulated Time: %s'%simulated_dt.strftime('%Y-%m-%d %H:%M:%S UTC'))
node.warn('Next rising: %s'%str(obs.next_rising(ephem.Sun())))
node.warn('Next setting: %s'%str(obs.next_setting(ephem.Sun())))

lights = obs.next_setting(ephem.Sun()) < obs.next_rising(ephem.Sun())

if lights:
    node.warn('Sun is up!')
else:
    node.warn('Sun is down!')

return {'payload':lights}

I’m using a DIY box that runs node red on a pi (great write up by @8k_feet here) :

I wrote a script to simulate a sun cycle at 40N. Right now, it’s just checking every 15m to see if the sun is up or down. I’ll probably reduce this to a 1-5min interval later.

Fortunately, node red has a python function box I can use. Strangely, the python3 function was still using the pi’s python2 version, despite having both. I had to scratch my ass on that one for a while before I figured it out.

Today is August 2nd, and the sun is now set. :sun_behind_large_cloud:

8 Likes
#16

:pensive: ohm…

…what then, of plants bred for many generations under the 'ole 18/6 -> 12/12 regimen? :thinking: …do they have conniption fits & freak out under natural light? Weeeeeeeeeiiiiird…

My intuition/instincts/lawngnomes suggest that plants would never adapt & ‘demand’ indoor light cycles, just all the other enironmental factors. :blush:

:evergreen_tree: Rising sea levels mean I need to get higher every day.

5 Likes
#17

I think you’re misunderstanding. I’m looking to identify plants natural cycles, and breed toward earlier outdoor flowering – but do it indoors. 12/12 isn’t a magic number, it’s just a photoperiod that will cause most plants to flower. It’s always 12/12 because the breeders who make the strains use 12/12 lighting indoors, and don’t know if it will flower under a longer cycle. As far as I’m aware, nobody has bred strains indoors to use a longer photoperiod.

Some varieties are known to have longer flowering photoperiods. Lebanese, Moroccan, Syrian, Crimean landraces and many others may initiate flowering at 14, 15 or maybe even 16 hours. It already exists in landrace varieties, so how can we identify it and base our selection on the ones that start flowering the earliest? Is it a heritable trait? If I have stabilized a 14 hour photoperiod in one strain, can I pass it on in a hybrid?

8 Likes
#18

I really dig your line of thinking here. Selecting for induction of flowering at >14 hours could provide an opportunity to develop strains that can flower in full sun in northern latitudes.

4 Likes
#19

Searched for something like that and exactly what I was thinking about recently.
Thought about selecting for onset of flower as well as flowering length to bring the harvest more near mid September.

1 Like
#20

Yeah, hopefully I’m going to get to try it out pretty soon. I have some strains ready to go that might do it – Lebanese, Moroccan, Syrian, Sinai and now Crimea. I’m going to see how it goes and hopefully find some that trigger early.

5 Likes