Oddballs or Innovators?

I spotted an upland sandpiper on top of a power pole last week.  In central Nebraska, that’s not really noteworthy – upland sandpipers are pretty common across much of the state.  They tend to nest in large open grasslands with short vegetation structure, and Nebraska has an abundance of that kind of habitat.  This particular sandpiper, however, was perched on a pole surrounded by what looked to be miles of contiguous cropland.  Seeing the sandpiper in that context got me thinking about how conservation scientists deal with patterns in data and, more particularly, the outliers that don’t fit those patterns.

This is not the upland sandpiper I saw surrounded by cornfields, but another one who was living where he was "supposed" to be living - in big open grasslands near Norden, Nebraska.

This is not the upland sandpiper I saw surrounded by cornfields, but another one who was living where he was “supposed” to be living – in big open grasslands near Norden, Nebraska.

My graduate research focused on grassland birds in fragmented prairies.  I categorized bird species by the size of prairie they tended to nest in.  Dickcissels and red-winged blackbirds seemed comfortable in really small prairies, grasshopper sparrows wanted a little more space, and bobolinks and upland sandpipers were usually in large prairies.  Now and then, of course, we’d find a bird in a prairie much smaller than it was “supposed” to be in.  An outlier.  I included those outliers in the data, and their behavior was averaged in with all the other sightings, but I treated them as an anomaly – not something important.  I wonder now if that was the right perspective.

As an ecologist, I see anomalies all the time.  Behaviors of plants or animals that don’t fit what I know – or think – to be the broad pattern of behavior of their species.  For example, during the spring migration of sandhill cranes, we tell visitors that cranes prefer to hang out in harvested fields or open treeless grasslands with short vegetation structure, but now and then we see a group of cranes feeding in tall grass beneath a grove of trees.  Plants can be surprising too.  Entire-leaf rosinweed (Silphium integrifolium) and Canada milkvetch (Astragalus canadensis) typically grow in lowland sites in our Platte River Prairies, but occasionally, some individuals will establish on top of a sandy ridge.  As a third example, I pay close attention to what plant species cattle graze in our prairies.  Forage selection varies by season, but there are some plant species cattle just don’t like to eat – except now and then when I find a clearly-grazed patch of Canada goldenrod, tall dropseed, or some other plant cattle “don’t like”.

It’s easy to dismiss those odd observations as unimportant results of unique circumstances.  Maybe cranes sometimes find a food source so fantastic it overrides their discomfort with tall vegetation.  Rosinweed and milkvetch plants might colonize dry sandy areas because of a lack of competition, but they might not survive for long.  And who knows why cattle do what they do sometimes…?

We usually see rosinweed in lowland areas of our prairies, surrounded by other lowland tallgrass prairie plants.

We usually see rosinweed in lowland areas of our prairies, surrounded by other lowland tallgrass prairie plants.

An agronomist friend of mine has shown me photographs of upland sandpiper nests in crop fields he works with.  It’s not an unheard of phenomenon, but it’s not representative of how most upland sandpipers act.  The birds that nest in those crop fields may be birds that were less able to defend territories in more suitable habitat.  Alternatively, maybe those birds are pioneers, forging a new path for the survival of the species!

Rather than dismissing anomalies, maybe we should be pursuing them with as much energy as we spend looking for patterns.  In this rapidly changing world, individual plants and animals that can survive where others can’t might just hold the key to conservation success.  Maybe those individuals are adapting to conditions in ways others of their species haven’t.  If upland sandpipers could figure out how to nest successfully in crop fields, for example, that would open up a great deal of nesting habitat for a species that has largely disappeared from large areas of North America.  If rosinweed can adapt to a wider range of habitat types, that might be a pretty important strategy for its survival in the face of a rapidly changing climate.  Should we be looking harder for ways to identify and facilitate that kind of adaptation?

It’s a big, beautiful, complex world out there.  It’s tempting to categorize everything we see into tidy little bundles to and simplify that complexity.  Oddballs can make life difficult, after all.  On the other hand, Nikola Tesla, John Lennon, and Steve Jobs were pretty odd, but turned out to have pretty good ideas in the end.

Maybe outliers are noteworthy after all…

Report from the 2011 Grassland Restoration Network – Part 2: Grassland Birds

The Grassland Restoration Network’s 2011 annual meeting was at The Nature Conservancy’s Kankakee Sands prairie/wetland restoration site in Indiana.  This year, we focused more than usual on creating habitat for various animal taxa, and I previously reported on the herpetology portion of the meeting.  Here is a summary of the discussions we had regarding grassland birds.

While the diversity of insects in a prairie is strongly tied to plant diversity, grassland birds have no such relationship.  The species richness, abundance, and breeding success of grassland birds are mainly related to variables such as vegetation structure, habitat patch size, and the amount of grassland in the neighborhood around their nesting sites.  At Kankakee Sands, the 6,000 acres of restored prairie/wetland habitat are obviously increasing both the patch size and total amount of grassland in the neighborhood, so our discussions focused on vegetation structure.  The ideal situation, of course, is to have a diversity of vegetation structure scattered across the site so that all grassland bird species can find the habitat they require for nesting.

This kind of tall vegetation is typical of most of the restored prairie at Kankakee Sands (and most other tallgrass prairie restoration sites). While valuable for many species, it's not much good for grasshopper sparrows and other animals that require short or patchy habitat structure.

We talked mainly about two different approaches to creating habitat structure in order to accomodate a rich variety of bird species.  The first approach is to design seed mixes for each desired habitat structure type, and the second is to design seed mixes that promote overall biological diversity, and then manage for structure with fire, grazing, and other tools.  Kankakee Sands is employing both strategies, so we were able to look up close at each.

We first looked at a restored prairie that had been seeded with about 60 species of short and medium-height plant species, and that included bunchgrasses like dropseeds and little bluestem instead of tall and strongly rhizomatous species like big bluestem.  The resulting structure was very favorable for grasshopper sparrows and other species that prefer that short to mid-height structure.  The seeding was nearly a decade old and seemed to be maintaining its structure and species composition with only a couple of prescribed fires as management.  The short patchy vegetation is definitely a contrast to the majority of other seedings at Kankakee Sands (and most other tallgrass prairie restoration sites) which are mainly tall and rank.  Those taller sites provide excellent habitat for Henslow’s sparrows and species with similar habitat preferences, but don’t do much for grasshopper sparrows.

Chip O'Leary (left) describes grassland bird habitat and research results with participants of the 2011 Grassland Restoration Network workshop. This restored prairie was seeded with short and medium-height plant species (including bunchgrasses instead of tall and strongly rhizomatous species).

The second example we visited was one of the first seedings done at Kankakee Sands (in the late 1990’s).  From the beginning, the several hundred acre prairie has been dominated by grasses and has been low on forb species diversity.  For the last several years, the Conservancy has been experimenting with patch-burn grazing as a way to create more heterogenous vegetation structure and to increase forb diversity in this prairie.  To date, forb diversity has neither increased nor decreased, but habitat structure has certainly changed.  The recently burned (and thus currently grazed) portions of the prairie provided excellent grasshopper sparrow habitat, while other portions were more tall and rank.  Though no change has been detected in forb abundance or diversity, the staff has noticed that a few forb species seem to bloom less abundantly than in the past – including compass plant, prairie dock, and Canada milkvetch.  Because of that, we discussed the value of fencing out a significant portion of the prairie each year to ensure that those species were given a complete break from grazing pressure periodically.

The two methods of creating bird habitat (seeding design vs. active management) both seem to be working well so far.  Both the prairie seeded with short/bunchy vegetation and the grazed prairie had significantly different vegetation structure than did the majority of the tall rank prairies around them – and birds are responding to that structure.  However, there are still plenty of questions about the long-term future of both approaches to creating bird habitat. 

In terms of the seeding design approach, one potential downfall is that the site was seeded with considerably fewer plant species than most other tallgrass prairie seedings at Kankakee Sands.  The potential effect of this lower diversity on insects and other species is unknown.  In addition, planting short and medium height plant species in soil/climate conditions that typically favor tall species could result in a relatively unstable prairie community.  In the long term, the fact that those shorter plant species aren’t using all of the available light/soil/moisture resources could lead to encroachment by either tall grasses (defeating the purpose of the design) or invasive species (which create obvious problems).  The planting we looked at was located on dry sandy soils with very low organic matter, so it probably is less at risk for that kind of instability than if it had been located in wetter or heavier soils. 

A final potential disadvantage of the seeding design approach is that the location of the short/medium habitat structure is static.  Grazing and other management tools for manipulating structure can be moved around a site from year to year, creating a shifting mosaic of habitat conditions.  That kind of mobility could help keep predator or pathogen populations from building up under consistently favorable conditions at any one site (this is speculation).  In addition, the staff will have to be careful to avoid repetitive management treatments aimed at maintaining the same structure year after year – that management could consistently favor some plant species over others, further reducing the plant (and insect?) diversity of the prairie.

There are plenty of concerns about the patch-burn grazing strategy as well.  To date, the plant diversity in the grazed prairie we looked at has not gone down, but neither has it increased – though increasing plant diversity in a grass-dominated prairie is very difficult with any strategy.  Because the prairie started with few forbs, it’s hard to know what the impact of patch-burn grazing would be on a more diverse plant community, but that needs investigating.  We discussed the possibility that a higher stocking rate and the addition of a large exclosure that changes location each year could help with both habitat and plant diversity over time. 

The burned patch of the patch-burn grazed prairie at Kankakee Sands. While the compass plant in this photo is blooming under grazing, other individuals (of compass plant, prairie dock, Canada milkvetch, and others) appear to be blooming less frequently than they did prior to the introduction of cattle. While this doesn't kill the plants over the short-term, it is a concern down the road, and the Kankakee Sands staff is considering strategies to mitigate those potential impacts.

A higher stocking rate would lead to more intense grazing of the dominant grasses such as big bluestem that are likely preventing existing forbs from becoming more abundant.  Currently many of the grass plants inside the most-recently burned patch are only being moderately grazed, and that can actually induce those plants to divert extra resources into rhizome production – leading them to expand their footprint (not the objective here).  More intense grazing on those grasses could create better opportunities for seed germination and seedling establishment around those plants, and would create even shorter vegetation structure, which might help attract upland sandpipers and as well as grasshopper sparrows. 

Regardless of stocking rate, the use of a grazing exclosure would help ensure a periodic break from grazing for plant species that otherwise be vulnerable to annual grazing – even in the unburned (and lightly grazed) portions of the prairie.  With a higher stocking rate, the exclosure would become even more important.  In addition to protecting plants from grazing, it would also protect fuel for the next year’s burn.  An exclosure roughly 1/4 or 1/3 of the size of the prairie would probably sufficient for protecting both plants and fuel, and would still leave cattle access to both burned and unburned portions of the prairie – something is important when promoting selective grazing. 

Even with those potential alterations to the current patch-burn grazing system, there are still plenty of unknowns about the long-term impacts of cattle grazing at Kankakee Sands.  It seems clear that grazing can create a shifting mosaic of habitat structure, but whether or not it can maintain the kind of plant diversity (and other diversity) desired by the Conservancy at this site is still an open question.

Discussing those kinds of questions while standing on the ground, however, is the best part of our Grassland Restoration Network workshops.  We don’t all agree on the best strategies, because we are all still experimenting with our own ideas on our own sites – and none of us feel like we have all the answers.  Being able to see for ourselves what various restoration and management treatment results look like helps us better compare those results to what we see on our own sites.  While we don’t have all the answers yet, we’re certainly moving much closer to them as a group than we would be individually.