Approaching a state shift in Earth’s biosphere part 2
In view of potential impacts on humanity, a key need in biological forecasting is the development of ways to anticipate a global critical transition, ideally in time to do something about it65. It is possible to imagine qualitative aspects of a planetary state shift given present human impacts (Fig. 1), but criteria that would indicate exactly how close we might be to a planetary-scale critical transition remain elusive. Three approaches should prove helpful in defining useful benchmarks and tracking progression towards them.
Tracking global-scale changes
The first approach acknowledges the fact that local-scale state changes—whether they result from sledgehammer or threshold effects—trigger critical transitions over regions larger than the directly affected area, as has been shown both empirically and theoretically66, 67, 68, 69, 70. On the landscape scale, tipping points in undisturbed patches are empirically evident when 50–90% of the surrounding patches are disturbed. Simulations indicate that critical transitions become much more likely when the probability of connection of any two nodes in a network (ecological or otherwise) drops below ~59% (refs 66, 67, 68, 69, 70). More generally, dense human populations, roads and infrastructure, and land transformation are known to cause ecological changes outside the areas that have actually undergone sledgehammer state changes68. Translating these principles to the planetary scale would imply that once a sufficient proportion of Earth’s ecosystems have undergone transformation, the remainder can change rapidly (Fig. 2), especially because emergent, larger-scale forcings (for instance changes in atmospheric and ocean chemistry, nutrient and energy cycling, pollution and so on) multiply and interact to exacerbate local forcings21 (Fig. 1). It is still unknown, however, what percentage of Earth’s ecosystems actually have to be transformed to new states by the direct action of humans for rapid state changes to be triggered in remaining ‘natural’ systems. That percentage may be knowable only in retrospect, but, judging from landscape-scale observations and simulations66, 67, 68, 69, 70, it can reasonably be expected to be as low as 50% (ref. 68), or even lower if the interaction effects of many local ecosystem transformations cause sufficiently large global-scale forcings to emerge. Figure 2: Quantifying land use as one method of anticipating a planetary state shift. The trajectory of the green line represents a fold bifurcation with hysteresis12. At each time point, light green represents the fraction of Earth’s land that probably has dynamics within the limits characteristic of the past 11,000 yr. Dark green indicates the fraction of terrestrial ecosystems that have unarguably undergone drastic state changes; these are minimum values because they count only agricultural and urban lands. The percentages of such transformed lands in 2011 come from refs 1, 34, 35, and when divided by 7,000,000,000 (the present global human population) yield a value of approximately 2.27 acres (0.92 ha) of transformed land for each person. That value was used to estimate the amount of transformed land that probably existed in the years 1800, 1900 and 1950, and which would exist in 2025 and 2045 assuming conservative population growth and that resource use does not become any more efficient. Population estimates are from refs 31–33. An estimate of 0.68 transformed acres (0.28 ha) per capita (approximately that for India today) was used for the year 1700, assuming a lesser effect on the global landscape before the industrial revolution. Question marks emphasize that at present we still do not know how much land would have to be directly transformed by humans before a planetary state shift was imminent, but landscape-scale studies and theory suggest that the critical threshold may lie between 50 and 90% (although it could be even lower owing to synergies between emergent global forcings). See the main text for further explanation. Billion, 109. In that context, continued efforts to track global-scale changes by remote sensing and other techniques will be essential in assessing how close we are to tipping the balance towards an Earth where most ecosystems are directly altered by people. This is relatively straightforward for land and it has already been demonstrated that at least 43% of Earth’s terrestrial ecosystems have undergone wholesale transformation1, 2, 34, 40, on average equating to ~2.27 transformed acres (0.92 ha) per capita for the present human population. Assuming that this average rate of land transformation per capita does not change, 50% of Earth’s land will have undergone state shifts when the global population reaches 8,200,000,000, which is estimated to occur by the year 202531. Under the same land-use assumption and according to only slightly less conservative population growth models, 70% of Earth’s land could be shifted to human use (if the population reaches 11,500,000,000) by 206031. Assessing the percentage change to new states in marine systems, and the direct human footprint on the oceans, is much more challenging, but available data suggest widespread effects38, 39. More precise quantification of ecosystem state shifts in the oceans is an important task, to the extent that ocean ecosystems cover most of the planet.
Tracking local-scale changes caused by global forcings
The second approach is the direct monitoring of biological change in local study systems caused by external forcing. Such monitoring will be vital, particularly where the human footprint is thought to be small. Observing unusual changes in such areas, as has occurred recently in Yellowstone Park, USA, which has been protected since 187271, and in many remote watersheds72, would indicate that larger-scale forcings38, 73 are influencing local ecological processes. A key problem has been how to recognize ‘unusual’ change, because biological systems are dynamic and shifting baselines have given rise to many different definitions of ‘normal’, each of which can be specified as unusual within a given temporal context. However, identifying signals of a global-scale state shift in any local system demands a temporal context that includes at least a few centuries or millennia, to encompass the range of ecological variation that would be considered normal over the entire ~11,000-yr duration of the present interglacial period. Identifying unusual biotic changes on that scale has recently become possible through several different approaches, which are united by their focus on integrating spatial and temporal information (Box 2). Breakthroughs include characterizing ecosystems using taxon-independent metrics that can be tracked with palaeontological data through pre-anthropogenic times and then compared with present conditions and monitored into the future; recognizing macro-ecological patterns that indicate disturbed systems; combining phylochronologic and phylogeographic information to trace population dynamics over several millennia; and assessing the structure and stability of ecological networks using theoretical and empirical methods. Because all of these approaches benefit from time series data, long-term monitoring efforts and existing palaeontological and natural history museum collections will become particularly valuable74.
Box 2: Integrating spatio-temporal data on large scales to detect planetary state shifts
Synergy and feedbacks
Thresholds leading to critical transitions are often crossed when forcings are magnified by the synergistic interaction of seemingly independent processes or through feedback loops3, 16. Given that several global-scale forcings are at work today, understanding how they may combine to magnify biological change is a key challenge3, 15, 16, 17. For example, rapid climate change combined with highly fragmented species ranges can be expected to magnify the potential for ecosystem collapse, and wholesale landscape changes may in turn influence the biology of oceans. Feedback loops also occur among seemingly discrete systems that operate at different levels of the biological hierarchy6, 8, 37 (genotype, phenotype, populations, species distributions, species interactions and so on). The net effect is that a biological forcing applied on one scale can cause a critical transition to occur on another scale. Examples include inadvertent, anthropogenic selection for younger maturation of individual cod as a result of heavy fishing pressure61; population crashes due to decreased genetic diversity75; mismatch in the phenology of flowering and pollination resulting from interaction of genetic factors, temperature, photoperiod and/or precipitation76; and cascades of ecological changes triggered by the removal of top predators62. In most cases, these ‘scale-jumping’ effects, and the mechanisms that drive them, have become apparent only in hindsight, but even so they take on critical importance in revealing interaction effects that can now be incorporated into the next generation of biological forecasts. Finally, because the global-scale ecosystem comprises many smaller-scale, spatially bounded complex systems (for instance the community within a given physiographic region), each of which overlaps and interacts with others, state shifts of the small-scale components can propagate to cause a state shift of the entire system21. Our understanding of complexity at this level can be increased by tracking changes within many different ecosystems in a parallel fashion, from landscape-scale studies of state-shifts12, 21 and from theoretical work that is under way20. Potential interactions between overlapping complex systems, however, are proving difficult to characterize mathematically, especially when the systems under study are not well known and are heterogeneous20. Nevertheless, one possibility emerging from such work is that long-term transient behaviours, where sudden changes in dynamics can occur after periods of relative stasis even in the absence of outside forces, may be pervasive at the ecosystem level20, somewhat analogously to delayed metapopulation collapse as a result of extinction debt77. This potential ‘lag-time’ effect makes it all the more critical rapidly to address, where possible, global-scale forcings that can push the entire biosphere towards a critical transition. Humans have already changed the biosphere substantially, so much so that some argue for recognizing the time in which we live as a new geologic epoch, the Anthropocene3, 16, 78. Comparison of the present extent of planetary change with that characterizing past global-scale state shifts, and the enormous global forcings we continue to exert, suggests that another global-scale state shift is highly plausible within decades to centuries, if it has not already been initiated. As a result, the biological resources we take for granted at present may be subject to rapid and unpredictable transformations within a few human generations. Anticipating biological surprises on global as well as local scales, therefore, has become especially crucial to guiding the future of the global ecosystem and human societies. Guidance will require not only scientific work that foretells, and ideally helps to avoid65, negative effects of critical transitions, but also society’s willingness to incorporate expectations of biological instability64 into strategies for maintaining human well-being.
Diminishing the range of biological surprises resulting from bottom-up (local-to-global) and top-down (global-to-local) forcings, postponing their effects and, in the optimal case, averting a planetary-scale critical transition demands global cooperation to stem current global-scale anthropogenic forcings3, 15, 16, 17, 19. This will require reducing world population growth31 and per-capita resource use; rapidly increasing the proportion of the world’s energy budget that is supplied by sources other than fossil fuels while also becoming more efficient in using fossil fuels when they provide the only option79; increasing the efficiency of existing means of food production and distribution instead of converting new areas34 or relying on wild species39 to feed people; and enhancing efforts to manage as reservoirs of biodiversity and ecosystem services, both in the terrestrial80 and marine realms39, the parts of Earth’s surface that are not already dominated by humans. These are admittedly huge tasks, but are vital if the goal of science and society is to steer the biosphere towards conditions we desire, rather than those that are thrust upon us unwittingly.