There is a problem in doing ecological studies that is too little discussed – what is the time frame of a good study? The normal response would be that the time frame varies with each study so that no guidelines can be provided. There is increasing recognition that more long-term studies are needed in ecology (e.g. Hughes et al. 2017) but the guidelines remain unclear.

The first issue is usually to specify a time frame, e.g. 5 years, 10 years. But this puts the cart before the horse, as the first step ought to be to define the hypothesis being investigated. In practice hypotheses in many ecological papers are poorly presented and there should not be one hypothesis but a series of alternative hypotheses. Given that, the question of time can be given with more insight. How many replicated time periods do you need to measure the ecological variables in the study? If your time scale unit is one year, 2 or 3 years is not enough to come to any except very tentative conclusions. We have instantly fallen into a central dilemma of ecology – studies are typically planned and financed on a 3–5-year time scale, the scale of university degrees.

Now we come up against the fact of climate change and the dilemma of trying to understand a changing system when almost all field work assumes an unchanging environment. Taken to some extreme we might argue that what happens in this decade tells us little about what will happen in the next decade. The way around this problem is to design experiments to test the variables that are changing ahead of time, e.g., what a 5⁰C temperature increase will do to the survival of your corals. To follow this approach, which is the classic experimental approach of science, we must assume we know the major variables affecting our population or community changes. At present we do not know the answer to this question, and we rely on correlations of a few variables as predictors of how large a change to expect.

There is no way out of this empirical box, which defines clearly how physics and chemistry differ from ecology and medicine. There are already many large-scale illustrations of this problem. Forest companies cut down old-growth timber on the assumption that they can get the forest back by replanting seedlings in the harvested area. But what species of tree seedlings should we replant if we are concerned that reforestation often operates on a 100–500-year time scale? And in most cases, there is no consideration of the total disruption of the ecosystem, and we ignore all the non-harvestable biodiversity. Much research is now available on reforestation and the ecological problems it produces. Hole-nesting birds can be threatened if old trees with holes are removed for forestry or agricultural clearing (Saunders et al. 2023). Replanting trees after fire in British Columbia did not increase carbon storage over 55 years of recovery when compared with unplanted sites (Clason et al. 2022). Consequently, in some forest ecosystems tree planting may not be useful if carbon storage is the desired goal.

At the least we should have more long-term monitoring of the survival of replanted forest tree seedlings so that the economics of planting could be evaluated. Short-term Australian studies in replanted agricultural fields showed over 4 years differences in survival of different plant species (Jellinek et al. 2020). For an on-the-ground point of view story about tree planting in British Columbia see:
https://thetyee.ca/Opinion/2023/11/02/Dont-Thank-Me-Being-Tree-Planter/. But we need longer-term studies on control and replanted sites to be more certain of effective restoration management. Gibson et al. (2022) highlighted the fact that citizen science over a 20-year study could make a major contribution to measuring the effectiveness of replanting. Money is always in short supply in field ecology and citizen science is one way of achieving goals without too much cost. 

Forest restoration is only one example of applied ecology in which long-term studies are too infrequent. The scale of restoration of temperate and boreal ecosystems is around 100 years, and this points to one of the main failures of long-term studies, that they are difficult to carry on after the retirement of the principal investigators who designed the studies.

The Park Grass Experiment begun in 1856 on 2.8 ha of grassland in England is the oldest ecological experiment in existence (Silvertown et al. 2006). As such it is worth a careful evaluation for the questions it asked and did not ask, for the scale of the experiment, and for the experimental design. It raises the question of generality for all long-term studies and cautions us about the utility and viability of many of the large-scale, long-term studies now in progress or planned for the future.

The message of this discussion is that we should plan for long-term studies for most of our critical ecological problems with clear hypotheses of how to conserve biodiversity and manage our agricultural landscapes and forests. We should move away from 2–3-year thesis projects on isolated issues and concentrate on team efforts that address critical long-term issues with specific hypotheses. Which says in a nutshell that we must develop a vision that goes beyond our past practices in scatter-shot, short-term ecology and at the same time avoid poorly designed long-term studies of the future.

Clason, A.J., Farnell, I. & Lilles, E.B. (2022) Carbon 5–60 Years After Fire: Planting Trees Does Not Compensate for Losses in Dead Wood Stores. Frontiers in Forests and Global Change, 5, 868024. doi: 10.3389/ffgc.2022.868024.

Gibson, M., Maron, M., Taws, N., Simmonds, J.S. & Walsh, J.C. (2022) Use of citizen science datasets to test effects of grazing exclusion and replanting on Australian woodland birds. Restoration Ecology, 30, e13610. doi: 10.1111/rec.13610.

Hughes, B.B.,et al. (2017) Long-term studies contribute disproportionately to ecology and policy. BioScience, 67, 271-281. doi.: 10.1093/biosci/biw185.

Jellinek, S., Harrison, P.A., Tuck, J. & Te, T. (2020) Replanting agricultural landscapes: how well do plants survive after habitat restoration? Restoration Ecology, 28, 1454-1463. doi: 10.1111/rec.13242.

Saunders, D.A., Dawson, R. & Mawson, P.R. (2023) Artificial nesting hollows for the conservation of Carnaby’s cockatoo Calyptorhynchus latirostris: definitely not a case of erect and forget. Pacific Conservation Biology, 29, 119-129. doi: 10.1071/PC21061.

Silvertown, J., Silvertown, J., Poulton, P. & Biss, P.M. (2006) The Park Grass Experiment 1856–2006: its contribution to ecology. Journal of Ecology, 94, 801-814. doi: 10.1111/j.1365-2745.2006.01145.x.