Sitting on the fence has become a metaphor for ambivalence, but actual fence lines are some of the clearest lessons in land management. Fencelines can be the best place to study ecology, because most fences divide two different land management histories. The easiest places to learn from fence line contrasts is where the land management history is known. For example, along highway right-of-ways (ROWs), the strip of land between the road and the fence is almost never grazed, whereas the private or public alotment on the other side of the fence has almost certainly been grazed and/or farmed.
However, just because the ROW hasn't been grazed doesn't mean that it has escaped disturbance. While comparing two disturbed areas can yield some insights, the multiple factors at work will make cause and effect deductions extremely difficult. To find a good comparison, look for areas that are relatively far from the road; immediately adjacent to the road is a zone of disturbance, which can include vehicle traffic, trash, mowing, and runoff from the road (i.e. increased moisture).
The best comparison areas occur where the ROW is relatively higher than road (so there is no possibility of runoff and little chance of other human disturbance). However, areas with cutbanks below them are not good for comparisons, because of excess erosion, different microclimates around bare rock or exposed subsoil, and lowered water table. A zone of depression in soil moisture can also occur around ditches, trenches, gullies, roadcuts, etc.
The actual fence-line itself may have different species due to fence-line drip of dew and the ability of fences to catch seeds, especially tumbleweeds. (photo).
On the ground immediately beyond the fence there may be an area of extra disturbance due to cattle trails, etc, and any areas near stock tanks or gates are also likely more heavily used (and hence a more extreme contrast). In cases with less grazing on the private land, such as on steep hill slopes, vegetation and soils may look quite similar across fence-lines. Of course, there will always be variable disturbance on both sides of the fence, but that is part of the challenge and opportunity of observing fence-line contrasts.
The best comparisons are between areas relatively far from disturbance, close to but not immediately adjacent to the fence-line. With a good undisturbed ROW as a control, the vegetation on the other side of the fence can be compared to the potential climax community of the site.
I was recently watching fence-lines along NM highway 285 from Vaughn to Clines Corners, and noticed that typical overgrazed areas are Grama Grass (Bouteloua gracilis) monocultures or low-stature annuals with large amounts of bare dirt. The ungrazed roadsides still have bare ground, but the vegetation has a starkly different structure and composition: multiple grass species occur with different growth forms. But even more noticeable than the grass growth is the shrub encroachment in an area that is pure grassland. Without fire or grazing, woody growth, especially saltbush (Atriplex canescens) and Chimisa (Ericameria nauseosa) increases markedly.
While some trees and shrubs (e.g. E. nauseosa) are resistant to grazing when mature, their seedlings are highly palatable. Cows can completely eliminate woody overstories from riparian areas in a single generation simply be eliminating recruitment (through both grazing and trampling) of Cottonwood and Willow seedlings. Grazing pressure on seedlings is important, but easily overlooked: as long as there are trees, we describe an ecosystem as a forest. And it may seem strange to say that cows are eating a forest. But without seedling regeneration, no ecosystem is sustainable.
That grazing impacts woody growth as much or more than herbaceous growth is well-known along rivers and wetlands, but I think has been less remarked on in uplands. From this brief study of fence-line contrasts, it appears that even more of our grasslands would support shrublands were it not for either grazing or fire limiting woody plant establishment.