I’ve seen a lot of prairie plantings. Many look reasonably good, a very select few are diverse and complex to the point they resemble remnant prairies, and a good number are little more diverse with native species than an abandoned agricultural field would be after a few years. Most abandoned fields will grow some asters (especially frost aster, panicled aster, and New England Aster), goldenrods (especially tall, Canada, grass-leaved, and giant goldenrods), and common milkweed after two to three years. If openness is maintained, fields may ultimately be colonized by, bergamot, native sedges, self-heal, and others. These meadows have value for pollinators and other wildlife. Prairie plantings, where investments have been made in preparation, seeds, and hopefully subsequent stewardship, should aim higher.

There will be some specifics in this post, but I’m more interested sharing some guiding principles. All of this assumes good site preparation and subsequent stewardship.

Creating a diverse, complex prairie planting that will be resistant to anthropogenic stressors ranging from climate change to invasive species is no simple task. The prairies that Europeans destroyed developed over thousands of years, with their component parts arriving (and sometimes arriving and departing…and arriving) each at their own pace according to their responses to conditions in North America over thousands of years. The species composition of an old-growth, remnant prairie should not be taken to suggest that a seed mix with the same composition will lead to a similar outcome.

For the following, I will assume that the land steward is making a long-term investment to help heal the land. The planting is not intended to be part of a shifting mosaic working lands destined to be brought back into cultivation within a decade or two. The steward wishes for a legacy not just of cleaner water or more wildlife on the landscape, but more life, complexity, and beauty on the focal property, and the steward dreams of what would emerge and the landscape if others did the same with their land. Biodiversity and the intrinsic value of all parts as essential to the whole are at the core of the steward’s ethics.

I also assume that the planting is not immediately adjacent to an existing high quality remnant. If it is, than the seed mix design should be heavily influenced by the goals set forth in a management plan for the remnant, and species selection and seed sourcing require more nuanced consideration.

I’ll start with what not to do. These are the actions that lead to outcomes that differ little from the alternative of simply allowing a fallow field to begin succession, or which yield warm-season grass over-dominance.

1. Do not boost amounts of early-maturing, showy species for a big show in the early years. Those species are great for the bees and the butterflies for a few years, but we’re in it for the long-haul, and those species will diminish after a few years. The can also diminish the initial establishment of the real long-term doers when it comes beauty and function in the prairie planting. In 10 or 20 years, we want a lot of false indigos, lead plant, compass plant, prairie clover, blazing stars, butterfly milkweed, and on and on. Some stewards achieve blue-eyed grass, wood betony, wood lilies, and prairie phlox! Many plantings only utilize species that peak early like false sunflower, black-eyed Susan, yellow coneflower, and bergamot; these also tend to be the species that can meet conservation standards for the lowest prices. Such plantings are usually over-dominated by grass and forb-poor by the time they are a decade old. The grass itself is good habitat for some species, but it could have been better, and it’s a problem that is difficult to undo without starting from scratch. I’m not saying not to use those early-maturing species, but they should be small components in the seed mix…enough to ensure they establish, but not so much they run rampant in the first three years. This would mean planting things at rates of a 1/64 oz. per acre that are more often planted at a rate of ounces or even pounds per acre. Their negative impacts can be lessened by growing season mowing/haying during years two and three, but that defeats the point of including them in large amounts by preventing their flowering, and there can be more impacts from mowing on ground-nesting birds.
2. Do not use the big grasses (big bluestem, indiangrass, and switchgrass); at least don’t use them in greater than trace amounts. They often show up regardless as minor contaminants in other seed. These grasses simply establish and occupy real-estate too quickly, and if they establish early in large numbers they prevent slower-developing, desirable species from becoming established in the first place. Even before they flower for the first time, they are reaching out laterally with their rhizomes.
3. Do not put up too many constraints when it comes to species and seed source selection. Prairie plantings go into altered soils in landscapes containing newly added species, all under a changed (we’ve already shifted things a couple hundred miles) and rapidly changing climate.
4. Do not introduce seed only once. It’s true that there are diminishing returns later, but there are a lot of valuable species that simply don’t seem to like establishing on the freshly prepared seed bed. Many of these species are the early bloomers and the partially parasitic (hemiparasitic) species that provide critical early resources to pollinators and just might be the glue that holds the high small-scale diversity of tallgrass prairie together.

Instead, do the following.

1. Do make a long-term investment in the seed mix, and plant 50, 100, or more species. Don’t plant pounds of common milkweed. Plant ounces of butterfly milkweed (or whatever more conservative milkweed is site-appropriate) and a fraction of an ounce of common milkweed. Plant more sky blue aster than New England Aster. Plant a little stiff sunflower, and plant trace amounts of early sunflower. If this means planting just a few acres at a time, do that. Weather can have a big impact on establishment, so it’s probably best to do a little at a time anyway. Buy what you can afford. Collect and add whatever you can. Serious effort in a small area can yield valuable seed for future efforts.
2. Do use bunch grasses and sedges (graminoids). Many old-growth prairies are dominated by species other than the big, rhizomatous grasses anyway. Clumping species like prairie dropseed, rough dropseed (eroded, overfarmed soils), little bluestem, prairie wedgegrass, Junegrass, and plains oval sedge should be mainstays. Rhizomatous species should be lower-growing and less aggressive, like side-oats grama. Graminoids should be no more than half of the seed mix by weight, but they are essential for the development of fuels that will carry the fires, which are themselves part of the prairie system.
3. Do use a wide range of species and genotypes that are native to prairies regionally (perhaps level II Omernik ecoregions). Use what is locally available. Use species that should be well suited to site conditions, even if they have to be sourced from farther away. Draw seed for individual species from multiple sources, both as local as possible and farther afield. Push the ranges of species, especially those from within a couple hundred miles to the south, so long as they were/are part of similar prairie systems, because odds are you’re in their climate envelope now or you’re about to be. Very little is actually pinned down about local adaptation in the context of restoration/reconstruction. Research has focused on plant size, seed set, genetic differences, morphological differences, etc. and almost universally sidesteps what is critical for at least the long-lived prairie species, and that is their initial establishment and subsequent year to year survival rates, which are what drive their establishment and persistence/population growth. The results that are out there are mixed. I believe variation in the context of all of the change and fragmentation on the landscape is the most prudent course. What does that look like? Well, I wouldn’t shame a planting for having rosinweed north of its pre-European range. A project might source seed from one producer or another, and the seed might not be super local, but the steward might manage to collect and add some local seed, and the steward might deliberately bias the acceptable seed source region a bit to the south, and that might mean sourcing more southern species as well as genes, so long as they come from similar communities and share recent co-evolutionary histories with the local flora and fauna.
4. Do keep seeding. Any good plan is going to see to it that a planting is burned as early as it has the fuels to carry fire and frequently, if not annually, for at least several years after. Whenever it is possible to burn in the fall, over-seeding with species that initially failed or which weren’t included initially should occur, even if just small amounts are available, and this should probably continue indefinitely, given the fragmented state of the wider landscape. Partially parasitic species like wood betony should be added during this time. If the steward is thinking across generations, species typical of remnant prairies, but which are difficult to acquire seed from (or grow from seed) may even be planted out as nursery-propagated transplants and tended to until they become established, things like prairie rose, hoary puccoon (now at least one producer is growing this species from seed and selling it in pots), Mead’s sedge, and wild strawberry.
5. Do keep records of what was planted where, when, how much, and from what source(s).