Can AI manage conflicting objectives in forest management?

At the start of the autumn term, landowners, technology companies and government agencies gathered in Stockholm to discuss whether, and if, conflicting interests could be tackled with the support of machine learning and AI? The question is particularly relevant for forestry, a sector facing increasing challenges in reconciling different interests such as climate change, timber production, nature conservation and reindeer husbandry.

At the same time, artificial intelligence (AI) and digitalisation have the potential to change the way we manage the forests. We can quickly analyse large amounts of data and use it to support decision makers in creating more precision-oriented forest management. In turn, this means – at least in theory – that we can manage and protect forests more efficiently, balancing different objectives, from timber production to the conservation of biodiversity.

AI is already being used to analyse forest information such as growth patterns, the inventory of tree species and the identification of forest damage. AI can also be used to monitor forest environments and detect anomalies, such as potential threats from forest fires, or from disease. This information can be used to optimise forest management and create a more sustainable production.

One of the most interesting technologies that is currently being discussed in forestry, is the development of digital twins. A digital twin is a virtual representation of a forest environment created by combining data from sources such as satellite images, and from sensors. This digital replica makes it possible to simulate forest management and to explore different scenarios in order to achieve specific objectives, such as reducing climate impacts or improving the health of the ecosystem.

Digital twins can provide us with a more dynamic and comprehensive picture of the forest. They make it possible to analyse the impact of different actions on the whole ecosystem, leading to better informed and more sustainable decisions. For example, AI can be used to assess the impact of a particular logging operation on water availability, or on biodiversity.

Hence, there is no doubt that AI and digitalisation can offer great benefits to forestry. In addition to streamlining forest planning and management, the technology can also potentially reduce conflicting objectives by helping us to make better, more data-driven decisions. For example, it may mean that we could better balance timber production with the need to retain forests for nature conservation and for climate measures.

Through AI-based solutions, we can also better anticipate and respond to problems such as forest fires or outbreaks of disease, which would contribute to more resilient forest management. In this way, technology can enable forest management that is not only more efficient, but also more sustainable.

However, despite its many benefits, digitalisation and AI also pose challenges within the forestry sector. One of the main dangers is the risk of an overly reductionist view of forests – one that relies too heavily on data and algorithms, and may miss important ecological or social aspects. A digital twin can never fully replace human judgement, or the local knowledge that is crucial for sustainable forest management.

Another challenge is the varying levels of digital skills within the forest sector. There is a digital divide, both between different organisations and individuals, but also within society at large. Many, especially older people , may lack the technical skills required to fully benefit from the new tools. This can create resistance to the implementation of AI-based solutions, despite their potential to streamline and improve forest management.

In the forestry sector, studies have shown that there is a variation of attitudes towards AI. Some stakeholders see the technology as an opportunity to improve forest management, while others are more hesitant. This uncertainty could be a barrier to fully embracing the new tools and to developing the digital skills needed to manage the forestry of the future.

In consequence, this was the focus of the workshop held around the theme of whether AI and digitalisation can help us to manage conflicting interests in the forest. The researchers are currently analysing the overall outcome. There were mixed views on the contribution of AI and digitalisation based on the availability of sufficiently good data for example, but also on the technological development itself, as well as its costs (not least future energy costs). Nevertheless, there was a general consensus on one thing: the need to invest in digital skills for AI and digitalisation to really help the forest sector to manage conflicting objectives. As continuous adaptation to new tools and technologies will be necessary, it is essential that we work to close the digital divide between different stakeholders. In turn, this increases the opportunities of ensuring that AI is used in an ethical, fair and sustainable way in forestry.

Camilla Sandström

Process leader social sustainability and professor of political science at Umeå University