Semi-natural habitat natural capital accounts, UK: 2021

Exploring the size, condition, quantity and value of semi-natural habitats and ecosystem services, as part of the UK Natural Capital accounts. These are our most natural spaces, although they have been altered by human activity.

This is the latest release. View previous releases

5 August 2021 15:30

A correction has been made to the carbon sequestration estimates. This was due to a error in the units of the physical values of carbon sequestration. This has affected Table 22, Figures 44, and the dataset. You can see the original content in the superseded version. We apologise for any inconvenience.

View superseded version

Email Hazel Trenbirth and Adam Dutton

Release date:
10 February 2021

Next release:
To be confirmed

1. Main points

  • Our most natural areas of the UK, semi-natural habitats, occupy 32.6% of the land area of the UK with 8.02 million hectares.

  • The annual value of services from the UK's semi-natural habitats is estimated at £7.0 billion in 2018.

  • There were an estimated 1.7 billion hours spent on recreation in semi-natural habitats in 2018.

  • The removal of air pollution by semi-natural habitats in the UK equated to an estimated saving of £634.1 million in health costs in 2018.

  • The asset value of UK semi-natural habitats was estimated at £269.8 billion in 2018.

Back to table of contents

2. Size of the area covered by semi-natural habitats

The Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services defines semi-natural habitats as:

"An ecosystem with most of its processes and biodiversity intact, though altered by human activity in strength or abundance relative to the natural state."

The European Investment Bank definition gives more detail on types of modification that could be, "in their composition, balance or function". The definition also gives more detail on how they come about:

"They may have evolved through traditional agriculture, pastoral or other human activities and depend on their continuation to retain their characteristic composition, structure or function."  

While definitions vary, the intensive history of human occupation across the UK means even the most natural ecological assemblages are generally described as being, "semi-natural".

As such, a semi-natural natural capital account for the UK can broadly be interpreted as an account of much of its natural spaces. Previous Office for National Statistics (ONS) natural capital publications have been for a single broad habitat, such as woodlands and urban. This publication focuses on a range of habitats that make up the semi-natural landscape as it provides a more holistic narrative than just semi-natural grasslands alone. Enclosed farmland, urban, marine and coniferous woodland are excluded from semi-natural habitats.

Given both the varied definitions of semi-natural and uncertainties in mapping such habitats, calculating their extent in the UK is not easy, meaning there is no definitive answer.

The Land Cover Map 2015 (LCM2015) was created to provide a consistent approach UK-wide (see Figures 1, 2, 3 and 4), and is based on UK Biodiversity Action Plan Broad Habitat definitions. Using the LCM2015 categories, 32.6% of the UK can be considered to be semi-natural (Table 1). Full definitions of each habitat class can be found in the dataset.

Different agencies within the UK use different methods. Wales is assessed by Natural Resources Wales, who estimate that 30.8% (640,827 hectares (ha)) of the land area of Wales is semi-natural. This estimate is from combined analysis of satellite imagery and existing habitat maps. Applying the LCM2015 classification to Wales, 34.9% is classed as semi-natural.

NatureScot intends to adopt the European Nature Information System (EUNIS) habitat classification as its framework for terrestrial habitat data and mapping in Scotland.

The Department of Agriculture, Environment and Rural Affairs (DAERA) in Northern Ireland considers semi-natural habitats that are protected to be old grassland, woodland, scrub, bogs, wetlands and moorland.

There is also an issue around woodlands. Forest Europe defined semi-natural forests are neither a plantation or undisturbed by man and display certain characteristics of the natural ecosystem. As we do not have data on the area of semi-natural woodlands in the UK, we are using the broadleaved woodland land cover category as a proxy.

Back to table of contents

3. Condition of UK semi-natural habitats

The System of Environmental Economic Accounting (SEEA) defines ecosystem condition as "the overall quality of an ecosystem asset in terms of its characteristics". Suitable condition indicators are needed so there is an understanding of the relationship between the ecosystem condition and the ecosystem services they deliver.

There are a range of metrics for measuring the health of semi-natural habitats, including:

  • condition of the soil

  • chemistry of the water

  • presence of different mammals

  • birds, bats and insects as well as the sizes of their populations

All of these metrics reflect ecological health.

Ecologists regularly use a simple set of descriptive words to summarise the overall status of a habitat or species. For instance, "favourable" being good, "unfavourable" being less good, and "destroyed" meaning the habitat at a site is no longer present and there is no prospect of being able to restore it. These descriptors are based on an assessment of an appropriate set of metrics depending on the habitat being measured.

The bird indices on population numbers for woodland and water and wetlands plus butterflies of woodland and semi-natural habitats are in long term decline, with a negative impact on biodiversity. While there has also been a decline in the average numbers of bees counted in the last decade in semi-natural habitats, in the same period the average number of bats counted has increased.

Wildfires in the UK also have a negative impact. There has been a reduction in the area burnt in Scotland between 2009 and 2013, while in Wales, there was an increase in the number primary fires between 2011 and 2019 and in England an increase in area burnt between 2009 and 2017.

The summary of the condition of the UK semi-natural habitats (Table 2 and Table 3) give an overview of the more detailed analysis that follows. The long-term trend looks at the whole time series, while the short-term is the most recent trend and is not a fixed time period and is relative to the length of the time series.

For instance, if a metric has been falling for the last three or five or even seven years within a 30-year dataset (in which the earliest period was actually still lower than the current year), it would have an increasing long-term trend and decreasing short-term trend. Within a five-year dataset the two might not differ.

The System of Environmental Economic Accounting (SEEA) is developing a typology of condition for ecosystem accounting. Included is the use of ancillary data, which is classed as data that are not included in the condition accounts, however, they are data that can be used as a proxy for missing metrics. SEEA classes data on protected sites and pressure indicators, such as wildfires, as ancillary data (Table 3.).

Physical and chemical state condition indicators

Soil indicators

Soil is important for providing many essential ecosystem services. Monitoring trends over time of specific soil indicators provides a suitable condition indicator for semi-natural habitats. The Environment Agency states that soil holds three times as much carbon as the atmosphere, absorbs water and therefore reduces the risk of flooding. It is also a wildlife habitat and delivers 95% of global food supplies. In 2019 the Environment Agency identified:

  • almost 4 million hectares of soil that are at risk of compaction (soil particles pressed together)

  • more than 2 million hectares of soil are at risk of erosion

  • intensive agriculture has caused arable soils to lose about 40% to 60% of their organic carbon

  • soil degradation was calculated in 2010 to cost £1.2 billion every year

The Environmental Agency used indicators including:

  • pH (acidity) levels

  • carbon levels 

  • loss of ignition (organic matter content)

  • bulk density (weight soil in a given volume)


The pH levels in soil is a measure of acidity. A low pH level is more acidic and can be damaging for some habitats. Mountains, moorland and heath (MMH) habitats, shown in Figure 5 as dwarf shrub heath and bracken, have always been naturally acidic. Between 1978 and 2007, the average pH level has increased by 8.6% for dwarf shrub heath and 12.3% for bracken, reflecting the recovery from acidification and reductions of sulphur emissions since the 1980s (Figure 5).

The pH levels of broadleaf, mixed and yew woodland has increased by 13% between 1978 and 2007 (Figure 5). Broadleaved woodlands dominated by oak and birch occur on more acidic and infertile grounds. Broadleaf, mixed and yew woodland pH levels in 1978 were at their lowest level of 5.09 but have since increased to 5.75 in 2007.

Forest Research investigated acidity exceedance and the data for 2004 revealed 69.1% of broadleaved woodland (managed) and 58.6% of unmanaged woodland (ancient and semi-natural) habitat area was exceeded for acidity critical loads. With acid deposition above the critical load there is a risk of damage to the ecosystem.

Neutral grassland and acid grassland have seen increases in pH levels of 11% and 8%. Fen, marsh and swamp have increased pH by 17.5% from 1978 to 2007 and bogs have had a 5.1% rise in the same period (Figure 5).

A report carried out on behalf of Natural England investigated soil pH and nutrient status on selected protected sites for acidic grassland, neutral grassland, calcareous grassland and fen, marsh and swamp. A total of 602 soil samples were taken in 2014. The results of the survey suggest that unfavourable condition was a result of high phosphorus and potassium levels in soil.


Soil organic carbon is an important indicator of soil quality and it is important for soils to retain their carbon. This plays a critical role in soils maintaining their structure and water retention. Carbon levels in all the semi-natural habitats increased in Great Britain between 1978 and 1998, then decreased between 1998 and 2007. So overall, there was no change in the levels of carbon.

Organic matter content

Organic matter content is plant and animal residues in different stages of composition and is another good indicator of the condition of soil.

Ideally, soils should be slowly increasing their levels of organic matter content over time as they have been doing since the last glaciation 11,000 years ago. However, as the climate changes, or grazing pressure, or burning or cutting intensity increases, there can be a loss of organic matter content because of increased decomposition rates of soil organic matter and/or increased erosion of the organic-rich top layer of soil.

Dwarf shrub heath has seen a 7% reduction in loss of ignition from 1978 to 2007 while acid grassland and fen, marsh and swamp have seen a 3% and 1% reduction in loss of ignition respectively during the same time period. Between 1978 and 2007, woodland has seen the largest increase of 42%, with bogs increasing by 5% and neutral grassland by 1%.

Bulk density

Bulk density is an indicator of soil compaction and is calculated as the dry weight of soil in a given volume. Compaction of soil prevents plant growth and increases erosion. A high bulk density is an indicator of poor condition as it can restrict root growth and increase soil erosion. The Countryside Survey only has 2007 data available (Table 4) and therefore cannot be used as an indicator of soil condition change over time.

Uplands water quality

The UK Acid Waters Monitoring Network was set up in 1988 to investigate the chemical and biological responses of acidified streams and lakes to changes in air quality and "acid rain". It later became the UK Uplands Water Monitoring Network (UWMN) in 2013 and now has a wider remit to also investigate impacts of climate change and land management on these systems.

Among a wide range of variables, the UWMN measures the acidity and acid neutralising capacity (ANC) of these waters, together with dissolved organic carbon (DOC), nitrate and non-marine sulphate. Year-on-year changes in these metrics across a wide range of sites provide a strong indication of how the water quality our upland streams and lakes has been evolving over recent decades.

Currently the UWMN samples around 26 sites regularly for water quality assessment. Sites from mountain, moorlands and heath areas were extracted and data were upscaled to a UK level for the purpose of this report. While the number of sites is limited, they show highly coherent behaviour in the metrics presented. This indicates that the changes observed should be broadly representative of changes in acid-sensitive upland waters more widely.

Non-marine sulphate

Non-marine sulphate in upland surface waters represents the chief acidifying anion (a negatively charged ion) and is mostly derived from the burning of fossil fuels by power stations and heavy industry. Upland water condition has improved from the reduction in concentrations of non-marine sulphate from 1996 to 2018, from an average of 61 to 29 microequivalent per litre in our dataset, with the most rapid reduction occurring in the late 1990s (Figure 8).

Climatic events can influence the sulphate pollutant signal in upland waters. During droughts, such as the one experienced in the UK in 1995, previously inert sulphur compounds deep in the catchment soils, can re-oxidise as sulphate, resulting in a temporary increase in sulphate concentrations when the rain returns.


pH is a measure of the acidity or alkalinity of water. It is measured on a scale of pH units between 0 and 14, with low numbers being acidic, seven being neutral and higher values being classed as alkaline. Over the time series (Figure 9) acidity has gradually declined in response to a reduction in acid anions like sulphate.

Figure 9 illustrates the gradual rise in the five-year mean of pH with clear inter-annual variability superimposed. The reduction of acidity is considered to be making these waters more favourable to a range of aquatic organisms, including algae, higher plants, macroinvertebrates and fish.

Acid neutralising capacity

Acid neutralising capacity (ANC) is a measure of the capacity of water to resist changes in pH levels. Catchments whose drainage waters have higher levels of ANC have a greater ability to neutralise acid deposition.

An increase in ANC is favourable for water quality because this represents an increases ability of a system to neutralise acid inputs. The timeseries (Figure 10) shows that on average, the ANC of these waters has increased from 26 to 69 microequivalent per litre, from 1989 to 2018.

Figure 10 shows that superimposed on gradual increase in the long-term mean ANC has shown significant inter-annual variations, much of which can be attributed to changes in rainfall and amounts of sea salt deposited during winter storms.

Dissolved organic carbon

Dissolved organic carbon (DOC) is derived largely from the degradation of plant and soil organic material. It forms a substantial part of dissolved organic matter (DOM) that often causes a brown staining to upland waters. Concentrations of DOC tend to be particularly high in waters draining from peatlands.

A large proportion of the UK's drinking water comes from the uplands and the water industry must remove most of the DOM at an early stage in the treatment process  to avoid the creation of potentially toxic disinfection by products. Water that is sourced from degraded peatland has dissolved organic carbon, representing a large cost to water companies to remove it.

As Figure 11 shows, mean concentration of Dissolved Organic Carbon (DOC) in our dataset has roughly doubled from 2.6 milligrams per litre in 1989 to 5.0 milligrams per litre in 2018, however, there is considerable inter-annual variation and it has been flat since 2004. The increase in DOC is thought to be largely because of an increase in the solubility of soil organic matter as upland soils recover from the effects of acidification.


Nitrate concentrations have fluctuated over the time series with substantial inter-annual variations (Figure 12) but also show a gradual long-term decline. This is consistent with a gradual reduction in the atmospheric deposition of nitrogen to these catchments over the monitoring period.

Nitrate concentrations tend to peak during the winter months, when nutrient demands within catchment soils are at a minimum. Analysis by the UKAWMN for data over the period 1998 to 2000 suggested that much of the interannual variation in concentrations could be linked to variations in winter temperatures, with the highest concentrations occurring in the coldest years.

There are health and environmental reasons for concern about the level of nitrates in drinking water. However, the concentrations measured across the Upland Water Monitoring Network (UWMN) are mostly considerably lower than the nitrate concentration limit (50 milligrams per litre) specified under the Drinking Water Directives of 1980 and 1998.

Water quality from Water Framework Directive

The Water Framework Directive (WFD) is an important tool for assessing the water environment across Europe for several types of water bodies, including rivers, lakes and canals. The overall condition of each body of water  is examined and categorised into one of five classifications, ranging from "bad" to "high" condition status.

The WFD operates in six-year cycles, with the second (2016 to 2021) cycle currently in effect. This framework helps to classify the quality of water bodies by measuring their ecological and chemical status.

Poor ecological and chemical condition of a surface water body can cause a range of damage, including drinking and bathing water quality, biodiversity, and marine life health. One of the main aims of the WFD, across all surface water bodies in the European Union, is that all bodies of water should be in a "good" condition.

Figure 13 illustrates the number of surface water bodies in the UK and their condition classification (excluding sites classified as "unassessed"). The surface water bodies data are comprised of data from lakes, canals, rivers, estuaries and coastal water bodies. Heavily modified water bodies are also included in this dataset.

There were 9,301 recorded sites in the UK in 2019, with 2,873 of surface water bodies in "good" condition. From 2016 to 2019, the proportion of sites in a "poor" or "bad" condition remained consistent. There has also been an increase in the number of sites classified as "good" in the period 2016 to 2019, rising by 93 sites. The number of sites classified as "moderate" has fallen in the period 2016 to 2019.

Across the time series shown, there was a decline in the number of water bodies observed from 2015 onwards. This is because of the second phase of the WFD adopting different classification standards and excluding measurement of any water body catchment area that is less than 10 square kilometres.

Figure 14 provides an overview of the condition of UK lakes as a percentage; lakes are a subset of the data in Figure 13. In 2019, approximately a quarter of the monitored lakes were in a "good" condition, while the largest individual proportion of lakes is classed as being in a "moderate" condition (54%). The number of lakes in each classification remains consistent in the period 2016 to 2019 (Phase 2 of WFD enacted for these years).

The Water Framework Directive is one of several approaches that national and local governments are using to monitor water quality. Although the initial aim of the WFD indicated that all surface water bodies should be in a “good” condition by 2015, the data presented indicates that small improvements in water quality have been made.

Compositional state condition indicators

Bird index

Bird populations are thoroughly monitored in the UK. Bird populations are considered as a good indicator of the broad state of the wildlife. Birds occupy a range of habitats and their behaviour responds to environmental pressures. The species are selected for the index if they have a population of at least 300 breeding pairs and are a native species. To find out more about how bird populations are counted, please visit the Breeding Bird Survey, British Trust for Ornithology (BTO) website.

In this subsection we present a range of habitat-based bird population indices. Many are official statistics produced by the RSPB and BTO. Some habitats do not yet have official indexes and where this is the case, we took earlier papers on potential species lists for those habitats and built new indices. ONS-built indices are presented as experimental statistics for information and should not be confused with the official indexes.


The woodland index covers all woodland including semi-natural broadleaved ones (approximately 55% of total woodland area). The woodland bird populations decreased by 25%, according to unsmoothed data, between 1970 to 2019 (Figure 15). We can see that there was a steep decrease in the early 1990s, then a flattening. Until 2013, when the index began to decline again (Figure 15). Some possible reasons for this decline are outlined in the following, but none of these can be confirmed and the Royal Society for the Protection of Birds (RSPB) is investigating the cause.

Since 1970, 22% of monitored woodland species have increased in population numbers, 32% declined and 46% showed little or no change. Two breeds of bird among those whose populations have increased are the great spotted woodpecker and the nuthatch. According to Defra's Wild Bird Populations in the UK, 1970 to 2019, this decrease may be a result of a lack of woodland management along with increased deer-browsing pressure. Both result in a lack of diversity within woodland and few suitable nesting and foraging habitats.

There has been a decline of more than 80% in five woodland specialist populations since 1970; the lesser spotted woodpecker, lesser redpoll, capercaillie, spotted fly catcher and the willow tit. While the reason for this large decline is unclear, the RSPB's managing willow tit habitat project highlights changes in habitat structures and the drying out of woodland soils as a possible reason.

Lowland acid grasslands

The lowland acid grassland bird index, shown in Figure 16, is an unsmoothed index calculated by the Office for National Statistics (ONS) from the Breeding Birds Survey data. The bird species associated with this habitat are similar to other lowland dry grasslands. There has been substantial decline in this habitat during the 20th century which Natural England has attributed this to agricultural intensification.

The index has remained consistent over the long-term. However, there is huge variation in the individual species that are included the index. The chough population, for example, has increased by 152% since 1994 and the hen harrier has increased 57% from 1994 to 2019. The skylark has declined 18% between 1994 and 2019.  This has been attributed to changes in farming practices.

Water and wetlands

This index includes rivers, ponds, lakes, coastal marshes, reedbeds and other wet grasslands and lowland raised bogs. There are 26 species of birds that live in water and wetland. The bird populations decreased by 15% between 1975 and 2019, according to the smoothed index (Figure 17).

There were declines of 1% in reed bed birds and 52% in wet meadow birds. Those in fast flowing water habitats declined by 25%.  The only increase in the index was shown in the slow standing water birds, at 23%.

According to Defra's Wild Bird Populations in the UK, 1970 to 2019, the decrease in breeding waders may be a result of land management changes. The yellow wagtail has declined by 97% and the snipe by 80% over the long-term. The British Trust for Ornithology suggests yellow wagtails are in decline because of changes in agricultural practices. However, species associated with bodies of standing water have benefited from the creation of new habitat from restoring gravel pits. The mallard and tufted duck are two species to have benefited over the long-term and shown an increase; the tufted duck has increased by 65% and mallard almost trebled.


The moorland bird index, shown in Figure 18, is an unsmoothed index calculated by the Office for National Statistics (ONS), using a species list developed in the early stages of potential development of an official index in UK natural capital: mountains, moorland and heath accounts in 2019. The index between 1994 and 2019 has increased by 13%, with the largest increases being seen in red grouse, raven and hen harrier. There is a noticeable drop in the index in 2001; this is because of reduced monitoring as a result of the foot and mouth outbreak that year when footpaths were closed on moorlands and the number of surveys dropped.

Although the disaggregated index fluctuates, there have been large decreases in some species of whinchat (51%), merlin (36%) and curlew (29%). The hen harrier has seen a 57% increase, helped by the success of Natural England's hen harrier recovery project established in 2002.

Though the hen harrier increases were from a very low baseline and Natural England chairman Tony Juniper stated, “Despite the great progress there is though no cause for complacency. Too many birds still go missing in unexplained circumstances and I urge anyone who is still engaged in the persecution of these magnificent creatures to cease at once.”

Butterflies and moths

According to the UK Butterfly Monitoring Scheme, butterflies and moths are good indicators to environmental change as they have short life cycles and react quickly to weather and climate change.

Moths hold vital roles in the wildlife ecosystem with more than 2,500 species present in Britain in a range of habitats. The State of Britain's Larger Moths 2013 report showed the total number of larger moths recorded in the Rothamsted traps national network declined by 28% from 1968 to 2007. Between 1976 and 2017, in the UK, the habitat specialists' butterflies index has reduced by 77% and the abundance of butterflies of the wider countryside has reduced by 46%.

Habitat specialist index

This index includes individual measures for 26 habitat specialist butterflies for low mobility species, restricted to semi-natural habitats. Semi-natural habitat loss and fragmentation has resulted in a decline in habitat specialist butterflies. Habitat fragmentation prevented recovery from the declines experienced in the 1970s, which were mostly to do with the knock-on effects of the drought in 1976.

Figure 19 shows a long-term decline of 59% between 1976 and 2019. The greatest decline since 1976 has been seen in the heath fritillary, Lulworth skipper, wood white, grayling and small pear-bordered fritillary.

Moorland moth index

The unsmoothed index has fluctuated over the time series (Figure 20) for moorland moths, with the highest count of 3,241 recorded in 2018. The index between 1991 and 2018 has increased by 80% from 100 in 1991 to 180 in 2018.

Woodland index

Both woodland moths and butterflies have seen significant declines in the unsmoothed (raw) index in the long-term between 1990 to 2018. Moths have seen a decline of 2% and butterflies have declined by 31% (Figure 21). (Data for butterflies available until 2019 shows a 41% decline).

Changing management practices over recent decades has been suggested for the decline in moths and butterflies in broadleaved woodlands. Reductions in open spaces and increased shade has changed the plant communities, vegetation structure and micro-climates.


Bees provide a range of ecosystem services as well as being useful indicators of wider ecological health. In particular, bees are one of the main groups of insects responsible for pollination of wildflowers, berries, orchards and crops. As a third of all crops in the UK are pollinator dependent, this is a particularly important service. Habitat loss and degradation is related to a decline in the bee population, making them a useful way of monitoring long-term changes in the condition and health of the environment.

The Bumblebee Conservation Trust runs the BeeWalk Survey Scheme, which uses citizen science volunteers to monitor the number of bumblebees on a monthly walk, along a set route of approximately one mile, from March to October.

The BeeWalk identified worker bees (the most common), drones (fertile males) and queens (the sole fertile female in any colony). The number of queen bees per kilometre were counted and reported over time (Figure 22).  The total number of bees per kilometre (Figure 23) per bee walk acted as indicators of the condition of semi-natural habitats. There was a drop in the average number of bees counted during 2011. This drop corresponds with a year in which England and Wales suffered a significant drought. Over the time series there has been a decline in the average number of bees, counted from 30 in 2010 to 26 in 2019 (Figure 23).

There has been an overall decline in wild bee diversity over the last 50 years in the UK. It is estimated that in the UK 97% of unimproved grassland was lost in Wales and England between 1932 and 1984. This means species-rich grassland in the UK is now fragmented and often degraded. This loss of unimproved flower-rich grassland is attributed as a cause of the decline in many bumblebee species.

Diseases in bees

The disease data are based on apiary inspections. However, some honeybee diseases can be spread into the wild bumblebee population, as there is growing evidence that pathogens and parasitic organisms can be shared to the wider pollinator community from managed bees.

The American foulbrood disease (AFB) is a contagious bacterial disease of honeybee larvae, which is spread by spores carried on drifting bees from nearby colonies. The number of bees being diagnosed with American foulbrood disease has been stable over the last 12 years. With the exception of 2009, less than 1% of tested apiaries testing positive (Figure 24).

European foulbrood (EFB) is also a bacterial disease that  affects larvae before the capped stage but is not the same as AFB, which is caused by different bacteria and produces different symptoms. By comparison, European foulbrood disease has fluctuated over the same time period, with 2012 and 2020 being greater than 6% (Figure 24).


Bat populations are widely distributed throughout different habitats in the UK, including semi-natural ones. Bats are considered a good indicator of the health of the environment as they depend on a range of habitats, are reliant on insect prey (in the UK) and are sensitive to changes in land use, climate and site management.

Annual surveys of bats are co-ordinated by the National Bat Monitoring Programme, run by the Bat Conservation Trust, which organises volunteers to monitor bats at survey spots and walks. The semi-natural bat index (Figure 25) is a composite of four bat species trends: Daubenton's bat, common pipistrelle, soprano pipistrelle and noctule. It is calculated from survey locations that intersect semi-natural habitats. The semi-natural bat index has increased by 21% from 1999 to 2019 (Figure 25), although this increase is not yet statistically significant.

Fresh water salmonids

There is an abundance of fresh waters in the UK, which provide extensive recreational fisheries on rivers and lakes. However, very few are commercial fisheries. Evidence of anthropogenic impacts of fresh water is widespread across the UK. A way of measuring the condition of fresh water is the abundance of fish. A proxy for this is to look at the numbers of fish caught in fresh waters.

The Environment Agency and Natural Resources Wales use catch and return data to access and manage salmon and sea trout stocks in a sustainable way. The Environment Agency states that salmon stocks in both England and Wales continue to be of concern, with 39 of 42 principle salmon rivers in England being "at risk" or "probably at risk" and 22 principle salmon rivers in Wales assessed as being "at risk" or "probably at risk".

In 1994, the number of salmon and trout caught by rod in Great Britain was 186,614 and 16% of those were released while 84% were retained. The numbers of fish being caught has fluctuated over the last two and a half decades and peaked at 200,782 in 2010. These numbers have since declined by 53% to 94,907 in 2019, with 89% released and 11% retained.

In 2019, the Environment Agency looked at the state of 64 principle salmon rivers in England and Wales and assessed them against the conservation limit, which is the minimum spawning stock level. This conservation limit measured the probability of the river's salmon population exceeding its conservation limit of four years out of five. Results in England showed 57% "at risk", 36% "probably at risk" and only 7% "probably not at risk". For Wales 73% were "at risk" and 27% were "probably at risk".

Between 1994 and 2019, the number of salmon caught in Scottish rivers has seen a decline of 43%. This compares with a 63% decline for England and Wales for the same period. In 1994, 108,476 fish were caught by rods in Great Britain but this has significantly declined by 48% to 56,678. Also, the number of fish retained has declined between 1994 and 2019. In 1994, 91% of the total salmon caught were retained compared with only 4% in 2019.

In comparison, the number of rod-caught trout has declined by 51% between 1994 and 2019. In 1994, 75% of trout were retained, with only 25% being released back into the rivers. This has changed significantly in 2019, with 14% being retained and 86% being released.

Fixed engine fishing is an ancient practice used in the UK as a general descriptor of stationary fishing gears, fixed to the bed or bank.

The total amount of salmon and trout caught by fixed engine has fallen by 84% between 1994 and 2018. There was a total of 213,474 fish caught by this method in 1994 and this has decreased 35,170 in 2018. Scotland has seen the largest decrease, with salmon fishing declining by 100% and trout fishing down by 97% between 1994 and 2018. England and Wales have seen smaller declines in the number of fish caught by fixed engine between 1994 and 2018, with declines of 84% for salmon and 60% for trout.

Although the numbers for fish caught by fixed engine method in England and Wales have fluctuated between 1994 and 2018 (with large increases in 2000, 2010 and 2015), the trend is downward and is now 73% lower in 2018 than it was in 1994. Scotland has also seen smaller fluctuations between 1994 and 2018 and is now 99% smaller in 2018.

National Forest Inventory (NFI) condition indicators

The National Forest Inventory (NFI) survey is based on data collected between 2009 and 2015 in Great Britain. Full details on the Forest Research's NFI survey methods are available. The data presented (Table 5) are for woodlands defined as native woodland, estimated to be 1,507,105 hectares in Great Britain. These are stands with 50% or more native tree species occupancy in the upper canopy that either form a discrete woodland parcel with a minimum area of 0.5 hectares, or form a woodland stand with a minimum area of 0.1 hectares that is part of a woodland that is 0.5 hectares or larger.

A varied tree age structure in woodland benefits biodiversity as differently aged trees provide different ecological habitats. For a woodland to be classified as favourable, it needs to have young, intermediate and old trees present.

The vertical structure is defined as the number of canopy storeys present. Woodlands with greater structural diversity provide a wider range of microhabitats and conditions, thus providing greater diversity of trees and other species.

Regeneration is an important indicator of biodiversity to predict the future health of woodlands. It is an assessment of seedlings, saplings and young trees. To be classed as favourable, the woodland areas sampled need to have trees with 4 centimetres to 7 centimetres diameter as well as having saplings and seedlings present.

Tree diseases and pests can have a negative impact on woodland biodiversity. While dead and decaying wood provides more light to reach the forest floor and an important micro-habitat, rapid widespread tree death can harm ecological health.

Landscape condition indicators

Habitat connectivity

Habitat connectivity is all about how well different species can move between habitats in the landscape. One definition is "the degree to which the landscape facilitates or impedes movement among resource patches". There is structural and functional connectivity. The structural is about the distribution of habitat patches across a landscape and the functional is about the ability of species to move around different habitat patches. For instance, birds might happily (functionally) move across a naturally (structurally) fragmented set of habitats many miles or even thousands of miles apart while some terrestrial mammals may struggle if a single road crosses their habitat.


A connectivity indicator was developed by Forest Research and the Centre for Ecology and Hydrology using the data from the Countryside Survey in 1990, 1998 and 2007. As this survey did not continue after 2007 it has not been possible to update this indicator, but a new one is under development.

This indicator was developed by looking at bird and butterfly species and their ability to move between patches. This developmental statistic showed an increasing trend in the connectivity of neutral grasslands. However, no significant change was shown in the connectivity of broadleaved, mixed and yew woodland.


A method has been developed to assess functional habitat connectivity at a national and regional scale and applying this method to calculate an indicator of habitat connectivity for Scotland; report contains a full methodology. Metrics were created for 10 catchment areas using Equivalent Connected Area (Probability of Connectivity) or ECA(PC), which shows the higher the value the greater the connectivity. The ECA(PC) as a percentage of total amount of habitat in the region was decided most meaningful way to present the connectivity value. It is not currently possible to detect changes in connectivity over time on a national scale because of inconsistent land cover data over time. However, this method has the ability to measure changes in connectivity at a local level over time.

Habitat connectivity varies considerably across Scotland for the different habitats (Table 7 and 8). Semi-natural grasslands for Forth and the North Highland have similar total areas, 60,359 hectares and 59,281 hectares respectively. However, the Forth has a much higher ECA(PC), indicating this area is better connected.


Initial research has been undertaken to identify habitat connectivity in Wales. Natural Resources Wales has developed CuRVe to investigate ecosystem resilience. CuRVe is an interactive map with many different layers, including connectivity, ecological networks and connectivity of overall layers. The 2016 State of Natural Resources Report identified that the mountain, moorland and heath habitat in Wales extends over 261,824 hectares. Fragmentation of this habitat has resulted in poor connectivity for lowland examples of mountain, moorland and heath habitats.

River naturalness

River naturalness refers to the extent to which a river has departed from its original state. By examining the chemical, biological, hydrological and physical habitat change indicators, we can begin to predict the overall condition of the river. This is particularly important as the quality of surface water bodies can affect many of the service flows, we, and natural life, derive from them.

A natural river may not create the largest ecosystem service flows and benefits to humans. Alterations, such as dams and weirs (a morphological change), may provide benefits to humans but may reduce the overall naturalness of the river as, for example, the migratory flow of fish up-stream is disrupted. Examining the naturalness of rivers assesses its ecological integrity. This is the river's ability to maintain and support its own ecological processes. Measures that quantify the overall naturalness of a river have been developed.


Natural England has developed the Priority River Habitat Map to assess the condition of a river. An aggregate score is comprised of hydrological, physical habitat, biological and chemical integrity scores. Each integrity score assesses a core component of a river's condition.

Figure 30 indicates that the naturalness score for 9,485 bodies of river water were limited by physical integrity. The number of river waterbodies limited by biological, chemical and hydrological integrity is 1,281, 897, and 452 respectively. The physical integrity represents the single largest cause of why a river is limited in its naturalness scoring.

Natural England are further developing an assessment of river naturalness and habitat resource. Additional detail of the development of this approach can be found here.


In Wales the naturalness of rivers was assessed using the Hydromorphological Impact Ratio, which looks at the channel substrate, flow regime, channel vegetation and geomorphic activity. The data were derived from the 2007 to 2008 baseline survey of Welsh rivers. South east Wales and south central Wales had the highest departure from naturalness for rivers (Table 9). These areas are linked to higher levels of urban modification. Generally, the rivers displayed higher levels of naturalness in the more rural settings of mid and north Wales.

A significant proportion of the engineering pressures on the water channels in Wales are from channel and bank resectioning, as shown in Table 10. There are a higher occurrence of engineering structures and resectioning in south Wales compared with north Wales.

Northern Ireland

We are currently unable to comment on the naturalness of rivers in Northern Ireland. Northern Ireland has 450 rivers and water bodies, which we hope to provide more statistical detail on in the future. However, data regarding Northern Irish Water Framework Directive assessments have been included in this publication for water quality.


Data on the physical status and ecological potential of all Scottish surface water bodies are presented in the Scottish River Basin Management Plans from 2009 and 2015 (SWB).

Table 11 shows that 25% of surface water bodies did not meet the "good" condition status (as targeted in the WFD) because of physical changes to the riverbeds and banks, such as straightening or culverting, and barriers to fish migration, such as dams and weirs.

Table 12 shows 255 water bodies in Scotland were substantially impacted in Scotland from modifications. Modifications for agricultural land use altered 242 of Scotland’s water bodies, representing the single largest cause for physical change to Scottish rivers. This illustrates a reduction in the naturalness of these water bodies.

Ancillary condition indicators

Protected sites

There are several formal designations, including Special Areas of Conservation (SAC) or a Site of Special Scientific Interest (SSSI) or Areas of Special Scientific Interest (ASSI) in Northern Ireland. The rare fauna or flora present or important geological or physiological features make it an area of interest to science.

In Northern Ireland the area of semi-natural habitat with a protected site designation of ASSI forms 24% of total semi-natural habitat area. Areas of semi-natural habitat with an SSSI designation comprises: 31% in England, 19% in Scotland and 27% in Wales of the total semi-natural habitat area. For the UK, 23% of the semi-natural habitat has an ASSI or SSSI designation.

England saw an increase in all semi-natural habitats area of SSSIs classed as favourable, except for coastal ones, from 2011 to 2019 (Table 13). All habitats had an increase in area classed as unfavourable. Wetlands rose from 3.2% of total area classed as unfavourable in 2011 to 6.6% in 2019. For a full breakdown of each habitat, see the condition supplementary tables.

In Scotland, the number of protected sites that have a favourable condition has increased for all semi-natural habitats (Table 14). Broadleaved, mixed and yew woodlands are the only habitat to have increased the number of sites classed as unfavourable, rising from 48 sites in 2007 to 161 sites in 2020.

In Wales, the only protected sites regularly monitored are the SACs (Table 15). However, insufficient sites were assessed in the 2013 to 2018 period to show if there has been any change since the 2007 to 2012 period. Most sites are classed as unfavourable, with only 4.7% being favourable when measured between 2001 and 2006, declining to 4.3% favourable in the period 2007 to 2012 for all semi-natural habitats (Table 15).

National Parks

In addition to protected sites, there are 15 National Parks in the UK. There are 10 in England, 3 in Wales and 2 in Scotland and these National Parks are to care for and protect unique habitats, rare wildlife and ancient monuments. Of the total area of National Parks in Great Britain (1.4 million hectares), 19% are semi-natural habitats (Figure 31).

Pressure indicators

Pressure indicators are defined here as damage inflicted on the landscape by humans. Wildfires can also be considered a pressure indicator. Most wildfires are anthropogenic in origin, with and without intent. In Wales, it was identified in 2018 and 2019 that around 7 out of 10 fires on woodland, grassland and crops were started deliberately.

Fire can be used as a management tool to control habitat diversity and disease control. Rotational burning is used on heather moorland for higher grouse breeding. However, this is controversial with opponents to burning on peatland, noting the impacts on the ecosystem services such as carbon storage and flood protection from burning the moorland.

There are two main sources of data on wildfires: reported fires and satellite data. Reported fires catch wildfires of all sizes attended by England's Fire and Rescue Services but may miss some remote fires that are addressed by land managers. Satellite data capture fires in both built up and remote places but might miss smaller fires under 30 hectares. 

Wildfires in England in open habitats (arable, improved grassland, semi-natural grassland, mountain, heath and bog) accounted for more than 70% of the total burnt areas for the financial years 2009 to 2010 to 2016 to 2017, with the exception of 2011 to 2012 where it accounted for 95% of the burnt area. The greatest number of incidents and area burnt occurring on semi-natural habitats (Table 16) occurred in 2011 to 2012, where there were 7,366 incidents over 12,415 hectares. This correlates to the drought in the same period and heat wave alerts in central, eastern and southern England and Wales.

Wildfires in Wales also had a high number of incidents in 2011 to 2012 (Table 18), as well as England. However, the Welsh data are not just for semi-natural grasslands, as they include all grasslands. The Welsh Government identified in July 2018 that there were almost nine times more fires on grassland, woodland and crops compared with July 2017. In July 2018, there were 40% more sunshine hours and half the rainfall that occurred in July 2017.

As with England and Wales, the drought in 2011 caused an increase in the number of wildfires in Scotland (Table 20). Between 29 April and 5 May 2011, the Highland and Island Fire and Rescue Services dealt with 70 significant wildfires. The woodland numbers includes all woodland, not just semi-natural ones.

Wildfires were investigated in Areas of Special Scientific Interest in Northern Ireland during the unusually dry spring and summer of 2011. A total of 3,801 hectares of designated semi-natural habitats were burnt.

The EU-wide reporting (Table 21) for the UK is based on satellite data. The periods of low wildfires correlate with heavy periods of rainfall in spring and summer, as well as wetter winters. A large number of small fires are excluded from the European Forest Fire Information System (EFFIS), meaning the number of fires is smaller.

Back to table of contents

4. Overall quantity and value of semi-natural ecosystem services

This section assesses the contribution semi-natural habitats services provides to the economy and society.

The value of semi-natural habitats ecosystem services was estimated at £7.0 billion in 2018 (2019 prices), as shown in Figure 33. Data are only available for the seven ecosystem services annual valuations in 2018. This is a partial valuation with exclusions such as tourism, education and flood protection for habitats other than broadleaved woodland.

Back to table of contents

5. Provisioning ecosystem services: quantity and value

Provisioning ecosystem services create products. For semi-natural habitats, these include food, water and renewable energies from wind and hydro.


Livestock grazing is one of the most common land uses for peatlands, however, most can only sustain light seasonal grazing. In the Fens, in the East of England, there is widespread use of conservation grazing cattle. Sheep can be farmed in almost every part of Wales because of their hardiness. In the uplands this may be the only feasible option. However, there are relatively low returns to farmers, despite low maintenance and capital costs.

Most livestock holdings in Wales are in Less Favoured Areas (LFA) for farming. A LFA is an European Union designation of an area of poor soils and low agriculture income. A significant proportion of Scottish land quality for agriculture is quite poor, with more than 5.73 million hectares (85%) of farmland classed as LFA. As a result, most of the agriculture is livestock grazing, with 3.6 million hectares classified as rough or common grazing.

The estimate for food from semi-natural habitats is derived from the different land uses data from the Land Cover Map 2015 and data from the Farm Business Survey (FBS). The FBS provides data for England on the outputs from agriculture, which excludes subsidies and agri-environment (a scheme that compensates farmers for the costs of carrying out wildlife friendly management on their land), and costs for agriculture excluding agri-environment activities and data on farm area. A rate is then calculated per hectare and applied to the different land use classifications for the UK. The method of valuing estimates for food from farming are currently being developed using farming rents for a future publication.

Agriculture on semi-natural habitats shows a negative contribution for ecosystem services (Table 23). This is only a crude estimate to give an idea of scale of agriculture on these habitats. Grazing is often used as a conservation management tool for improving the ecology, to provide a range of benefits beyond food production. Natural England identified a need for grazing on grassland, heathland and coastal marshes to maintain their structure and composition.


Surface sources such as upland reservoirs and lowland rivers provide around 70% of the UK's drinking water, while 30% is sourced from ground water. The uplands areas have around 80% of the large dams for public water supply.

Peat is dominant on the higher ground in the uplands, therefore, a significant proportion of the UK's water supplies lands or flows through peatlands. It was estimated in the UK Natural Capital for Peatlands that 27% of the UK's drinking water supply was sourced from peatlands. It is currently estimated around 80% of the UK's peatlands are in a damaged and deteriorating condition. Water sourced from degraded peatland has dissolved organic carbon and peat sediment, which represents a large cost to water companies to remove it.

In the UK, water abstraction for public water peaks in 2005 at 7,345 million cubic metres (Figure 34). A possible reason for the decline from 2005 is the more efficient and sustainable use of water, as advocated in the Water Act 2003. However, there has been a steady rise from a series low in 2014 of 6,443 million cubic metres to 6,796 million cubic metres in 2018. This has been driven by an increase in abstraction in England.

Annual monetary estimates are based on resource rents calculated for the Standard Industrial Classification (SIC) subdivision class: water collection, treatment and supply. The resource rent can be interpreted as the annual return stemming directly from the natural capital asset itself, that is, the surplus value accruing to the extractor or user of a natural capital asset calculated after all costs and normal returns have been considered. The annual value fluctuates over the time series, being valued at £3,355 million in 2018. The UK asset value in 2018 was £98,178 million.

Renewable electricity generation


The typical environment favoured for hydro schemes are in semi-natural upland areas. The number of gigawatt hours (GWh) varies over the time series (Figure 36). The reduction in power generated in 2010 (3,591 GWh) is in the year where there was a reduction of 30% in rainfall between 2009 and 2010.

Other factors, in addition to rainfall, also affect the amount of hydro power produced, including the timing of the rainfall and the local groundwater conditions. In 2019, hydro generation accounted for 5% of all the renewable electricity generated in the UK.

The number of hydroelectric sites in the UK has increased from 339 in 2008 to 1,520 in 2019, with 50% of those sites in Scotland. This is expected to increase as numerous applications are currently going through the planning process and awaiting construction.

As at September 2020, in Scotland there were three applications submitted, 25 sites were awaiting construction and three sites were under construction. In England, two applications were submitted, eight were awaiting construction and one was under construction. In Wales, one application was submitted, two were awaiting construction and one was under construction.

The annual value from hydroelectricity was £61 million in 2018. Part of the annual variation is because of fluctuations in the residual value resource rent approach used (see methodology publication) in addition to variations in the generation of gigawatt hours (GWh) produced. The asset valuation for hydroelectricity has increased by 97% between 2008, when it was £919 million, and 2018, when it was valued at £1,808 million.

Onshore wind power generation

Electricity generation from onshore wind turbines on semi-natural habitats is approximately six times greater in 2019 than it was in 2008. In 2019, electricity generation on semi-natural habitats was 6,449 gigawatt hours (GWh), a rise from 1,129 gigawatt hours in 2008. The apportioning of onshore wind turbines to semi-natural habitats uses mapping of wind farm locations involving a combination of Department for Business, Energy and Industrial Strategy (BEIS) wind farm points and Ordnance Survey wind turbine points (see methodology publication for full details).

Between 2015 and 2016, there was an 11% reduction in the generation of electricity from wind, which can be attributed to the fall in average wind speeds from 9.4 knots to 8.4 knots. This is also reflected in the wind generation load factor, a measure of generation efficiency being the utilisation of total generation capacity, which decreased from 29% to 24%.

Scotland generates 95% or 6,138 gigawatt hours (GWh) of the total electricity produced on semi-natural habitats, while England generates 3% or 201 GWh and Wales generates 1% or 109 GWh.

Comparison between onshore wind generation on all sites with semi-natural habitats

Between 2008 and 2019, the electricity generation from semi-natural habitats has been consistent. In 2012, semi-natural habitats produced the highest generation of electricity (at 21.2%) before declining for the next four years to the lowest level of 19.4% in 2016. This has since recovered and in 2019, 20% of electricity was generated on semi-natural habitats.

18% or 1,068 gigawatt hours (GWh) of the UK's electricity generation from onshore wind turbines was produced in Scotland's semi-natural habitats in 2008. This steadily increased and peaked in 2012, at 20.3%, before decreasing again. It then stood at 19% or 6,138 GWh in 2019. Semi-natural habitats in England and Wales produced 1.0% and 0.3% respectively in 2019.

Annual value

The annual value of onshore wind generation on semi-natural habitats increased from £3.8 million in 2008 to £36 million in 2018 and is now approximately 10 times larger.

Scotland’s proportion of the annual value has increased from £3.6 million in 2008 to £34.3 million in 2018 and accounts for 95% of the total Great Britain valuation. England’s share of the annual value in 2003 was £0.1 million and has increased to £1.2 million in 2018 and accounts for 3.3% of the total valuation. Wales’ share of the annual value has increased from £0.1 million to £0.5 million in 2018 and accounts for 1.5% of the Great Britain valuation.

The asset value of onshore wind generation on semi-natural habitats is £535 million in 2018 (2019 prices).

Back to table of contents

6. Regulating ecosystem services: quantity and value

This section discusses the benefits provided by the regulation of natural processes, including air quality regulation, climate regulation, pollination and natural hazard regulation such as flood mitigation.

Carbon sequestration

Vegetation can sequester carbon, that is, remove it from the atmosphere. Capacity for vegetation to do this depends on its characteristics. Forests capture the carbon dioxide from the atmosphere and through photosynthesis transform it into biomass (for example, wood) and so carbon accumulates in the form of biomass, deadwood, litter and soils.

We use the Department for Business, Energy and Industrial Strategy's (BEIS) non-traded carbon prices, which approximate to the cost of removing carbon by other means -- including reducing emissions. This valuation does not consider the impacts to global society outside of the UK of carbon sequestration.

The amount of carbon sequestered has risen from 11,871 thousand tonnes carbon dioxide (CO2) equivalent removed in 1998 to 12,420 thousand tonnes CO2 equivalent in 2018.

Broadleaved woodland removed approximately four times more carbon in the UK than semi-natural grasslands. The value of carbon has been increasing over time, with the annual value at £867 million in 2018 (Figure 45). The asset value of carbon sequestered by semi-natural lands in 2018 is £44,036 million (2019 prices).

This analysis does not capture all carbon sequestration for all semi-natural habitats. We currently only have data on broadleaved woodlands and semi-natural grasslands. A lack of data means values for sequestration by saltmarsh ecosystems are not included in current estimates.Peatlands, which are a significant source of emissions, are only partially represented in the data.

A report for the Department for Business, Energy and Industrial Strategy (BEIS), estimates that emissions from damaged peatland (23 million tonnes of CO2 equivalent) negate all terrestrial carbon sequestration in the UK. For more information on data gaps, please see the methodology guide.

Air pollution removal

The removal of air pollutants by vegetation is measured, in these accounts, in terms of the avoided healthcare costs associated with exposing the public to the pollutants removed. Air pollution is associated with a range of cardio-vascular and respiratory diseases in humans. The pollutants covered in pollution removal are:

  • PM2.5

  • PM10 (which includes PM2.5)

  • nitrogen dioxide (NO2)

  • ground-level ozone (O3)

  • ammonia (NH3)

  • sulphur dioxide (SO2)

The most harmful pollutant is PM2.5 (fine particulate matter with a diameter of less than 2.5 micrometres, or 3% of the diameter of a human hair), which can bypass the nose and throat to penetrate deep into the lungs, leading to potentially serious health effects and healthcare costs.

An estimated 480,000 tonnes of pollution was removed in 2018 by semi-natural vegetation (Figure 46).

The pollution removal annual value has fallen from £951 million in 2009 to £634 million in 2018 (Figure 47). These changes are not driven by changing conditions or the extent of vegetation but are linked to falls in UK air pollution levels.

Although the removal of PM2.5 represents a small part of total pollution removed, it represents nearly 90% of the avoided health impacts. The physical amount of pollution has remained fairly flat, however, the value of this has declined because of the reductions of PM2.5 in the atmosphere to remove.

In 2018, it is estimated that the asset value of pollution removal was £38.9 billion in semi-natural habitats.


Pollination by wild species provides benefits to the UK population through enhanced crop productivity, quality and nutrition, as well as its biodiversity benefits further up the food chain.

The National Pollinator Strategy: 2014 estimated pollinators added approximately £600 million per year to the value of UK crops, through increased yield in oilseed rape and the quality of various fruit and vegetables. A previous study, in 2007, estimated insect pollinators contributed around £430 million in production value to the UK economy or around 8% of the total market value of crop production.

Beyond the direct effects of pollinators on UK market values, movement by pollinating insects from semi-natural habitats to agricultural areas expands the positive externality effects of pollinator presence. Semi-natural habitats support ecosystem services beyond its immediate vicinity to produce additional benefits, including pollination and biological control. The establishment of quality semi-natural habitats, and insect movement from them, provides increased levels of pollination in neighbouring agricultural areas, improving marketable yields.

Flood mitigation


Forests are known to reduce flood flows, according to a systemic review  by the UK Centre for Ecology and Hydrology, which looked at 71 studies. There is broad support for the conclusion that increased tree cover in catchments results in decreased flood peaks, and that decreased tree cover results in increased flood peaks.

To capture the flood regulating service for woodland in Great Britain, Forest Research examined how much it would cost to install equivalent man-made flood water storage (that is, reservoirs) in an area where there was no woodland. For more information on the method, see the methodology report.

An annual average estimated that it would cost £218.5 million per year to create reservoirs able to capture that water if woodland in flood risk catchment (FRC) areas were replaced with grassland. In this study, 83% of the land area in Great Britain fell within an FRC area. The total Great Britain asset valuation of this service over 100 years equated to £6,513 million, in 2018 prices. However, there are a number of caveats with this calculation. For all limitations with this research, see the methodology report.

Semi-natural woodlands formed around 54% of woodlands and a crude estimate of the annual value, at 2019 prices, stands at £120 million, with an asset value of £3,583 million.


Saltmarshes provide a range of ecosystem services, such as facilitating waste remediation, carbon sequestration, providing seascapes for individuals to enjoy, and mitigating flood hazards. Further research is required to provide more accurate data on the extent of and ability of saltmarshes to allay flood risk.

We aim to provide estimates on the economic benefits of UK saltmarsh flood mitigation in the future and appropriate methods for their valuation. Research identified three factors for the capacity of a saltmarsh to provide natural hazard protection:

  • width of saltmarsh

  • distance travelled by waves

  • habitat contiguousness and degree of homogeneity

Back to table of contents

7. Cultural ecosystem services: quantity and value

Cultural ecosystem services provide non-material benefits, such as enjoyment of the landscape and its wildlife, recreation, education and cultural heritage.


The estimates for outdoor recreation refer to people aged 16 years and over and exclude overnight stays. We measure both the physical flow of recreation as the amount of time people spend outdoors as well as how much money people spend to go.

In 2018, it is estimated that 1,670 million hours were spent in the semi-natural habitats of woodland, mountains, moorland and hills and freshwater (Figure 48). This has increased significantly over the time series, from 1,237 million hours in 2009. It is an underestimate as not all semi-natural grasslands are included; we are unable to separate out the visits classed as farmlands that are on grasslands within the farm.

The annual monetary value of visits to semi-natural habitats fluctuates over time. This is because of reductions in expenditure per visit, with travel costs declining and people choosing cheaper outdoor activities. Semi-natural habitat visits in 2018 were estimated to be worth £1,893 million, excluding visits to semi-natural grasslands that may have been classed as visits to farmland. The majority of the value of visits is in mountains, moorland and hill areas (£1,164 million).

The asset value of recreation in semi-natural habitats was estimated to be £82,742 million in 2018. That is the valuation of the stream of services that are expected to be generated over the lifetime of the asset.

Back to table of contents

8. Asset value of semi-natural habitats

The asset account includes selected semi-natural habitats ecosystem services, valued at £270 billion in 2018 (Table 24). See methodology publication for full details.

Back to table of contents

9. Semi-natural habitat, UK data

Semi-natural habitats, condition indicators
Dataset | Released 10 February 2021
A short summary of the dataset and what it covers.

Semi-natural habitats, ecosystem services
Dataset | Released 10 February 2021
Physical (non-monetary) and monetary estimates of the ecosystem services provided by natural assets in the UK between 1998 and 2019.

Back to table of contents

10. Glossary


Asset valuation is an estimate of the stream of services that are expected to be generated over the life of the asset. It looks at the pattern of expected future flows and the time period over which the flows of values are expected to be generated.

Ecosystem services

Ecosystem services are the flows of benefits that people gain from natural ecosystems. This includes provisioning services such as food and water; regulating services such as flood protection and pollution removal; and cultural services such as recreational and heritage.

Natural capital

Natural capital is a way of measuring and valuing the benefits that the natural world provides society. These benefits from natural resources include food, cleaning the air of pollution, sequestering carbon and cleaning fresh water.

Back to table of contents

11. Measuring the data

In this release, the semi-natural habitat accounts are presented in four sections:

  • the size of the area covered by semi-natural (extent account)

  • indicators of the quality of the semi-natural ecosystem and ability to continue supplying services (condition account)

  • quantity and value of services supplied by the semi-natural ecosystem (physical and monetary ecosystem service flow accounts)

  • value of semi-natural habitat as an asset, which represents the stream of services expected to be provided over the lifetime of the asset (monetary asset account)

The data underpinning semi-natural habitats, natural capital, UK come from a range of sources with different timeliness and coverage. This release is based on the most recent data as at December 2020.

Data sources include:

  • Bat Conservation Trust

  • British Trust for Ornithology (BTO)

  • Bumblebee Conservation Trust

  • Butterfly Conservation (BC)

  • Countryside Survey

  • Department for Business, Energy and Industrial Strategy (BEIS)

  • Department for Environment, Food and Rural Affairs (Defra)

  • Environment Agency

  • European Forest Fire Information System (EFFIS)

  • Forest Research

  • Forestry Commission Scotland

  • Forestry England

  • Joint Nature Conservation Committee (JNCC)

  • Natural England

  • NatureScot

  • Marine Scotland

  • Natural Resources Wales

  • Rothamsted Research

  • UK Centre for Ecology and Hydrology

  • UK National Atmospheric Emissions Inventory (NAEI)

The Office for National Statistics' (ONS's) natural capital accounts are produced in partnership with the Department for Environment, Food and Rural Affairs (Defra).

An image of the Department for Environment, Food and Rural Affairs (Defra) logo.

Back to table of contents

12. Strengths and limitations

Data quality

The ecosystems services are Experimental Statistics. Currently, there is no single data source for the UK for the individual ecosystem services. They are calculated from data from the four countries with different timeliness.

Ecosystems provide a diverse range of services and not all have been included in this publication, either owing to unavailability of data or the need for new methods of evaluation. We intend to continue to develop our ability to report on all services.

Back to table of contents

Contact details for this Statistical bulletin

Hazel Trenbirth and Adam Dutton
Telephone: +44 (0) 1633 580051