Anyone who has conducted field research knows that the very process of collecting data alters the system that you are studying. As you walk across a field, forest or stream to collect data, your footfalls trample vegetation, they compact the soil, they scrape algae off the rocks. Survey work usually involves a single visit to a site - as long as you avoid sampling from the areas you have trampled, it’s usually pretty safe to assume that your presence is unlikely to have affected the data that you have collected. Permanent plots are a different matter - because these plots are repeatedly sampled, there is cumulative damage. In larger plots, permanent trails may be established within plots.

Since we can’t avoid these effects, the real question is whether the effects are significant. Ecological systems are inherently heterogeneous. Does the effect of disturbance fall within the range of natural variability within the sample? That’s what really matters when it comes to data collection. In a forthcoming paper in the journal Biotropica, Liza Comita and Gregory Goldsmith “sought to quantify the significance and spatial extent of research trail impacts on the structure and dynamics of the seedling layer in the 50-ha permanent forest dynamics plot on Barro Colorado Island (BCI), Panama”. This site is very heavily used:

The number of people working within the 50-ha plot on any given work day ranges from six to 12, with up to 20 people present in the plot during the main census of trees, which occurs every 5 yr.

While seedling densities were (unsurprisingly) significantly lower on trails, they were significantly higher within 5 m of the trails. Seedling recruitment showed a similar trend, but the differences were not statistically significant. Between 5 and 20 m from the trails seedling densities were lower than the average for the plot, and recruitment was significantly lower. Comita and Goldsmith suggest suggest that this difference in seedling recruitment is due to trampling effects as researchers move away from the trails to collect data. They suggest that this effect is masked in areas adjacent to the trails; despite the fact that there is a closed forest canopy above the trails, they may still allow more light to reach the seedlings, resulting in increased seedling density. They also suggest that human usage of the trails may cause changes in animal behaviour and that this, too, may be responsible for the “masking” within 5 m of the trails.

Since the proportion of the plot that falls within 20 m of the trails is a fairly small portion of the total area of the plot, Comita and Goldsmith conclude that “calculations of vegetation structure and dynamics at the scale of the entire BCI 50-ha plot are not altered by the impacts of research trails”, but that this may not be true in smaller sites, since more of those plots are likely to be in close proximity to trails.

Comita, L.S., Goldsmith, G.R. (2008). Impact of Research Trails on Seedling Dynamics in a Tropical Forest. Biotropica DOI: 10.1111/j.1744-7429.2007.00337.x