The data for this paper are from the experiment described in @Flory2009. They put these data in the context of how roads and forest successional age may affect plant invasions. They reported the data in aggregate, i.e., grouping the three native species and three non-native species.
In this paper we want to provide the breakdown by species, which will provide insight into the (dis)similarity in species responses.
Seedlings of six species (3 invaders, 3 natives) were planted at 12 forest sites classified as mature (>60 years) or young (<40 years). A 60-meter long transect perpendicular to the road was established at each site. We planted four seedlings of each species at distances of 0, 5, 10, 20, 40, and 60 meters from the roadway edge. The 24 seedlings were planted in a 8 x 3 grid with the long edge parallel to the road at each distance, with at least four meters between plants.
-
Are edge effects more prominent in young vs. mature forests, or is there no difference?
-
Are differences in species reponses expected based on their natural history?
-
What are differences in performance of native species vs. non-native species?
In @Flory2009 analysis of these data aggregated into native vs. non-native grouping they found that 1. germination of non-native species was overall 3x greater than native species, with the difference larger in mature vs. young forests. Germination was also greater overall in mature forests compared to young forests. 2. Native seedlings had greater survival in mature forests, while non-native seedlings had greater survival in young forests. 3. All seedlings were taller nearer to the road across both forest ages, but taller in young forests where non-native seedlings were also relatively taller than native seedlings. 4. Basal area and DBH was larger in mature vs. young forests. Non-native species were also larger than native seedlings. 5. Seedling biomass was greater nearer to roads for both native and non-native species, and especially greater in young forests. Non-native species produced more (aboveground) biomass than native species.
@Harper2005 wrote a review on edge effects ("edge influence") on forest structure. They suggest a number of definitions, including:
-
Edge effect ("edge influence") as the effect of processes at the edge resulting in detectable differences in composition, structure, or function near the edge compared to the ecosystem on either side of the edge.
-
The magnitude of edge effects:
$(e-i)/(e+i)$ where e is the parameter value in the "edge" and i is the parameter value in the "interior." This approach requires defining the edge vs. the interior. -
The distance of an edge effect as the set of distances from the edge over which there is a statistically significant effect
They also distinguish between primary and secondary responses to edge creation. Primary include changes in productivity, nutrient cycling, evapotranspiration (processes), and canopy cover, tree density, biomass (structural). Secondary responses include recruitment, growth, mortality, reproduction (processes), and sapling density, understory cover (structural), and composition.
-
Based on these definitions, we are experimentally testing secondary process responses of potential recruitment (germination), growth, and mortality.
-
I don't know if we can make an argument to distinguish between the age of the edge in terms of creation. It's likely that the roads were already there and the forests have grown up with this defined edge.
- In this case, edges might be more similar between the forest ages than other locations along the transects.
We are working with data from "maintained" edges, i.e. roadways. These edges often develop a "sidewall" of dense vegetation/side canopy, potentially reducing the depth (distance) of edge effects. The "edge zone" may become deeply shaded. In eastern North American forests the magnitude of edge effects was slightly higher for maintained vs. regenerating edges. Overall, older maintained edges may typically have large magnitude and short distance, whereas older regenerating edges may more frequently have smaller magnitude and short distance.
- Not sure if this information has been updated in the last few years with a meta-analysis...
Geography and topography can have substantial influence on magnitude and distance of edge effects. Also, open edges will have greater distance of edge effects than closed/sealed edges.
- We don't have these data. While we could potentially obtain it through remote sensing/GIS I don't think it is necessary for this paper.
@Avon2010 found that the main road effect on understory plant diversity extended <5 m into the forest stand. Ultimately, they delineated 3 habitats: the road verge leading up to the exact edge between the roadway and forest stand, the forest-road edge (0 - 5 m), and the forest interior.
-
Maybe we should define our edge/interior classifcation to have three classes (e.g. 0, 5-10, 20-60, or 0-5, 10-20, 40-60)?
-
It seems that something to separate out the 0 m location would make sense.
@Haddad2015 showed that >70% of all forests are within 1 km of a road, i.e., most forest patches have an edge ≤1,000 m away from