Food production in the Fens: navigating towards Net Zero?

Reposting here an article that I wrote recently for the International Peatland Society’s (IPS) Peatlands International (PI) quarterly publication. You can get access to the full publication (after becoming a member), access the back-catalogue for free, and find out who to contact if you want to write for PI here. I post this with thanks to Prof. Sue Page, for commenting on the drafted version, and to Susann Warnecke, for sending me to the FenlandSOIL gathering on behalf of the IPS, and generally for running the IPS ship so fantastically.

If you have ever enjoyed a fresh salad grown on English shores, it is likely to have comprised ingredients harvested from the Fens. A third of the country’s fresh vegetable produce comes from this region; an area of c. 3,900 km2, rich in peat. The Fens, situated across the counties of Lincolnshire and Cambridgeshire, and small parts of Suffolk and Norfolk, comprise lowland agricultural peat soils, the working of which generates some £3 billion each year and employs over 80,000 people. These people and resident communities share this region with 13,000 species of plants and animals, which live within and outside of the agricultural matrix.

Another key characteristic of the Fens is that its use in food production is “an obstacle” to achieving Net Zero by 2050. Centuries of farming in this peat-rich landscape has led to vast, largely unquantified carbon emissions and to extensive wastage of the peat soil. With United Nations Framework Convention on Climate Change (UNFCCC) signatory nations now required to measure, report, and reduce their greenhouse gas emissions across different sources in line with their Nationally Determined Contributions (NDCs) to mitigating climate change, emissions from farming peat soils need to be addressed.

Cue the formation of FenlandSOIL: a cross-sectoral group tasked with exploring how farming in the Fens can be achieved in a carbon-neutral way. This farmer-led consortium was established in 2021, and now has over 80 members from the farming community, academic institutions, and multiple other public and private sector organisations.

One of the FenlandSOIL associated partnerships is that between the World Wildlife Fund (WWF) and UK-based supermarket chain, Tesco. The goal of their collaboration, and of the FenlandSOIL consortium, is to answer the billion-dollar question: can we mitigate emissions whilst maintaining food production?

On 17th and 18th April, 2023, over 200 delegates gathered to explore this question, in the small city of Ely in East Cambridgeshire, perched on an island of hard sandstone within a fenland scape. Over the two days, attendees had a chance to mix with individuals from UK Government agencies, universities (including the key collaborators from the University of Cambridge), the National Farmers Union of England and Wales, Wildlife Trusts, supermarket chains, farm equipment suppliers, and an inspiring mix of others.

Alongside the incoming Scientific Officer, Dr Örjan Berglund, I was fortunate to attend this fascinating, inspiring, and at times frustrating meeting of minds on behalf of the International Peatland Society. After attending the two days of presentations, observing smaller group discussions and conversing with a range of different stakeholders in the conference breaks, I identified some common themes that seemed to emerge in this cross-sectoral space. Here are some of my learnings from the event:

  • No ‘one size fits all’ when it comes to developing interventions that will reduce emissions whilst enabling the food production to continue across farms in this peat-rich landscape. We need a framework to support the development of local solutions, which are bottom-up….
  • …and farmer-led. Of course, there is no one size of farmer, with each having a different relationship with the landscape they are farming, but every farmer will have knowledge and experience of that multidimensional space, which must feed heavily into each stage of intervention planning and practice. The depth of knowledge, understanding and passion of farmers attending the meeting was evident. Their voices must be present and centred in policy-facing discussions.
  • One skill-set farmers are often lacking is that required to carry out effective carbon management, being a relatively new role that the already hyper-skilled individuals are being tasked to take on. One farmer I spoke to was confused as to what the best approach to reducing emissions was for his farm – with his particular production system and land cover – after listening to multiple presentations advertising different emissions outcomes of different interventions for diverse production systems on different farms. An evidence-based approach is needed.
  • There was a call for that evidence-base to focus on “field-scale trials and innovations”; a continuation of the no ‘one size fits all’ principle. When evidence is often place-specific, incomplete, and associated with many uncertainties, packaging it into useful guidance for farmers is one of the ultimate challenges.
  • But whatever specificity of interventions are proposed, monitoring the changes in emissions, food production, soil health, species abundance, etc., and reporting those verified changes is essential, i.e., MRV – a feasible and effective Monitoring, Reporting and Verification procedure. We need to have standard, transparent, and feasible ways of assessing whether interventions are going any way to reducing GHG emissions over time, whilst not jeopardizing food production, livelihoods and other emergent properties of these systems.
  • The somewhat unpredictable elephant in the room – which could undermine even the most well-designed peatland carbon management plans – is climate change. This is seen as a large risk to food production and to climate change mitigation interventions. We need to understand more about how future climatic drying and erratic weather patterns may influence peatland ecosystem health, in order that current and near-future investments in restoring wasted peats are not themselves wasted.
  • The best way of climate-proofing any peatland is to manage the water table. There were plenty of discussions on storage, sharing, and managing risks associated with water across the Fens. From being a resource in abundant supply in this wetland-scape in the past, the lack of water resulting from hundreds of years of drainage is now a significant risk. Internal Drainage Boards (IDBs) are now the institutions who wield the power in these agricultural landscapes; they are in charge of decisions that can determine the economic and literal productivity of farms through water abstraction licencing. Although with transformative consequences on farming, imminent reform in licencing to reduce the exploitation of river water may create opportunities for peatland restoration. However that reform manifests, a short-term reduction in demand for water is needed across the Fens, alongside long-term local planning of water resource management, where the restoration of rivers to “good ecological status” is set as the achievable goal. As an aside, we were also reminded that water-level management in drains is not the same as water-table management in fields; each process plays a separate, yet interconnected role in peat soil conservation and food production.
  • The IDBs are, of course, not acting independently but in line with national legal frameworks for water management. When it comes to policies, Government intervention, be it through legislation or financial support, is seen as a double-edged sword. Although it may not be clear how the Government could support a strategy for food production in the Fens and the variety of lowland peatlands across the UK, there were proposals for how top-level support could reduce barriers to farming in these landscapes. There were discussions relating to England’s reformulated agricultural payment schemes, e.g., the post-Brexit Environmental Land Management scheme (ELMs), and how multiples of these could be ‘stacked’ together on one farm to increase the resources available to farmers to manage these complex landscapes. Where do farmers get the equipment, the seed-stocks, and other materials and expertise (in some cases) necessary for restoring their peatlands? Could logistical barriers also be reduced through policy change?

  • Or ultimately, is the lack of financial support the key challenge? Certainly, comments were made about the current lack of financial models that account for low-emissions farming practices on peatlands. We need a financial vision and framework, to accompany a logistical one.
  • One of these frameworks is carbon financing, and more specifically, the Peatland Code 2.0. The IUCN UK Peatland Programme has worked to revise this standardised procedure for valuing the carbon held in peatlands under protection, with areas of the Fens now eligible for financial investment through the voluntary carbon market under this scheme.
  • Regenerative agriculture is the future, we are told! I would like to believe this. But I am unsure what this is, exactly, and what it might look like in lowland peatland settings. I was reminded of the need to carefully define the terms we are using, lest they become straw men and lose their meaning, and thus power.
  • Whilst sharing learning and experiences across peatland regions can be valuable, we also need to appreciate the unique nature of the Fens. Lowland peatlands behave differently to those in the uplands; the latter being the subject of the majority of financial calculations and modelling for restoration. Lowland peatlands themselves come in a wide range of shapes and sizes….
  • Nuance matters. Variability in soil characteristics, water availability, and management practices across space and through time in the Fens need to be accounted for in any planning and practice. For example, the volume of Nitrous oxide emissions resulting from agriculture on peatlands may depend on the crop being cultivated and its in-field management; this detail matters.

To enable continued food production from the UK’s lowland peatlands, whilst mitigating (to some extent) carbon emissions from damaged peat soils, we need action now. We need a framework for local solutions. We need a field-scale, mosaic approach to interventions. We need to connect up communities and sectors across IDBs, and landscapes. And we need to create opportunities for social innovation. The FenlandSOIL gathering made an inspiring start.

Dr Lydia Cole
Coordinator of IPS Expert Group Peatlands and Biodiversity, University of St Andrews
lesc1@st-andrews.ac.uk

For Peatlands’ Sake

After many years in the planning, we have an exhibition in St Andrews all about peat. It even contains cores of peat extracted from down the road at Bankhead Moss and from across the Atlantic in the Peruvian Amazon. The exhibition is the brain-child/-bog of Katy Roucoux. She’s been working hard, alongside the Wardlaw Museum staff and a bunch of us peat-minded colleagues, for several years, to put together a room of stories about the ecosystems and people associated with the peatlands of Peru and Scotland, with reference to a splattering of other locations. It’s inspiring to see how research can be translated into a variety of interactive media, and how lush-looking walls of greenery can be created from plastic(!) (see image below).

The ‘lush’ green wall….

To accompany the exhibition, there are all sorts of events happening until the close on 7th May. One set of associated events are the Critical Conversations, where a small group of staff and students come together online to chat about a relevant topic. I was asked to chair a set of three conversations associated with the exhibition. Well, one to start with … and then somehow said yes to the set. (I’m not sure what happened to my resolution of saying yes less.)

In the first conversation, held on 20th February 2023, Katy Roucoux, Shona Jenkins and I discussed our thoughts in response to the question: can peat use be sustainable? We didn’t talk about the use of peat as a substrate in which to grow our tomatoes – a hot topic of discussion in UK policy circles at the moment – but focused on peatland use and whether that could be sustainable in the geographies which we are more familiar with: the Peruvian Amazon and Central Congo Basin. The answer is “yes”, but it depends. To find out about the circumstances under which the use of peatlands can be sustainable, have a listen to the podcast!

The second conversation, on 14th March, explored the multifaceted topic of ‘ethical’ fieldwork: what constitutes ethical fieldwork practices and how we might achieve them. Nina Laurie and Euridice Honorio shared their thoughts, developed through many decades of fieldwork in Peru and other locations, working alongside people with different life experiences, opportunities, and aspirations. Dennis del Castillo Torres, the Director of Forest Research at the Research Institute of the Peruvian Amazon (IIAP), also contributed some thoughts in a pre-recorded conversation that we (just about!) slotted into our live chat. It was interesting for me to have this conversation only one night after facilitating a panel discussion on how to avoid ‘helicopter’ science for the British Ecological Society (keep an eye on the Conservation Ecology blog for a summary article on that in the coming months).

The final Critical Conversation in For Peatland’s Sake trilogy, scheduled for 11th April, will ask whether museums can influence behavioural change. Answers on a postcard, please.

Although the two topics of the first and second Critical Conversations are nebulous, with the impossibility of reaching a one-size-fits-all answer, the conversationalists managed to articulate some really important points and provide plenty of food for thought, certainly for me. The gazillion £/€ question underlying all of the conversations is how do we achieve ‘real’ sustainability, and equity, in practice – where behaviours can continue indefinitely between passing generations – and generations of all people, across cultures, societies and social classes. Answering this question involves understanding how peatlands (in this case) function and what they need to be healthy, alongside understanding how we can nurture their good health whilst sharing out the gifts of these ecosystems amongst each member of our communities. Simple?! We’ll keep working on the answer.

It’s not rocket science.

I have just read this.  The Malaysia Palm Oil Board (MPOB) is hooking up with the Sarawak Oil Palm Plantation Owners Association (SOPPOA) to figure out what is going wrong with palm oil production in Sarawak.  Amongst other issues, members of SOPPOA have been complaining of: “oil palm estates (having a) …. serious infestation of Tirabatha (a moth of the pest variety), particularly in the lower Baram and central coastal regions, poor fruit set, bunch failures, lower frond dessication and acid sulphate soil problems”.  The lower Baram and coastal regions are mostly peatlands.  Acid sulphate soils result from the draining, oxidation and resultant subsidence of peat, exposing the underlying acid sulphate soils.  It’s not rocket science.

MPOB, the oil palm research and management organisation in Malaysia (Government-funded and led) that directs the production of palm oil across the three States of Malaysia, will conduct the research and report to SOPPOA.  SOPPOA is somewhat at MPOB’s mercy to suggest how farmers across the State of Sarawak can maximise yields and profit, especially from the coastal peatland zones, for which there is limited to zero knowledge on best management practices, from an optimum profit point of view.  From an environmental point of view, I’m not sure there is a point of view.

I wonder what MPOB will conclude.

 

NHS for Soils needed!

Every month, the Parliamentary Office of Science and Technology releases several publications, known as POSTnotes, that aim to provide an easily digestible overview of research in different areas of science and technology, as a tool for policy makers.  One released this month is all about securing UK soil health, in additional to the principally-important parts about peat.  I’d definitely recommend reading it if you’re interested in learning of the current status and threats to this ‘renewable resource’.

Renewable is a slightly misleading word.  Peat is a renewable resource if we wait about 3,000 years between harvests.  Fossil fuels could also be renewable if we could hold off popping the kettle on again for another 300 million years or so.  There should probably be a time frame attached to each use of renewable, and a conservative one at that, based on the Precautionary Principle.

A maturing sugarbeet field in East Anglia.

A maturing sugar beet field in East Anglia.

We basically need some, or even one coherent and policeable policy that governs sustainable soil management in the UK (and Europe), so that we can adhere to the Government’s plan to “grow more, buy more and sell more British food” over the next 25 years.  The world needs our sugar beet and broad beans.  And we all so desperately need our soils.

Retreating from peat!

A degraded tropical peatland in Borneo, with an approximately five year old oil palm plantation in the distance.

A deforested and drained tropical peatland in Borneo, with an approximately five year old oil palm plantation in the distance.

(I also just published this on the UK Tropical Peatland Working Group website.  Worth a quick gander!)

Asia Pulp and Paper (APP) has decided to ‘immediately retire’ approximately 7,000 hectares of its acacia plantations in Indonesia, with the goal of restoring them to intact peat swamp forest and developing a peatland best management practice model.  This is a bold move, forfeiting profits to comply with their Forest Conservation Policy (FCP).  Just over a year ago, they proved to be conservation forerunners again, (loudly) announcing to ‘protect and restore’ one million hectares of forest.  These come as welcome actions from APP, after it spent many years (and still is?) leading the deforestation frontier across Sumatra and Kalimantan, replacing hugely diverse ecosystems with monoculture plantations, and draining many a peatland along the way.

As Wetlands International say, there’s still a long way to go before APP can claim to be conserving, rather than destroying peatlands.  For example, how do they plan to rewet the peatlands?  What species are they going to plant into the current monocultures, and when?  How will they manage fire risk (heightened this year by ENSO) and potential flooding?   What will be the likely carbon emissions under different restoration strategies?  These are all important questions that researchers can help to answer.  Members of the UK Tropical Peatland Working Group are certainly on the case (watch this space).

But APP have given us a goal to hold them accountable to….and we must.

More information on the restoration mission from Deltares, APP’s independent peat expert team, can be found here.

Contenders for the Darwin awards?

Imagine this: you work for the Northern Powergrid, replacing wooden electriticy poles across the Cumbrian landscape.  Your specific role is driving the big diggers – the power behind the project.  You’re wanting to get your digger across the valley to the pole-replacement ground in time to get back home for tea.  So why not take a short-cut across that 1000 acres of flattish area you see infront of you?

Because …. YOU’LL SINK!  That’s what happened when Digger #1 attempted to cross the Butterburn Flow (what a name!) upland peat bog back in September.  What’s more, it’s a Site of Special Scientific Interest (SSSI); there were fears that oil leakages and the general disturbance would threaten the ecosystem.  Cue Digger #2.

Two diggers are stuck in the bogImage credit: ITV Border.

Digger #2 went in after Digger #1 on a rescue mission, and guess what?  IT SUNK!  Two huge diggers stuck-in-the-mud.

According to reports, there was a retrieval plan involving a temporary metal road surface, probably more diggers and a lot more people.  So they are likely to be out now, but I doubt they made it home in time for tea.

Contenders for the Darwin Awards?