Land provides many critical services for humanity ranging from food to water to energy. It also provides a range of services1 that are socially valuable – such as the use of land for recreation and its contribution to cultural identity – but may not have a market value.
Over the coming decades, the need for the critical, market-based services from land – food, water and energy - will intensify. It is therefore imperative to identify viable pathways to transform the land-based sectors so that the global population will continue to benefit from these, and the other, services into the future.
However, unlike the energy or the transport sectors, there is less understanding, or consensus, over what land-based sector transformation might look like. In addition, any future planning would need to take into account both global and local environmental boundaries, the crossing of which could cause irreversible harm not only to the environment but also the services it provides. While it is likely that these limits will constrain total outputs - such as the maximum yield of crops that could be farmed from the same piece of land without damaging soil quality or ecosystem health - it is often unclear where the environmental limits are in any given locality or on a planetary basis. Furthermore, the expectation that the world will exhaust the ‘carbon budget’ compatible with a 1.5°C trajectory in the next few years, would lead to an increase in the rate and scale at which negative emissions technologies (NETs), including large-scale afforestation, are deployed. This will further intensify the competing demands among the services derived at the nexus of food, water, energy and land resource interactions.
The transformation of the land-based sectors into more sustainable development pathways is complicated by the challenge of identifying the best possible options in a given situation on the ground. This is in part because food, water, energy and other services are interdependent, and that any specific interventions could affect the provision of services in complex ways due to the distinctiveness of geographies and local ecologies. In some cases an intervention could bring synergistic benefits, while in others they could result in difficult trade-offs.
On a more positive note, there is hope that tools like integrated assessment models (IAMS2) will soon be sufficiently sophisticated to explore how interventions may affect services in holistic ways. Such tools could be used for picking the most appropriate mix of policy interventions in particular geographies. Until such approaches become more developed however, practical frameworks are needed to help policymakers assess the potential outcomes from different options in order to identify interventions that are feasibly ‘low/lowest regrets’, ‘do no/little harm’ or ‘future-proof’.
Increasing awareness of the complexity of calibrating policies, and auditing the potential outcomes along the different axes or through different lenses, can help reduce the risk of unexpected negative outcomes and enable selection of the ‘lowest regrets’ options. These lenses or axes include: ‘wickedness’; scale and scalability; timescale of effects; and reversibility.
A ‘wicked’ problem is one where there is no straightforward solution.3 This may arise due to interconnectedness, when intervention in one area would affect others, potentially bringing both positive and negative outcomes. ‘Wickedness’ also arises through uncertainty, when the cause-effect pathways from intervention to impacts are unknown or unpredictable.
Goods and services provided by land are typically intimately connected, because they result from shared environmental processes. This gives rise to synergies among some services, and also trade-offs, so interventions to boost one service often decrease other services.4 For example, more intensive farming to produce more food may reduce the availability, or quality, of water. Changing farming practice also often affects biodiversity: an intervention may increase the biodiversity in some locations whilst reducing biodiversity in other locations.5
Interactions between services could also lead to uncertain outcomes, which are not well understood. This makes it difficult to predict how an intervention may affect the totality of services, even within specific contexts. Such difficulties could lead to a paralysis in decision-making. The idiosyncrasies of each place can add to this uncertainty, where specific outcomes may depend on the prevailing soil, climate, and topography. The same intervention may even have different impacts depending on neighbouring land uses and practices being adopted.6
Where such ‘wickedness’ exists there are no simple solutions to a problem meaning that it is difficult to identify any specific intervention in land management that can provide unambiguous positive attributes for all potential stakeholder groups and ecosystem services.
Examples of ‘wickedness’
Storing carbon vs reducing pesticides
Increasing soil carbon is beneficial for carbon storage, soil structure, fertility and water holding capacity. No-till agriculture is also a common intervention which saves energy, however, the downside of no-till is that it requires using pesticides which may have implications for biodiversity and consumer acceptability. This exemplifies trade-offs: as an intervention can impact the biophysical, social and economic environments in multiple ways, is storing carbon through no-till agriculture better or worse than the impacts of pest control on biodiversity?
Sustainability vs resilience
Sustainability is about avoiding environmental degradation, resilience is about maintaining services as external drivers vary. Building resilience may decrease sustainability. In a bad year, for example, a farmer may build resilience by:
- Over-producing to ensure enough end-product, creating waste;
- Using extra ‘insurance’ inputs to boost production;
- Using machinery on wet ground, compacting the soil.
All of these options decrease sustainability.
2. Scale and scalability
The place, time and scale of an intervention will also affect the impact of that action. This dependence arises from a range of factors:
First, the scale of the outcomes may vary depending on the scale considered: whether smaller (e.g. patch-within-field) or larger (e.g. landscape, national, regional or global). For example, 250kg/ha of Nitrogen applied on a single square meter will have no noticeable ecological impact, but the same rate of fertilizer application across a landscape will greatly affect soil, water, air quality and biodiversity. Likewise, the impact of an agri-environment scheme to plant flower-rich margins for pollinators, or provide food for birds, will be limited to feeding existing animals if there is only one in a landscape, but if there are many, they may provide extra resources sufficient to maintain new nests and grow the population.7
Second, the proportion of an area adopting an intervention affects outcomes. A single organic farm in isolation with insufficient supporting infrastructure may not fulfil market demand; but a larger number of organic farms might reach critical mass to stimulate a local market. Alternatively, in a large undisturbed area of natural land, the first 1 per cent of land converted to agriculture can be lost without noticeable impact on the other services that land provides. But if 99 per cent of the land is already converted, converting the final 1 per cent will cause extinction of the native biodiversity. The same intervention therefore has different impacts depending on the frequency of implantation: this is frequency-dependence.
Third, technical efficiency can have scale-dependent benefits. If things are equal, increasing technical efficiency produces more output from fewer inputs and is, therefore, sometimes seen as, by definition, an intervention that enhances sustainability. However, an important operational route to increasing efficiency is through economies of scale as the marginal costs decrease as the scale increases. Increasing the scale of production can bring significant benefits but also negative outcomes. For example, in crop production, increasing the spatial scale creates landscape of monocultures, removing the diversity of crops - and habitats - that support a range of wildlife. Similarly, large scale intensive livestock farms may be highly efficient, but there may be negative consequences to animal welfare, or disposal of the amount of waste.
At a global scale, liberalized trade incentivises technical efficiency via stronger competition. In addition to creating large-scale, intensive and homogenous landscapes, global food production has become concentrated in a few breadbasket areas which may accentuate the risk of disruption in the face of a drought or disease.8 Furthermore, this has also lead to the global homogenization of diets.9
The scalability of solutions is influenced by the factors above, such that innovations that appear positive at a pilot scale may not deliver the expected benefits on larger scales. One reason is that every location is different e.g. soils, topography, climate and micro-climate, local biodiversity, human population, cultures and access to technology. A recent analysis showed that on average soils with more carbon gave larger crop yields, but for any particular location this association could be negative or positive.10
Finally, feasibility – whether something is easily replicated at scale is another consideration. For example, ‘conservation agriculture’ (CA) is a mechanism for building climate-resilient agriculture, particularly in the developing world. It is based on mulching the old stems, protecting the soil from heat and evaporation and building soil carbon. However, it makes weeding difficult – so whilst it is a good practice, it implicitly relies on pesticides or labour availability. So despite incentivization and education programmes, CA is rarely taken up at scale.11
3. The timescale of effects
Not only do the impacts of an intervention depend on the spatial scale as above, they also depend on the temporal scale. This is because it often takes time for impacts to become apparent, as well as their effects accumulating over time. An example of the former would be that as a landscape becomes more agricultural, the services provided by insects that support farming (like pollination and natural pest-control) can be maintained despite gradual loss of biodiversity. However, at a certain point, ecosystems pass an ‘extinction threshold’ beyond which extinction becomes inevitable. At a larger scale, the Green Revolution was rightly lauded for its positive impacts in providing cheap food and bringing people out of food poverty. However, over time, the global costs of agricultural intensification have mounted: whether in soil degradation, biodiversity loss, pollution by pesticides, nitrates or GHGs, as well as the impact on human health in the developed world. This suggests the need for a reappraisal of its benefits.
Land use challenges
When practices are rare, their impact may be diluted, but over time as they become common, their costs and benefits become evident.
Cumulative degradation of important natural capital
Gradually degrading the environment can lead to ‘tipping points’ where the services suddenly decline. For example, in many intensively farmed areas, soil carbon is declining as synthetic inputs have displaced organic manure. Soil microbial fauna then declines, leading to the loss of soil structure, and increased erosion risk. This can create ‘dust-bowls’ from which it is difficult to recover.
The production of goods from the land economy is supported by services provided by the ecosystem. This includes natural pest control by small wasps, spiders and hoverfly larvae and pollination by insects. However, due to the intensification of land use and habitat loss, populations tend to become fragmented into small isolated pockets. While these can survive, it is likely that even without further environmental degradation, they will become extinct. This is known as the 'extinction debt', when populations sufficiently decrease and become remote, they are likely to go extinct.
When making ‘low regrets’ decisions in land-management, the question of reversibility is key. If the impact is unexpectedly negative, for example, can the land be restored to how it was? And if so, how feasible is it? Clearly, some decisions are easier to reverse than others: removing concrete foundations of an abandoned building is feasible, reclaiming the land from under a city is not. Bioremediation of an industrial site is feasible, cleaning up country-wide water courses suffering diffuse agricultural pollution would take time and incur a huge expense. Ecosystems in an area can be restored through habitat restoration, however extinction of species cannot.
As an example, soil loss through erosion is occurring much faster than new soil is being formed. Across Europe, soil formation rates range from ~0.3 to 1.4 t ha−1 12, yet, as several studies concur, many agricultural soils are eroding at a net rate of ~20 tons ha−1 yr−1 13. This implies that reversing the condition of eroded soils may take 20 times the timescale over which they degraded.
An important issue is to know when interventions are negatively affecting the long term sustainability of land use. This requires using metrics that can highlight when environmental degradation may need remediation. However, there are no easily used metrics that enable a holistic assessment of what constitutes ‘sustainable land use’.14 Nonetheless, the mantra of if you can’t measure it, you can’t manage it implies that there is a broader need, if land managers are to make ‘low regrets‘ decisions, to monitor the impacts of their management on more than the single product, like food or water.
Due to the factors outlined, it is difficult to design an intervention in land use that will not have some downsides and therefore be ‘wicked’ in some aspects. For example, improving food yield may:
Require technologies that have uncertain associated risks;
Impact soils with long-term detrimental effects;
Require more water and nutrients due to the enhanced growth rates of the plants that impact on the quality and quantity of water in rivers, and its availability for industry and cost to consumers;
Be taken up at scale because they are efficient, thus impact on biodiversity through reducing habitat availability, and changing the cultural value of the landscape;
Make a profit for farmers, in a food economy that provides cheap food but also encourages over-consumption, weight gain, ill health and food waste.
To navigate these complexities requires not only due diligence but also risk assessments along all the axes of regretability. It also requires engagement with all affected parties in discussions and decision-making. Once a decision has been implemented, it also requires more sophisticated monitoring, appropriate to place, than is typical.