When I started working for The Nature Conservancy in 1997, many of our prairies along the Platte River were unplowed but in fairly poor condition. Chronic overgrazing and broadcast herbicide use prior to the Conservancy’s ownership had greatly reduced plant diversity and allowed invasive grasses to gain dominance. It became quickly apparent that simply managing those sites with fire and/or moderate grazing was enough to reduce the dominance of the invasive grasses, but not enough to bring back many missing wildflower species.
Because we were actively harvesting large amounts of local wildflower seed to restore crop land to prairie, we started experimentally broadcasting seed onto our degraded prairies as well. After much trial and error, we came up with a technique that was effective and fit into our broader management strategies. Essentially, we broadcast seed onto recently-burned prairies and then used grazing to reduce competition from grasses. In the early days, progress was slow and hard to see, but over time the grass-dominated grasslands started to gain more and more color.
Seeing more wildflowers as we walked through those prairies made us feel good, of course, but there’s a big difference between feeling good about a technique and knowing it is providing actual benefits. We had data showing long-term persistence of the new plants that emerged after our seeding – in other words, our management was sustaining higher plant diversity once we got it established. However, we also wanted to know whether or not those more colorful prairies were actually providing better habitat for pollinators or wildlife.
Over the last two years, we’ve collected data on flower abundance across both our Platte River Prairies and Niobrara Valley Preserve. We walked belt transects (divided into multiple segments) each month and counted every flowering stem within a meter on either side of us. Those data give us the ability to see how plant diversity and abundance changed through the seasons, something that is obviously important for pollinator populations. The data also let us look at how various management treatments and systems affect flower availability in our prairies.
As an additional step, Mike Arduser (former Missouri Department of Conservation ecologist) looked at our Platte River data and categorized our flower species in terms of how they are used by bees. While that is a messy process and there are various ways to approach it, we have (among other things) a list of the flower species that are used by a moderate to high number of bees. For the sake of this post, I’m calling that a measure of “quality”, but of course there’s a lot more to the value of a flower to pollinators than the number of bees using it.
Regardless, all of that description is just background for the main point of this post, which is: we can use those data to see whether or not our overseeding work is meaningful to bees! Even better, the answer seems to be yes – with a few interesting twists.
The three graphs below show different parts of the same prairie/pasture (also featured in the three above photos), which is being managed with patch-burn grazing. The west patch was burned in 2016, the middle in 2017, and the east in 2018. Following each burn, that area was grazed intensively by cattle for a full growing season. Grazing intensity generally decreases with time after a burn. Thus, in 2018, the west patch (burned in 2016) was grazed least intensively and the east patch (burned in 2018) was grazed hardest.
The orange bars show flowering stem numbers for species that are used by a moderate to high number of bees. The blue bars show only those species that were there before we did our overseeding work. In other words, looking at the difference between the orange and blue bars lets us see how much our overseeding has added to the availability of important flowers for bees.
We confidently say now that our overseeding work has increased the number of available flowers for bees. The increase varies somewhat by season and location, but there is an especially strong and consistent difference in early September. Much of the increased flower abundance during that period comes from one species – Maximilian sunflower (Helianthus maximiliani). That species should be a major component in our prairies but was missing from this one because of past management. In addition to being an important resource for bees, Maximilian sunflower is also listed by the Xerces Society as an important nectar plant for monarchs, which migrate through Nebraska in September. Because of that, it seems likely that our overseeding has increased the value of our degraded prairies to migratory monarchs as well.
Another story that comes out clearly in the above graphs is the lack of bee-friendly flowers during the early part of the growing season. This is not restricted to our degraded prairies, and it’s not unusual for this part of the country. Many of the flower species used most by bees in May and June don’t typically grow in large patches, but rather as widely-scattered individuals or small patches (which means we miss most of them with our narrow belt transects). We’ve done a lot of brainstorming, but haven’t identified many bee-friendly flower species we would focus on if we wanted to really boost blossom numbers during the late spring. The best option we’ve come up with is prairie ragwort (Senecio plattensis), a short-lived plant that can become temporarily abundant under some conditions – including after intensive grazing – but not at all sites. Regardless, these data help highlight that continuing issue and provide a reminder to keep up our discussions.
Collecting monthly data on flower abundance/diversity was a pilot project for us to see whether or not we could gain valuable information from that technique. I’d say the effort was very rewarding, and for much more than just confirmation that our overseeding work was worthwhile. Hopefully, we can look at some different sites in the coming years and continue to learn about how our work is affecting pollinator resources. I would encourage others to try something similar, even if you just use one or two short transects. The data have really helped us evaluate our sites in a new way.
To be clear, the degraded prairies we have overseeded along the Platte River have not regained their full wildflower diversity. We’ve put some species back into the community, but the history of broadcast herbicide use and overgrazing is still evident. We may never get those prairies back to full strength, but we feel good about the progress we’ve made.
…And now we have data to back that up!
The Prairie Ecologist 
Thanks for the update… so valuable to have these posted and available. Succinct and well-written.
We overseeded forbs in 2013 and used grass-specific herbicide in a +/- similar way you used grazing, to reduce competition from NWSG. Mowing is another strategy… (I can picture long strips at given intervals, mowed a couple times a year, and let forbs colonize out from there).
Will be presenting 6th year data at SER meeting in Pella, IA this April.
Nice, Adam, and good to hear from you. I’m glad you’re having some success there as well! We’ve tried mowing here, and it works ok, but not nearly as well as grazing. I think we’re more moisture limited here than you are, and we might have to more strongly suppress competing vegetation to get success – but that’s just a hypothesis.
I am assuming “overseeding” means planting on top of what is there after a burn not overdoing the application of seed to the ground. Is this correct?
Also, Joel Sartori runs an Instagram post where he photographs endangered animals. Are there Nebraska native prairie species of animals that are endangered? Could you talk about them?
Yes, see my answer to Steve below about overseeding.
Joel’s work is fantastic, and draws much needed attention to endangered species. We have some animals in Nebraska that are federally listed as endangered, but not many in prairies. Here is a full list: http://rarespecies.nebraska.gov/wp-content/uploads/sites/2/2014/02/Nebraska-TE-Species-Complete-List.pdf
The American burying beetle and black-footed ferret are the two endangered animals that (could) live in Nebraska prairies, though some of the other occur in wetlands/streams that are in/near prairie landscapes. Burying beetles have scattered populations and we’re still trying to learn more about what they need to thrive. Keeping prairies free of woody encroachment is certainly one important strategy for them. Black-footed ferrets rely on very large prairie dog towns, which have become rare now.
Impressive. Your graphs and photos give evidence of what your intervention has done. All across the country others are working on spreading native plants for pollinators and other benefits. Thanks for sharing the information.
Really interesting! I wonder why the history of the sites prevent totally successful overseeding now. Residual herbicide in the soil? It seems like the condition of the soil might be decisive. Has there been much analysis of current soil conditions vs less degraded sites? What soil conditions are needed that don’t exist yet, and how can those conditions be more effectively brought back? Questions, questions…
So far, we’ve limited our overseeding work to plant species that should be abundant in this kind of prairie but are missing. We haven’t tried all species that are missing. That said, some species certainly respond better (are more successful at germinating and establishing) than others. I think that’s mainly related to plant competition, rather than soil conditions, though the two are related and we don’t know all we need to about the soil part of the equation. I don’t think there are herbicide residues causing problems, I think it’s just we’re trying to introduce seeds and get them to germinate where there is already a full complement of existing plants. Even with the grazing we’re using to suppress the vigor of those other plants, there are only so many places a seed can land and have a chance to compete successfully. Getting the seed to the soil in-between existing plants is the first challenge, and higher seeding rates are most helpful mainly because they help us hit more of those potential spots (but we “waste” a lot of seed that lands where it can’t possible germinate and grow). Once the seed germinates, though, it still has to fight all the surrounding plants for survival, and some plant species are better suited than others to do that. Not all prairie plants are designed to compete in this way – they’re better at spreading by rhizomes, for example, or for other reasons don’t easily insert themselves successfully into a site where other plants already dominate. Figuring out how to get those species back into these prairies will be a challenge, and frankly, may not be something that is worth our time/effort relative to the other challenges we’re dealing with. We might have to live with these degraded prairies being better, but not perfect.
Makes a lot of sense. Thanks!
This is fantastic, Chris. I love the methodology and data, and your concern about early season forbs. Managing revived prairies as large gardens — what a task!
Very nice work in the field and in compiling this report. It’s always gratifying to learn that your efforts do pay off.
I too am curious about “overseeding”. How does overseeding differ from broadcasting?
Sorry for the terminology confusion. Overseeding simply refers to our practice of taking seeds and broadcasting them across the site (usually on burned sites or at least hayed/recently grazed sites) where bare soil is exposed. It is similar or identical to interseeding, but I usually think of interseeding as being done with a seed drill, as opposed to broadcasting seeds on the surface of the ground.
Thanks Chris!! Now I have a little more knowledge on the subject.
Happy New Year!
I have long been interested in how to plant wildflower meadows. And I’ve been working to get Refuges to let me do bee surveys and to manage for bees. One of the things I’ve learned is that many native bees only use the plants of one or maybe two genera. So the number of bee visits would also be a function of the number of a bee species present. Therefore, some of the plants with less documented visits, may, in fact be more important to the continued existence of one bee species than some of the plants used by more generalist bees. But I agree, that any extra bee and butterfly activity is cause for celebration.
You’re absolutely right, and we’ll be looking at individual flower species and how they’re impacted by season and management as well. As you say, some of those may not be important to a lot of bee species but critical to a few bee species.
Interesting data Chris. In these degraded sites, is there a lot of competition with cool season grasses? I wonder whether germination of earlier season flowers which normally would bloom during that period are inhibited by this competition. Perhaps the success of later blooming species is related to senescence of cool season grasses?
Yes, we have a lot of bluegrass and brome at those sites but I’m not sure they’re the cause of our missing bee-friendly wildflower species in the spring. I think it’s more that the wildflowers blooming that time of year tend to occur in small scattered patches or aren’t visited by lots of bees. Remember also that there are fewer bee species active during that time of year, so the system is kind of self-calibrated in that way. And, species like bumblebees, which are just starting to grow their new colonies for the year have fewer mouths to feed in May and June, so they need fewer flowers. Combine that with flowers on shrubs like wild plum and others, and there may be sufficient nectar and pollen resources, even though our graphs don’t show it. Still a lot to learn.
Yes, fair points. Perhaps an interesting exception to the paradigm you mention is pasque flowers. On my prairie, they are abundant and attract many bee/wasp species. Perhaps because they bloom earlier than most other forb/tree species?
If I lived nearby, I’d come out and collect Lithospermum seed for you. This would be a good nectar source in June. I live way too far away, so someone else should consider doing this for your sites (blatant hint to your volunteers).
Other spring/early summer flowers that would be good to have seeds from include violets, blue-eyed grass, spiderwort, pale spiked lobelia, small skullcap, penstemon, yellow-star grass, and others.
Lithospermum produces lots of seeds later in the season because of its self fertile flowers it produces. It would be an excellent plant to get introduced. I also think spiderworts are used by pollinators big time but spiderwort is a monocot and isn’t killed by broadleaf herbicide. I’d think it would still be in the prairies. I think Lomatium is another good early spring wildflower. Does it grow in the platte valley?
We have quite a bit of Lithospermum incisum in our prairies (remnant and restored), but it’s not actually visited by many bees, so it’s categorized as “low” in the particular analysis I used in the blog post. That said, it’s certainly important for the bees that do use it, and is (according to Mike) a species very important to the development of burgeoning bumble bee colonies. So I’m not discounting it – it just didn’t fie the profile for this particular analysis. Lots of spiderworts in our restored prairies, but not so many in the degraded remnants. Probably a function of chronic overgrazing. Spiderworts can handle some grazing, but cattle really like them and I would guess they’d get knocked out by too many years of intensive grazing in a row. Also, they don’t have any nectar – just pollen, and tend to be visited much more by flies than bees. Not unimportant, but another species categorized as “low” for this particular analysis. We don’t have any Lomatium in the Central Platte Valley’s lowland sandy soils, so that one isn’t a possibility.
It is interesting that your native spiderwort (T. bracteata) does not produce nectar. Upon doing a google search I am finding some people say the more eastern species produces nectar and attracts various bees and others who write no nectar is produced. Other species that might be good early season sources of nectar for bees in the Platte River Prairies include Callirhoe, Astragalus crassicarpaus, and Psoralidium tenuiflorum. Antenaria neglecta might be useful to some bee species but is probably already present in formerly over grazed prairie. Ranunculus might also be useful to some bee species if the area is appropriate for a given species.
Very interesting. I notice here in Kansas heath aster is even later to bloom than Maximilian’s, and it will attract copious bees into October. It’s one of my favorite wildflowers, so I let it grow all over my yard. I never thought of it as a monarch food, until this year when I did have several over two days in early October.
Hey, I really like this story, it’s great to see that kind of progress. One question though. Have you done anything to quantify whether bee numbers and species diversity are responding to the changes in plant species richness – pan trapping, camera trapping etc? Do you have baseline data on bee or insect numbers? Sometimes it’s almost sad to do something like this without getting the baseline data first (but still better than not doing it at all of course!)
In reading the full SWG-C report, it appears that regal frits are targeting burned areas to lay eggs in the fall based on no regals the year of the burn (during peak flight) with regals in that same unit the year after the burn. If that isn’t the case, perhaps the structure of a previously burned area (the year before) is attracting regals to the area that weren’t “born” on site (shorter vegetation/less “rank” grass etc.). As regals were surveyed into the fall, it would be interesting to note where regals are in late August/September as those locations would likely be where the females are laying eggs. Sorry my post isn’t directly related to this post, but the post led me to your final report which is excellent.
I haven’t been keeping up with blog reading lately. I always enjoy your posts when I get to them, and am so thankful for the work you and others do for the prairies and their inhabitants.
How would you characterize the availability of flowering shrubs near these prairies? In my part of the world (SW WI), I tend to see insects nectaring on flowering shrubs (e.g. Prunus spp., Cornus spp.) in that May – June period when there are less flowering herbs available.
Hi Bill,
We definitely have flowering shrubs in and around our prairies, but fewer than I’d like. During the early spring, we have patches of wild plum and sand cherry that provide really important pollinator resources. We also have coralberry (Symphoricarpus orbiculatus), dogwood, sumac, and a few others later in the spring/summer. When our prairies were smaller and more isolated, we worked really hard to open them up as much as possible (clearing all trees and shrubs from edge to edge) to maximize effective size. Now that we’ve done a lot of restoration work to reconnect and enlarge many of the sites, we have room for shrubs again, so we’re trying to be more accommodating of them.
I intend to try some sort of seed mix on a strip of bare ground exposed by removal by hand of a strip of overgrown fence line. Cattle have continual access to this area, but my stocking rate is lower than that of most cattle farmers in my area (West KY). Any species suggestions welcomed.
Hi Frank,
Sounds like you’re doing good work. I can’t really give you specific suggestions since I don’t know that area well, but here are a few thoughts. If you just want the area to fill in quickly with grass, that’s pretty easy – just use a mixture of native grass at a high seeding rate. You might throw in a some kind of annual grass or oats to help get things started. However, if you’re hoping for something with more of a wildlife/pollinators benefit, here are some things to think about in that regard: I’d suggest putting in a mixture of native grasses that include both cool-season and warm-season species, but at a fairly low seeding rate. Then pick out native wildflower species (as many as you can afford) and choose a mix that ensures a few species will be blooming at all times during the year. Seed those at a fairly high seeding rate. If you include some annual wildflowers, they’ll give you a quick response and also act as a kind of cover crop for the slower establishing species. If things go well, you’ll end up with a strip of wildflowers that will add some interesting wildlife cover and valuable pollinator resources. Eventually, the strip will probably become more dominated by grasses as your seeded species establish and grasses from the surrounding area fill in – then it’ll come down to management to help keep the stirp relatively diverse. Again, all of this is being suggested from a long way away, so it would be smart to run it past someone with local expertise. If you have particularly nasty week species that are likely to invade, for example, that would change the strategy… Good luck!
It seems to me grazing by animals would play an important part in your work. How often, when, and how many. Are cows the only grazing animal being looked at? Do they eat different plants than Bison? You don’t seem to mention grazing very often, considering what an important role it obviously plays.
Fascinating work. You mention that earlier flowering plants are scarce. Had work been done to try and incorporate tickets of early blooming plants into sections of edges of these prairies? Plants like Prairie Willow, Prunus, Viburnum, Hawthorn, etc? I’d imagine even a few of these added to a prairie would make an incredible difference in pollinator diversity because each plant is producing thousands of flowers in spring while the Prairie itself is still mostly growing.