Read on Twitter
One of the things that interests me most about weeds is the seeds. Weed seeds are a pain in the ass.
Most weeds produce a LOT of seeds. Thousands. Hundreds of thousands. Maybe even a million. Per plant.
And those weed seeds can survive in the soil for a LONG time. At minimum, 2 to 3 years. But some seeds will live for much longer. Decades even. Some of the weed seeds produced this season will still be viable and ready to grow when my kids are my age.
We call that the weed seed bank. Like a real bank but for seeds. Except without tellers lobbies. OK, not really much at all like a real bank, except we think of new seeds as deposits to the bank, and seeds that die or germinate as withdrawals from he seed bank.
The weed seed bank is really why annual weeds are a persistent problem. There's always more waiting to germinate.
And there are a lot of weed scientists who study the weed seed bank, trying to understand it better. The more we can learn about how long weed seeds live, and what practices can cause weed seeds to germinate or die, the better we can manage those weeds over time.
But weed seed banks are a pain in the ass to study.
Scientists have developed ways to try to study weed seed banks, but many of our methods are pretty artificial.
For example, we'll put 100 weed seeds in a mesh bag or screen, then bury the bag. Come dig it up later and see how many seeds survived.
Or we'll create an 'artificial' seed bank, by sterilizing an area of soil and putting just our own seeds in.
This is useful! We can learn a lot about seeds and mechanisms of germination and expiration from these methods. But also, these methods are not always predictive of what would happen in the real world.
A really cool study by Ramon Leon and @mdkowen demonstrate this difference between a artificial and natural seed banks with respect to seedling emergence. The artificial seed banks WAY overestimated seedling emergence. https://www.cambridge.org/core/journals/weed-science/article/abs/artificial-and-natural-seed-banks-differ-in-seedling-emergence-patterns/5887872D94DE9F744BFE645FF895A9EE
The mesh bag method also tends to overestimate seed death - because if one seed in the mesh bag gets attacked by a pathogen and dies, then the other seeds are likely to get attacked and die. https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/j.1365-2664.2005.01016.x
So this is a problem if you want to understand what weed seeds are really doing down there in the soil.
One of the first projects I started working on when I got my job here at UW was developing a method to 'track' weed seeds in the soil without using these more artificial methods.
This is one of those ideas that I think is good, but we have never been able to get single damn cent of grant dollars to pursue the idea. @ERECWeedScience Did some of this work while he was here. I'm tagging him now to remind him we should publish this some day.
We decided to try using 13C, a stable isotope of carbon that is relatively rare in the atmosphere, to 'tag' the seeds as the plant was producing them. If successful, we'd be able to tell those seeds apart from others that were in the soil.
So we bought one of those things you see in the contagion movies. Air tight, and you can put your hands in the box through the built-in gloves. We put plants in here at early seed development stage, and sent 13CO2 in so the plants would fix a high proportion of 13C in the seeds.
And it worked. We only had to put the plants in the little carbon chamber for 2 hours, and the seed ended up with plenty of 13C to differentiate it from plants that weren't exposed to the 13CO2. [results table for the isotope peeps]
![And it worked. We only had to put the plants in the little carbon chamber for 2 hours, and the seed ended up with plenty of 13C to differentiate it from plants that weren't exposed to the 13CO2. [results table for the isotope peeps]](https://pbs.twimg.com/media/EoMD7fwXEAMbnaI.png "And it worked. We only had to put the plants in the little carbon chamber for 2 hours, and the seed ended up with plenty of 13C to differentiate it from plants that weren't exposed to the 13CO2. [results table for the isotope peeps]")
This data was promising, but we had to test it in the field. So we put up a 'tent' and did the same thing outside.
I like this photo.
We put the tents on and put the 13CO2 in when the seeds were filling. We then went back to collect seeds from the plants to make sure they had a 13C signature. And they did.
And then the cool part: we went back repeatedly over the next couple of years and plucked the seedlings and analyzed them for 13C content. We also did the same thing from an area of the field that we didn't tag with 13C to be sure we knew what the 'ambient' 13C signature was.
For the next 3 years after we tagged the area, most of the emerged seedlings had the elevated 13C, suggesting those seeds were produced the year we tagged.
So we showed this method can work! But then we ran out of bootlegged funding sources to continue the work, and all of the grants I submitted to continue this work were not funded. So we stopped before we could do anything truly useful with this method.
And, well, that's life as an academic scientist, I guess.
