The Middle Fails To Thrive

Sometimes fine successes can blind us to lurking, festering flaws. We have been heartened by demonstrable increase in the health of our woodland and very open savanna areas. We have focused efforts on continuing their ecological recovery. But what about the large intermediate areas where our efforts have failed?  

For the positive, consider the two inspiring photos below:

Restored Somme savanna grassland: June 28, 2021

Twelve days later: July 10, 2021

Prairie-like open savanna areas are the pride of Somme Prairie Grove. Their diversity is such that every two weeks a new set of highly diverse conservative species blooms from spring through fall. Starting in 1980 as mostly alien old field vegetation, the most open parts of the Somme savanna restorations, year after year, have seemed to be more and more similar to the best remnant prairies. That is, they now seem like high-quality, rare restored nature. They feature such conservative species as cream false indigo, leadplant, yellow star grass, hoary puccoon, bastard toadflax, Leiberg's panic grass, and all those species with prairie in their names: prairie violet, lily, coreopsis, alumroot, dropseed, gentian, and white and purple prairie clovers. Oh, yes, they also include savanna specialist species, but we’ll get to that later. (See Endnote 1: Evolving Seed Mixes.)

Parts of Somme’s bur oak woodland areas also thrive with year-by-year increasing quality. (For photos and details of the darker parts of the ecosystem, see Endnote 2: Open Woods Flora.)

But … now comes the confession.  As we walk from grassy, bright open toward the woody shade, we must pass through dismal failure. 

Heavy dominance by one species - tall goldenrod

Between the brightest and shadiest areas, bands of intermediate brightness, as shown above, have little diversity or quality. Initially, following buckthorn control and seed broadcast, they seemed to have good potential. But after a few years they tended to become dominated by weedy native tall goldenrod (Solidago altissima). Such areas are unstable. They don’t allow the establishment of the grasses that are the base fuels for a good burn. Every time grassy or "oak-leafy" areas burn, the diversity, conservatism, and health of the ecosystem seems to increase. But, as the fires skip these intermediate areas, that health doesn't improve, and many have reverted to brush. 

In other cases, we’ve controlled the invading woody species in hope that conservative diversity would eventually emerge. Instead, after many years, the stands of tall goldenrod have often been replaced by woodland sunflower – which also impedes the development of enough grassy fuel for regular burns.

Heavy dominance by woodland sunflower 

It is perhaps positive but not remotely sufficient that the relatively conservative woodland sunflower replaces weedy tall goldenrod. When sunflowers (Helianthus strumosus or hirsutus and hybrids) take over these areas – after a decade or two – Floristic Quality calculations typically go up, at least temporarily, because tall goldenrod rates a weedy 1 while woodland sunflower rates a mid-conservative 5. But is that “progress” an illusion? As these areas continue not to burn - will brush take over?

Along with goldenrods and sunflowers, these Intermediate areas may be covered with briars, saw-tooth sunflower, willow aster, and other aggressive species. In some areas, these low-quality communities may support some of the site's endangered species populations. Such areas can be colorful at times (when their few species are in bloom). 

But we study these areas closely and ask questions. Are the low-diversity patches advancing “cancers” that will spread into and degrade the quality areas around them? Or are they, once succeeded to sunflower, gradually gaining sustainable conservative diversity? In the absence of fire, are they stalled in their recovery and unsustainable in the long run? Whatever, how to we decide what’s the best management for them?  

In 2021 we began an ambitious new approach. (Are we overly hasty? See Endnote 3: Too Impatient?) 

New Hypothesis

Our new working hypothesis is that the recovering savanna at Somme now consists of four parts – two successful, and two not, at least so far. These parts differ in a) their response to fire, b) their community structure, c) their health or success, and d) management needs. They can be described as: 

• Prairie-like savanna: There is enough sun to support tall grasses sufficiently dense to burn regularly. Herb diversity and health improve over time with our current stewardship.
• Intermediate "open savanna" – more shady than the above: Currently, not burnable under moderate conditions. Diversity and health do not much improve over time. May lose diversity to dense tall goldenrod, woodland sunflower (esp. H. hirsutus), or brush. 
• Intermediate "closed savanna" – less shady than woods-like savanna: Currently, not burnable under moderate conditions. Diversity and health do not much improve over time. May lose diversity to dense tall goldenrod, woodland sunflower (esp. H. strumosus), or brush
• Woods-like savanna: The trees are sufficiently dense that a continuous carpet of dried oak leaves regularly supports fire. Herb diversity and health improve over time with our current stewardship. (See Endnote 4.)

Until now, our ongoing hypothesis (developed with mentors Professor Robert Betz, Forest Preserve staff, and a long list of others) has been that if we cut brush, controlled invasive weeds, planted diverse rare seed, and burned frequently … then diverse quality vegetation will increasingly take over and become largely self-sustaining. This overall hypothesis here seems to have been confirmed over the decades – as the site’s number of native species has doubled and per plot conservative diversity for most areas has doubled, tripled, or quadrupled. This healing proceeds in both the sunniest and shadiest areas, as we’ve seen. But, oops, not in the intermediate. Experts and new visitors alike admire the beauty and complexity of many parts of Somme Prairie Grove. But in the Intermediate, we mostly avert our eyes and move on.  

Dr. Betz emphasized that even the best tallgrass “natural areas” have been profoundly diminished in their biodiversity by a century and a half of human disruption. Some types recover more readily than others. Perhaps “Intermediate Savanna” recovery is just slower, or perhaps these areas need restoration approaches that have not yet been developed.

Three New Restoration Approaches.

We rejoice in the splendid biodiversity gains of perhaps half of the site. Maybe health (sustainable diversity) will gradually spread into the ailing, dismal areas. But we stewards have poked away at the intermediate problem indecisively for long enough. After 41 years of restoration, we stewards have added a new focus to our work.

In the areas outlined in orange on the map below, we have adopted three new strategies to wrestle with the challenge of the intermediate. 

Strategy 1: New seed mixes

“Why so many seed mixes?” some people ask. “Why be so finicky? Why not just throw all seeds everywhere and let the nature decide what wants to grow where?”

The problem is that we don’t have large enough quantities of the rare seed that’s required – especially for many conservative species. When a newly sown turf fails to achieve sufficient diversity and strength to hold its own against the “aggressives", the diverse species die. The rare seed was wasted. We try to enhance our seed effectiveness by focusing it on the right places.

The new mixes shown in Endnote 1 were developed principally by Sai Ramakrishna, Eriko Kojima, Christos Economou, and Katie Kucera – based on literature review and observation. One of the main new approaches is to give the intermediate areas more seed of the grasses, the warm- and cool-season prairie, intermediate, and woodland grasses and sedges.

Strategy 2: Facilitate restoration of widely-separated oaks in higher-quality areas that border the ailing intermediate areas.

For four decades we did wait for one element of nature to take its course. Our hypothesis had been, “It’s best to let the fires decide” how savanna oak density should be restored. Following previous timber harvest and grazing and even plowing in some areas, the site faced two very different challenges – no oak areas and too-dense oak areas. 

At one end of the shade continuum, we decided to speed a return to widely spaced oaks by thinning (especially in the artificial white oak tree-plantation areas). Here we also use Strategies 1 and 3 to see if we can avert establishment of pesky tall goldenrod or sunflower. 

At the other end of the shade continuum, in some of the long-restored, high-quality grassland areas, hundreds of young oaks seemed to be waiting their turns. For decades, their tops have burned off every couple of years. But the roots survived. After each fire, those roots had a chance to get bigger and healthier and to put up larger, faster-growing young trunks. In the 1800s, on original, prairie-like open savanna, these re-sprout oaks were common and called “oak bushes” or “oak grubs” because farmers would have to “grub them out” to prepare a field for plowing. Over the millennia, chance variation in fire frequency would occasionally have given some such oak bushes the opportunity to mature fire-resistant trunks. Our expectation was that sooner or later, these shrubby oaks would be big enough to withstand the fires. That’s perhaps true, but how long would we have to wait? 

Part of their problem these days was that these trees had an additional new stress to deal with. In addition to fire, overabundant deer sought out the vigorous new shoots in springs after a burn and consumed them repeatedly, mercilessly. Over the decades we’d come to know well some of these little trees, as we used them to mark the corners of some experimental plots. We noticed that, as we checked on those experiments, the oaks that were one foot tall in 1980 were still one foot tall in 1990 - and still in 2010. We decided that, for this experiment, we had waited long enough. To give young oaks more of a “head start,” we began to protect some from deer with cages.

We also began to protect these few oaks from fire with a two-step process. Step 1: We raked fuel away from young trunks in years when we planned a burn for their areas. We also needed another step, because however laboriously we raked away the dense fuel, the trunks were killed anyway as 20-foot flames swept by. So, Step 2: On the day of the burn, we ignited little backfires on the upwind edge of the raked areas shortly before the main flame fronts (headfires) got to them. 

When we choose trees to favor in this way, another hypothesis came into play. Morton Arboretum scientists have expressed a belief that the genetics of our sugar maples gradually changed once landscape burning stopped. The old “pink-barked black maple” element had been swamped out by the faster growing (in the absence of fire) eastern sugar maple. We noticed great variety among our bur oaks and wondered if something similar was happening to them. Some (slower growing?) young trees had very thick corky bark; some (faster growing?) trees had smooth thin-looking bark. Hypothesizing that the corky bark was related to fire tolerance, we tried especially to assist young trees that had thicker bark.

For photos and more specifics, see Endnote 5 - Restoring Bur Oaks.

Strategy 3: Scything the aggressive species

To give conservative diversity an opportunity to out-compete aggressive species, we have begun to scythe away these "aggressives" once or twice a year. Typically we scythe tall goldenrod and woodland sunflower along with smaller amounts of sawtooth sunflower (H. grosseseratus), willow aster (Aster praealtus) and sometimes briars, ironweed, tall coreopsis, tall boneset, Joe Pye weed, and others. When there is some diversity mixed in, we selectively avoid cutting the less aggressive species. Following scything, the aggressive are likely to grow back, but perhaps the diverse other species will increasingly establish a competitive turf that will limit their aggressiveness and allow the establishment of the fire-fueling grasses or sedges. 

We will continue to scythe the rank vegetation for the next few years and try to get these areas burned more often. In small experiments, in previous years, we have found that repeated scything somewhat reduces tall goldenrod and dramatically reduces woodland sunflower. 

We can’t predict the results; this post just describes the current thinking about a challenge and our strategy to meet it. To ascertain how well it works on the ground may take years. We'll report back. If you’re doing this kind of work, we’d appreciate hearing your results.

Our goal, of course, is an ecosystem with sufficient diversity that it's self-sustainable with a minimum of our effort, beyond prescribed fire and deer control, which seem needed for the foreseeable future if these hundreds of plant species and thousands of animal species are to survive and thrive here. 

Endnotes

Endnote 1: Evolving Savanna Seed Mixes.

The seeds of our new savanna restoration seed mixes are divided by light intensity into four parts and labeled:

1. "Prairie (Savanna)" for full sun areas of the savanna

2. “Open Savana” for areas with bright sun for about two-thirds of the day

3. “Closed Savanna” – full or “bright dappled” sun for about one-third of the day

4. "Woodland (Savanna)" for areas of bright but mostly dappled light all day

These mixes are further divided into three wetness categories: mesic, wet-mesic, and wet. (Somme has little dry-mesic and no dry ground.) The term “mesic” means “medium wetness” – half way between “wet” and “dry.”

Our current list of mesic species and mixes is provided in Table 1. This table omits wet-mesic and wet lists, to make it easier to compare what's germane to the subject of this post, experiments in mesic savanna.

Note that the species of Somme's Open Savanna mix seem much like Prairie list but with the addition of some savanna specialist species like Penn sedge, hairy green sedge, purple milkweed, cream gentian, violet bush clover, savanna blazing star, etc.

Also note that these mixes reflect planting strategies and not community composition lists. For example: 1) We expect that big bluestem will be a part of our prairie and open savanna flora. But we do not include it in our prairie mixes, because it is likely to be too aggressive in early stages. On the other hand, we include it in our Open and Closed Savanna mixes because we're desperate for fuel there and because we notice that it's less aggressive there. 2) Some species on these mesic lists are more characteristic of wet-mesic areas, and we do put more seed of such species in the wet-mesic mixes, but if these species are sometimes in mesic, and we have enough seed gathered, we include them. 3) For many species, no one knows likely historic roles in savanna communities, because examples to study are slim to none. If our experience is that a species seems to reproduce in recovering mesic savanna communities, then we include it in that mix. If it becomes a part of the long-term community there, then we've learned something. If it later is outcompeted in a given community, perhaps its presence has in the meantime helped contribute niches and competition that has fostered the recovery of conservative, sustainable diversity.  

In the table below, C = Coefficient of Conservatism. MPS = the species planted in the mesic-prairie-like, most-open parts of the savanna. MOS = Mesic Open Savanna - the species planted in areas that receive full sun about two-thirds of the day. MCS = Mesic closed Savanna - the species planted in areas that receive full sun about one-third of the day. And MWS = seed for areas darker, yet still sunny, dappled shade all day, what might also be called open oak woodland. 

Revised Mesic Savanna Seed Mixes

C	SCIENTIFIC NAME	COMMON NAME	MPS	MOS	MCS	MWS
7	Actaea pachypoda	White Baneberry				X
10	Actaea rubra	Red Baneberry				X
8	Agalinis auriculata	Eared False Foxglove		X
3	Agalinis tenuifolia	Slender False Foxglove		X
5	Agastache nepetoides	Yellow Giant Hyssop			X	X
5	Agastache scrophulariifolia	Purple Giant Hyssop			X	X
5	Agrostis perennans	Thin Grass			X	X
7	Allium burdickii	Chicago Leek			X	X
3	Allium canadense	Wild Garlic		X	X	X
7	Allium cernuum	Nodding Wild Onion	X	X
9	Allium tricoccum	Wild Leek				X
10	Amorpha canescens	Lead Plant	X	X
5	Andropogon gerardii	Big Bluestem		X	X
8	Anemone cylindrica	Thimbleweed	X	X
7	Anemone quinquefolia	Wood Anemone				X
5	Anemone virginiana	Tall Anemone		X	X
7	Anemonella thalictroides	Rue Anemone			X	X
3	Antennaria plantaginifolia	Pussy Toes	X	X	X
5	Apocynum androsaemifolium	Spreading Dogbane		X
4	Apocynum cannabinum	Indian Hemp	X	X	X
6	Aquilegia canadensis	Wild Columbine		X	X	X
6	Arisaema dracontium	Green Dragon				X
5	Arisaema triphyllum	Jack-In-The-Pulpit				X
8	Arnoglossum atriplicifolium	Pale Indian Plantain		X
	Asclepias purpurascens	Purple Milkweed		X
8	Asclepias sullivantii	Prairie Milkweed	X
8	Asclepias tuberosa	Butterfly Weed	X	X
1	Asclepias verticillata	Whorled Milkweed	X	X
8	Astragalus canadensis	Canadian Milk Vetch	X	X
10	Aureolaria grandiflora var. pulchra	Yellow False Foxglove		X	X
8	Baptisia lactea	White Wild Indigo		X
10	Baptisia leucophaea	Cream Wild Indigo	X	X
5	Blephilia hirsuta	Wood Mint				X
9	Brachyelytrum erectum	Awned Wood Grass				X
10	Bromus kalmii	Prairie Brome	X	X
5	Bromus latiglumis	Ear-leaved Brome			X	X
5	Bromus nottowayanus	Glossy-leaved Brome			X	X
7	Camassia scilloides	Wild Hyacinth		X	X	X
4	Campanulastrum americanum	Tall Bellflower			X	X
5	Cardamine concatenata (Den lac)	Cut-leaved Toothwort				X
8	Carex albursina	Blunt-scaled Wood Sedge				X
8	Carex bicknellii	Prairie Oval Sedge	X	X
1	Carex blanda	Common Wood Sedge			X	X
5	Carex cephalophora	Oval-headed Sedge		X	X
4	Carex davisii	Awned Graceful Sedge			X
9	Carex festucacea	Fescue Oval Sedge		X
10	Carex formosa	Awnless Graceful Sedge			X	X
7	Carex gracillima	Graceful Sedge			X	X
3	Carex granularis	Pale Sedge		X	X
7	Carex grayi	Common Bur Sedge				X
3	Carex grisea	Wood Gray sedge				X
5	Carex hirsutella	Hairy Green Sedge		X	X
5	Carex hirtifolia	Hairy Wood Sedge			X	X
5	Carex jamesii	James’ Sedge			X	X
2	Carex molesta	Field Oval Sedge	X	X
5	Carex normalis	Spreading Oval Sedge			X	X
5	Carex pensylvanica	Penn Sedge		X	X	X
5	Carex rosea	Curly-styled Wood Sedge				X
6	Carex shortiana	Short’s Sedge			X
5	Carex sparganioides	Bur-reed Sedge			X	X
8	Carex sprengelii	Long-beaked Sedge			X	X
5	Carex swanii	Downy Green Sedge			X	X
7	Carex tribuloides	Awl-fruited Oval Sedge			X
10	Carex woodii	Wood’s Stiff Sedge				X
10	Castilleja coccinea	Scarlet Painted Cup	X	X
8	Caulophyllum thalictroides	Blue Cohosh				X
8	Ceanothus americanus	New Jersey Tea	X	X
4	Celastrus scandens	Climbing Bittersweet		X	X
4	Chenopodium simplex	Maple-leaved Goosefoot			X	X
5	Cinna arundinacea	Common Wood Reed			X	X
8	Cirsium altissimum	Tall Thistle			X	X
3	Cirsium discolor	Pasture Thistle	X	X
4	Claytonia virginica	Spring Beauty			X	X
4	Clematis virginiana	Virgin's Bower			X
9	Comandra umbellata	False Toadflax	X	X
10	Coreopsis palmata	Prairie Coreopsis	X	X
5	Coreopsis tripteris	Tall Coreopsis		X	X	X
4	Cryptotaenia canadensis	Honewort				X
7	Cuscuta glomerata	Rope Dodder		X
5	Cuscuta gronovii	Common Dodder		X
10	Dalea candida	White Prairie Clover	X	X
9	Dalea purpurea	Purple Prairie Clover	X	X
3	Danthonia spicata	Poverty Oat Grass		X	X
8	Dasistoma macrophylla	Mullein Foxglove		X	X	X
4	Desmodium canadense	Showy Tick Trefoil	X	X
10	Desmodium cuspidatum	Bracted Tick Trefoil		X	X
7	Desmodium glutinosum	Pointed Tick Trefoil			X	X
5	Desmodium perplexum	Smooth Tick Trefoil		X	X
6	Desmodium paniculatum	Panicled Tick Trefoil		X	X
4	Dichanthelium acuminatum (Pan imp)	Old-Field Panic Grass		X
8	Dichanthelium latifolium	Broad-Leaved Panic Grass			X	X
10	Dichanthelium leibergii	Prairie Panic Grass	X	X
5	Dioscorea villosa	Wild Yam				X
6	Dodecatheon meadia	Shooting Star	X	X	X	X
10	Drymocallis arguta	Prairie Cinquefoil	X	X
4	Elymus canadensis	Canada Wild Rye		X
8	(Elymus) Roegneria subsecunda	Bearded Wild Rye	X	X
7	Elymus villosus	Silky Wild Rye			X	X
3	Epilobium coloratum	Cinnamon Willow Herb				X
8	Erigeron pulchellus	Robin's Plantain			X
9	Eryngium yuccifolium	Rattlesnake Master	X	X	X
5	Erythronium albidum	Trout Lily				X
4	Euphorbia corollata	Flowering Spurge	X	X
10	Eurybia furcata	Forked Aster			X	X
9	Eurybia macrophylla	Big-leaf Aster				X
3	Eutrochium purpureum	Purple Joe Pye Weed			X	X
5	Festuca subverticillata	Nodding Fescue				X
5	Galium circaezans hypomalacum	Hairy Wild Licorice			X	X
2	Gaura longiflora	Common Gaura		X
9	Gentiana alba	Yellowish Gentian		X	X
10	Gentiana puberulenta	Prairie Gentian	X	X
8	Gentianella quinquefolia occidentalis	Stiff Gentian	X	X
4	Geranium bicknellii	Northern Cranesbill		X
1	Geranium carolinianum	Carolina Cranesbill		X
5	Geranium maculatum	Wild Geranium			X	X
4	Hedeoma pulegioides	American Pennyroyal				X
9	Helianthus pauciflorus	Prairie Sunflower	X	X
7	Heliopsis helianthoides	False Sunflower		X	X
5	Heracleum maximum	Cow Parsnip			X	X
8	Hesperostipa spartea	Porcupine Grass	X	X
10	Heuchera richardsonii	Prairie Alum Root	X	X	X	X
5	Hydrophyllum virginianum	Virginia Waterleaf				X
7	Hylodesmum glutinosum	Pointed Tick Trefoil			X	X
4	Hypericum punctatum	Spotted St. John’s Wwort		X	X
8	Hypoxis hirsuta	Yellow Stargrass	X	X
5	Hystrix patula	Bottlebrush Grass			X	X
8	Koeleria macrantha	June Grass	X	X
9	Krigia biflora	False Dandelion	X	X
1	Lactuca canadensis	Wild Lettuce				X
8	Lactuca floridana	Wood Lettuce			X	X
10	Lathyrus ochroleucus	Cream Vetchling			X	X
8	Lathyrus venosus	Veiny Pea		X	X
5	Leersia virginica	White Grass				X
4	Lespedeza capitata	Round-Headed Bush Clover	X	X
10	Lespedeza frutescens	Violet Bush Clover		X	X
8	Liatris aspera	Rough Blazing Star	X	X
9	Liatris scariosa var. nieuwlandii	Savanna Blazing Star		X	X
7	Liatris spicata	Marsh Gay Feather	X	X
10	Lilium philadelphicum var. andinum	Prairie Lily	X	X
6	Liparis liliifolia	Purple Twayblade			X
8	Lithospermum canescens	Hoary Puccoon	X	X
10	Lithospermum latifolium	American Gromwell				X
5	Lobelia inflata	Indian Tobacco			X	X
4	Lobelia spicata	Pale-spiked Lobelia	X	X
7	Lonicera reticulata (prolifera)	Yellow Honeysuckle		X	X
7	Luzula multiflora	Wood Rush			X	X
4	Monarda fistulosa	Wild Bergamot	X	X	X
8	Oenothera perennis	Small Sundrops		X	X
10	Oenothera pilosella	Prairie Sundrops	X	X
3	Oligoneuron rigidum	Stiff Goldenrod	X	X	X
4	Osmorhiza claytonii	Hairy Sweet Cicely				X
8	Oxalis violacea	Violet Wood Sorrel	X	X	X
8	Oxypolis rigidior	Cowbane			X	X
4	Packera paupercula	Balsam ragwort			X
8	Parthenium integrifolium	Wild Quinine	X	X	X
9	Pedicularis canadensis	Wood Betony	X	X	X	X
9	Pedicularis canadensis var. rubrum	Wood Betony (Red)				X
4	Penstemon digitalis	Foxglove Beardtongue			X	X
8	Perideridia americana	Thicket Parsley		X	X	X
5	Phlox divaricata	Woodland Phlox				X
9	Phlox glaberrima var. interior	Marsh Phlox	X	X
8	Phlox pilosa var. fulgida	Prairie Phlox	X	X
6	Phryma leptostachya	Lopseed				X
4	Physostegia praemorsa	Fall Obedient Plant	X	X
8	Polemonium reptans	Jacob’s Ladder			X	X
10	Polygala senega	Seneca Snakeroot	X	X
4	Polygonatum biflorum	Smooth Soloman's Seal		X	X	X
5	Prenanthes alba	White Lettuce			X	X
7	Prenanthes altissima	Tall White Lettuce				X
10	Prenanthes aspera	Rough Prairie Lettuce	X	X
10	Prenanthes racemosa	Smooth Prairie Lettuce	X	X
7	Pycnanthemum tenuifolium	Narrow-leaved Mountain Mint	X	X
1	Ranunculus abortivus	Kidney-leaf Buttercup				X
4	Ratibida pinnata	Yellow Coneflower	X	X
5	Rosa carolina	Pasture Rose	X	X
5	Rosa setigera	Illinois Rose		X	X
1	Rudbeckia hirta	Black-eyed Susan	X	X
8	Rudbeckia subtomentosa	Sweet Black-eyed Susan		X	X
1	Rudbeckia triloba	Brown-Eyed Susan		X	X	X
4	Sambucus canadensis	Elderberry			X
5	Sanguinaria canadensis	Bloodroot			X	X
5	Sanicula marilandica	Black Snakeroot		X
5	Schizachyrium scoparium	Little Bluestem	X	X	X
4	Scrophularia marilandica	Late Figwort			X	X
7	Scutellaria leonardii	Prairie Skullcap	X	X
9	Scutellaria ovata	Heart-leaved Skullcap				X
6	Silene stellata	Starry Campion			X	X
10	Silene virginica	Fire Pink				X
5	Silphium integrifolium var. deamii	Deam's Rosin Weed	X	X	X
5	Silphium laciniatum	Compass Plant	X	X
5	Silphium terebinthinaceum	Prairie Dock	X	X	X
6	Sisyrinchium albidum	White-eyed Grass	X	X
5	Sisyrinchium angustifolium	Stout Blue-Eyed Grass		X	X
5	Smilacina racemosa	Feathery False Solomon's Seal		X	X	X
5	Smilacina stellata	Starry False Solomon's Seal	X	X
5	Smilax ecirrhata	Upright Carrion Flower			X	X
5	Smilax lasioneura	Common Carrion Flower		X	X	X
8	Solidago caesia	Blue-Stemmed Goldenrod				X
7	Solidago flexicaulis	Zig-zag Goldenrod				X
3	Solidago juncea	Early Goldenrod	X	X	X
3	Solidago nemoralis	Old Field Goldenrod	X	X	X
8	Solidago speciosa	Showy Goldenrod		X	X	X
5	Solidago ulmifolia	Elm-Leaved Goldenrod			X	X
5	Sorghastrum nutans	Indian Grass	X	X	X
3	Sphenopholis intermedia	Slender Wedge Grass		X	X	X
8	Spiranthes magnicamporum	Great Plains Ladies Tresses	X
10	Sporobolus heterolepis	Prairie Dropseed	X	X	X
6	Symphyotrichum ericoides	Heath Aster	X	X
9	Symphyotrichum laeve	Smooth Blue Aster	X	X
8	Symphyotrichum oolentangiense	Sky-Blue Aster	X	X
7	Symphyotrichum shortii	Short's Aster				X
5	Symphyotrichum urophyllum	Arrow-leaved Aster				X
10	Taenidia integerrima	Yellow Pimpernel			X	X
7	Thalictrum dioicum	Early Meadow Rue				X
8	Thaspium trifoliatum	Meadow Parsnip		X	X
3	Tradescantia ohiensis	Common Spiderwort		X	X	X
9	Trillium grandiflorum	Large-Flowered Trillium				X
5	Trillium recurvatum	Prairie Trillium			X	X
5	Triosteum aurantiacum	Early Horse Gentian		X	X
4	Triosteum perfoliatum	Late Horse Gentian		X	X	X
1	Turritis glabra	Tower Mustard		X	X
7	Uvularia grandiflora	Bellwort				X
2	Verbena urticifolia	Hairy White Vervain		X	X
4	Vernonia missurica	Missouri Ironweed		X	X
8	Veronicastrum virginicum	Culver's Root	X	X	X
4	Vicia americana	American Vetch		X	X
5	Viola labradorica	Dog Violet			X	X
5	Viola pubescens	Downy Yellow Violet				X
10	Zizia aptera	Heart-Leaved Alexanders	X	X
5	Zizia aurea	Golden Alexanders	X	X	X	X

Seeds for additional specialized seeding experiments are not reflected above. This list reflects the regular mesic seed mixes. 

Endnote 2: Open Woods Flora

Some rich bur oak woodland areas increasingly thrive at Some Prairie Grove. Not just spring flora – throughout the summer and fall uncommon, conservative wildflowers, grasses, and sedges also rise and bloom.

Following the typical early spring flora including trilliums, toothworts, wood betonies, and shooting stars, the later spring woods (above) is thick with wild hyacinth, golden Alexanders, wild geraniums and a long list that includes what had been thought of as "typical" prairie and woodland species.  

In close-up (below), notice the growing abundance of woodland sunflower (large paired leaves). In some areas, that species comes to largely replace all other vegetation.

 

In fall the bur oak woods are dense with zigzag and blue-stemmed goldenrods (in flower) and woodland puccoon (in seed) as shown above. Other frequent species include awned woodgrass (Brachyelytrum), big-leaf aster, and Short's aster. 

 

Endnote 3: Too Impatient?

Why not just wait and see if success comes in time? Perhaps that’s all it takes, but two concerns convinced us to try new approaches. 

First, we have apparently wasted much precious seed by broadcasting it into areas where goldenrod or sunflower shade would doom it to fail. Either a) we should save that precious conservative seed for a later stage in succession/restoration, or b) we should use some improved management for these areas so that the thugs don’t win out, or c) we should revise our seed mixes for a solar-powered solution that depends on more-competitive, diversity-promoting, early-stage species. 

Second, we need to face the fact that true ecosystem recovery proceeds with a grand slowness. Perhaps we would be showing more apt humility if we showed more patience. Yet, our oak woods and savanna plants and animals are rapidly dying – losing their extent, species populations, and, probably, specialized genetic alleles. Thus, we are motivated to speed up our learning. It’s easy to restore a mix of native plants that looks good for the short term. But for how long? 

Thus we struggle with the big questions. Darwin’s most important insights may have centered around his appreciation of time: his recognition of how long it took for earthworms to transform soils or, of course, for evolution to function. To paraphrase Darwin, “There is a grandeur in this view of restoration.” We biodiversity conservationists have rightly assumed that the conservation or rare natural remnants should take first priority. We have suspected that those areas are all we will ever have ... that larger areas to high quality cannot be restored. But perhaps human facilitation can restore natural ecosystem processes, and then nature will restore itself.  

Endnote 4 - Somme Prairie Grove's "Woodland" Areas

In groves outlined in green on the map below, judging from the 1839 Public Land Survey, widely scattered bur oaks grew - in other words, savanna. Groves outlined in orange are today mostly white oaks, planted by Forest Preserve staff in the mid 1900s. Today these groves have a "woodland" rather than savanna density. All the heavily wooded areas are "unnatural" results of fire suppression and/or planting. The planted groves, when restoration began, included birch, pine, black locust, honey locust, basswood, silver poplar, and many oaks species (white, swamp white, post, chinkapin, rock chestnut and hybrids). Over the years, cutting, fire, and competition have removed most such trees except the white oaks. 

Since fallen oak leaves facilitated fire and we planted woodland grass and wildflower seed here, both the bur and white oak areas today have a fairly diverse and conservative woodland flora. Since such areas were not problems and gradually improving (unlike the "intermediate" areas), we’ve otherwise paid them little attention, except to pull garlic mustard and more recently as a seed source for woodland areas east of Waukegan Road. But the original natural community and current conservation priority here is savanna. The original oaks of this site are bur and Hill's. The Hill's (also called scarlet) oaks were likely originally mostly shrubby re-sprouts. The consensus biodiversity conservation goal at Somme Prairie Grove is to expand savanna areas to take advantage of savanna soil biota and to expand the populations of savanna-dependent species. (We restore original prairie along the western edge.) Thus, in the savanna areas we thin white oaks, plant bur oaks, burn, seed and scythe. Some former planted oak areas now have sufficiently open tree canopies that they're mapped as savanna. 

On the edge of groves here and there are diverse, healthy 'intermediate" areas. An example is shown below:

Species in the photo include spreading dogbane (in bloom), early goldenrod (about to bloom), big bluestem, woodland sunflower, Hill's oak, heath aster, and wild quinine (leaf in bottom right corner). Although diverse relative-stability seems to take a long time, we seem to be noticing that in some areas of conservative diversity, the woodland sunflower becomes just another species in that diversity. 

Also in the photo above, notice Asian honeysuckle (brown leaves in top left corner). Those dead leaves had been foliar-sprayed with herbicide last spring by stewards led by the dedicated Eriko Kojima. Somme's regularly burned areas included thousands of small, annually or biennially re-sprouting buckthorns and honeysuckles. We'd hoped fire and competition would eventually eliminate them. Here again, we decided that 40 years of patience had been enough. Eriko and colleagues have now largely eliminated them from large areas.   

Endnote 5 - Restoring Bur Oaks

In the high-quality open, grassy areas shown above, scores of young bur and scarlet oaks thrive among open savanna vegetation. In "woodland or "intermediate areas, we don't get bur oak reproduction. Apparently the the shade of the existing trees or aggressive forbs ("wildflowers") is too great for the sun-loving young oaks. The oaks reproduce in rich grassland, but here they don't get any bigger than this. Yes, they thrive in the years between fires. But their trunks burn back to the ground with every burn (and then the deer diminish them further by eating off spring re-sprouts).

The photo below features one oak we have protected from fire (and deer) for a few years: 

This heroic little oak (like scores of others in the "protection program") may now be on its way to becoming a true savanna oak. So far, so good. But, in the foreground, an expanding patch of woodland sunflower has blotted out most other species, especially the grasses and sedges. As a result, the foreground area does not burn when the grassland behind it does. We have now begun to scythe the sunflowers of course. Restoring "the Intermediate" seems to require different work. We're happy to be working to figuring it out. 

Other Notes, Tidbits, and Questions

Old Pasture Plantings

Here’s a warning - if you are initiating a remnant restoration effort in an old pasture: You'll need both late fire and seed. The normal component of Eurasian grasses – timothy, bluegrass, redtop, smooth brome, orchard grass, etc. – will burn well enough to initiate restoration. In late spring, such fires will be hot enough to inhibit these grasses’ growth, reduce some weed species, and promote establishment of seeded savanna species. In other words, such a fire can promote establishment of warm season grasses and other species crucial to the restoration. But that will happen only if you provide enough seed, preferably broadcast the fall before the fire. With enough seed in early years, the natural ecosystem will win out. Without it, the Eurasian cool-season grasses will die out and be replaced by tall goldenrod and its ilk, and then it will no longer burn. Woody brush may gradually take over, and you’ll have lost your chance for early success.

Somme Burn Strategy

We try to burn a bit more than half the site every year. All that is readily flammable thus burns on average every second year. Of course, these efforts are foiled in the Obstinate Intermediate, because on most years it just won’t burn. (See maps below.) The fires burn merrily through the oak leaves of the groves and roar through the tall “prairie grass” areas; then they fizzle out on reaching the Depauperate Intermediate. Such problem areas divide our two burnable types from each other. Burn crews typically need to ignite oak leaf areas and "prairie grass" areas separately. 

Three maps below show results when we attempted to burn the north half of the preserve. In the first two maps, the unburned areas are sketched with brown lines. 

2001 fire map

2006 fire map

2008 fire map

Fire behavior may have played a major role in the development of our "Intermediate" areas. (It seems likely that the more intense fires of previous times would, in the long history of evolution, have produced different results.) The overall quality map, below, seems to reflect the fire patterns, recorded above.  

The Meaningless and the Stochastic

We note changes, but are they meaningful to us? Multiple changes from multiple forces beset every little area. In the photo below, tall goldenrod (not in bloom yet) seems to be giving way to (now blooming) saw-tooth sunflower (upper left center),  woodland sunflower (upper right), and purple Joe Pye weed. Varied amounts of shade come from hawthorns (upper left) and white oak (top), planted in early Forest Preserve days. An ephemeral stream slices through (in the photo) the mid-right edge to bottom-middle edge. Wet-mesic species along the stream blend into mesic a few feet back. In various parts of this little area, shade is rapidly decreasing from time to time (as we stewards cut trees) and elsewhere increasing (as remaining trees grow). Fire frequency here is reduced by stream and wet-mesic areas - but differently depending on wind direction during each fire. If we observe changes in an area like this, we are kidding ourselves if we think we can attribute them to one or a few causes. For now, the best we think we can do is to measure changes and develop management strategies based on less complex areas.

 

A Wet-mesic Example

The photo below is a close-up from among saw-tooth sunflower in this same area: such conservatives as yellow star grass (Hypoxis hirsuta) and violet wood sorrel (Oxalis violacea) thrive densely under that aggressive. Will increasing diversity reduce the dominance by the sunflower? Or will the shade of the sunflower kill off the conservatives? Does the answer depend on management regimes? Time will tell, if we do the right studies. Or, as an alternative, perhaps the site-wide random sampling we do every four years will continue to show overall improvement. In that case, we'll continue with our management strategy; we won't really know what's causing what; but we'll have some confidence that preserve floristic quality is on the rise. 

Control of Aggressive Species by Parasites 

In the above photo, a parasitic morning glory called dodder is reducing the dominance of mountain mint and tall goldenrod. We've spread the seed of this native parasite (Cuscuta gronovii) in recent years, as this strategy has seemed promising, but we have no data on it, only our impressions. 

Results of a Preliminary Scything Experiment

The photo above shows the corner of a rectangular patch in Vestal Grove where woodland sunflower has been scythed once a year since 2017. The scything was done when the sunflower was nearly full height. The photo was taken when the sunflower outside the scythe area was in bloom. After the first year, there were only a few short sunflower stems to scythe. The remaining vegetation in the scythed area is not dramatically different from the vegetation under the sunflower. Trees are gradually being thinned in this heavily shaded area. We hope that ongoing monitoring of this and many experimental areas will lead to helpful data and analysis in time.  

Acknowledgements

Who did the on-the-ground work on which these results are based? For the first 15 years, the answer was “almost entirely the volunteers of the North Branch Prairie Project.” We got strategic advice from Dr. Betz, Forest Preserve and Nature Conservancy professionals, and others. For the past two decades, Forest preserve staff and contractors played an increasing role, especially for prescribed burns and large tree removal. Thus, when this post says, “We did X,” the implication is that “Some combination of volunteers and professionals did X.”

Seed harvest in recent years has been led by Eriko Kojima (planning and harvest) and Jim Hensel (seed prep). Seed mix revisions were led by Sai Ramakrishna with input from Eriko Kojima, Christos Economou, and Katie Kucera. 

Monitoring is being done by Sai Ramakrishna, Eriko Kojima, Karen Glennemeier, Lisa Musgrave, Matt Evans and many others. 

Revised site map and computer coaching by Linda Masters is much appreciated. 

Editing and proofing credits for this post go to Christos Economou and Eriko Kojima.