RiVR: Reference data for riparian forests


Since I haven’t yet published my thesis, this post is where it will live for those curious about the reference method I developed as the project for my Master of Forestry.

I’ve always been fascinated by the spatial structure of natural forests. I grew interested in reference information while at Urban Creeks Council. I wondered why the planting palette for every restoration project looked the same, and whether that was reflected out there in the undisturbed ecosystems we were attempting to emulate. Other types of reference information, particularly geomorphic data, is used in every stream restoration project, so why didn’t we have quantitative data for the ecosystems we were attempting to restore? No one had ever looked at that type of data before. As it turned out, the reference systems I studied bore very little resemblance to the restoration projects we were building.

RiVR, as I named it- the Riparian Vegetation Reference index- quantifies the species assemblages in riparian forests. It’s so specific, in fact, that it can tell us the exact number of species and individuals that we should be aiming for in restoration. While other reference programs like CRAM and SWAMP can give us (in CRAM’s case) general vegetation data and a site score, and (in SWAMP’s case) aquatic data by which the measure the effectiveness of restoration work, RiVR tells us exactly which species should be used on a restoration site and how numerous they should be.

One of the most important discoveries that came out of the project was the concept of relative species abundance- that is, how numerous the species on a given site are in relation to one another. This is a basic biological concept, but it had never been applied to riparian ecosystems before. My data indicate that there are typically three to four dominant species in a given assemblage (e.g., on a typical 100-meter site) and all other species- eighty-two percent of all the species I detected in all of my reference sites- were actually quite rare on their sites. I call these minor species. As rare as these species were on their reference sites, their universal prevalence on these sites suggest their critical importance to their ecosystems.


So, in contrast to the “plant only what survives” approach used by many restorationists, we can’t in good conscience call what we are doing restoration if we’re not restoring 82 percent of the species found in our reference system. And I only studied woody species, so imagine what the diversity would look like if

So the takeaways are, essentially:

  • Riparian forests have specific structural attributes that can be measured and replicated.
  • Relative abundance is an important component of these systems.
  • Riparian forests are highly diverse: reference sites average 15 species of trees and shrubs in the space of only about an acre.
  • Riparian forests are composed of different assemblages that shift along the elevational gradient of watersheds, depending on substrate, geomorphic conditions, and probably also simple chance.
  • These assemblages, while consisting of different species, follow predictable biological patterns.
  • Reference sites- preferably multiple ones- should be used in restoration to help us do restoration more effectively.

For the curious, here’s a short presentation on the study design, results, and potential applications. I hope to get around to actually publishing one of these days!


Sorry about the inelegant link: I promise it works. I’m always here, so if you’d like to learn more about RiVR, or use it in a project, email me at river.why.ecology -at- gmail and I’ll be happy to respond. You can also contact me through the Urban Creeks Council Facebook page.