"Predict and survive" - or not
There's an intriguing question asked in the pages of the US journal Proceedings of the National Academy of Sciences (PNAS) this week.
In a nutshell, it is this: can we forecast sudden, possibly catastrophic environmental changes by monitoring long-term trends?
As Reinette Biggs and her co-researchers point out, what can happen is that a trend gradually worsens for a while, but manageably, until there is a quick, catastrophic flip over into another state - the starkest example being from a species in existence to a species in extinction.
The term de jour for these changes is "tipping points". And although it's been bandied around in the arena of global warming for the last four or five years, the example this group uses in the PNAS paper involves fisheries, in particular the collapse of the Grand Banks cod stocks off the coast of Newfoundland in the early 1990s.
The collapse appears to have produced a "regime change". Once, adult cod kept numbers of smaller fish such as capelin down by the simple expedient of eating them. The depletion of adult cod has enabled capelin to thrive - and now, they are taking their revenge by eating juvenile cod so fast that few make it to adulthood.
The predator has become prey; the ecosystem has flipped over into another "regime", and may never return.
Scientists and fishermen - and Canadian politicians - knew in advance that the cod population was declining. But if there had been something to tell them it would result in a probably irreversible collapse, and by when, would they have done something about it?
Part of Jonathan's job is to devise new ways of monitoring declines in biodiversity. He pointed out to me that for all the talk of global changes, human society only has one continuous long-term record of a key global environmental trend - the carbon dioxide measurements made at Mauna Loa in Hawaii for more than half a century.
When I first thought about it, I decided Jonathan must have erred. The Central England Temperature Record dates back to 1659, and measurements of sea level using tide gauges began in Amsterdam not long afterwards. Britain's "first phenologist" Robert Marsham began recording the arrival of spring in 1736, while historical documents recording the timing of events such as grape-harvesting go back even further.
But after a bit more deliberation I realised that he had a point. Interesting and valuable as they are, all of these longer historical records are registering consequences of some environmental change; they are telling us nothing about the causes, and nothing about whether sudden change lies ahead and whether it will be reversible.
So what Reinette Biggs and her colleagues set out to ask in their simple model of an ecosystem very like the Grand Banks cod and capelin conumdrum was this: is there anything in the record, any signals that in retrospect could have shown that a huge change was coming, and indicated whether anything could be done to stop it?
If so, what would that tell us about using a similar prediction technique in the future?
Their conclusion basically was "no", for the Grand Banks fishery, or for anything else: "Indicators cannot at present be relied upon as a general means for detecting and avoiding ecological regime shifts."
The situation regarding projections of climate change impacts would appear to back up their view.
For all the talk of tipping points analogous to the cod collapse - irreversible melting of the Greenland icecap or drying out of the Amazon rainforest - uncertainty surrounds the precise temperature rises that could make them happen, and the precise greenhouse gas concentrations that would lead to these temperature rises, and whether reductions in emissions could reverse them.
The situation becomes even more complex when you look at the multiple pressures that crowd in on ecosystems and species. The most acute issue for the Wyoming Toad is the fungal disease chytridiomycosis that is threatening amphibians worldwide; but the Red List of Threatened Species also lists salinity changes (possibly linked to climate), invasive species, "predation, pesticide use, irrigation practices, and lack of genetic diversity".
Sort out an early warning system for that lot if you can.
Environmental monitoring is increasing. Satellites now watch over a plethora of parameters including forest cover, clouds, ice extent, and gases in the atmosphere.
Countries such as the UK have set up "biodiversity indicators" such as the numbers of key species, water quality, the extent of sustainably managed woodland, and even the amount of time that people spend volunteering for nature conservation tasks.
But for all that, we are a long way from knowing what to look for in the streams of data these projects produce; a long way from being able to tell what will survive without our help, what are the most important interventions to make, and when the deadlines for making those interventions will arrive.
So where does that leave us? In the words of Reinette Biggs: "While this research develops, management for unwanted regime shifts will depend on existing approaches that hedge, avoid risk, maintain ecological resilience, or build social resilience to cope with unexpected change."
In other words, "predict and survive" is for now an unfeasible doctrine for environmental protection; deploying a bit of prudence and a bit of planning is, she suggests, probably the best we can do.