Go to content

Information about the service

Welcome to Luonnontila!

Welcome to view the state of Finnish nature through indicators divided by main environment types.

The questions formulated as headings below describe the background, functionalities, and key operating principles of the Luonnontila web service. You can get acquainted with each topic by clicking on the links below or by scrolling down the page.

Who are the parties behind the Luonnontila web service?

The Luonnontila web service is a collective effort of several experts and research institutions operating in the environmental field, striving to present an up-to-date picture of the state of Finnish nature. Key players behind the content and indicators of the Luonnontila web service include the Finnish Environment Institute(you are switching to another service), the Finnish Museum of Natural History(you are switching to another service), the Natural Resources Institute Finland(you are switching to another service), and Metsähallitus(you are switching to another service). The Finnish Environment Institute is responsible for the maintenance of the Luonnontila web service and its related coordination. If you have any questions or comments related to the entirety of the Luonnontila web service, please contact the persons listed on the “Contact Us” pages.

The re-design of the Luonnontila web service has been implemented with the support of the Finnish Ecosystem Observatory(you are switching to another service), funded by the Ministry of the Environment(you are switching to another service).

Why does the Luonnontila web service exist?

The Luonnontila web service was launched for the first time in 2010. It was originally designed to facilitate the Clearing-House Mechanism of the Convention on Biological Diversity. The basic principle of the Luonnontila web service, with indicators divided by main habitat types, traces back to this first version of Luonnontila.

However, the previous version of the Luonnontila relied on substantial manual updates, which made its long-term maintenance practically impossible. Moreover, by the 2020s, the website’s technical solutions had reached their end. Yet, the demand for and relevance of nature related information and data in societal discussion clearly continues to grow in the 2020s, and there have been no easily accessible services that compile various sources to communicate about the state of Finnish nature other than the Luonnontila web service.

What’s new?

In the updated Luonnontila launched in October 2023, several significant improvements have been made. One of the most crucial developments is the automation of indicator calculations. Nowadays, many datasets related to nature and its diversity are freely available in machine-readable interfaces from various operators. This allows the production of indicators to be executed on servers and to repeat the analysis whenever the dataset is updated.

A central reform is also the quantitative definition of the current status and recent development of an indicator. As these two features are aimed to be defined for the indicators, they can be placed in a quadrant, from which the state of Finnish nature can be viewed at a glance. This overview is available on the “Luonnontila web service Overview” page. The principles for defining the state and direction are described in later sections of this page.

Another novelty is the integration of experts from various research and expert institutions into the introduced indicators. In this way, the Luonnontila web service aims to ensure that the state of nature is communicated in a multifaceted manner, considering various perspectives. Experts are also given a voice in the site’s new blog-like “Perspectives from Experts” section.”

Why indicators?

In the Luonnontila web service, the approach chosen to form an overall picture of the state of nature is through indicators. Indicators are thought to condense crucial information related to habitats, such as, for example, the physio-chemical state of a particular habitat, the population fluctuations of species specialized in that habitat, or processes affecting them, such as the effects of human activity.

Indicators enable the quantitative measurement of the state of nature and, therefore, also the setting of concrete goals related to nature. To be effective, indicators must have a clear connection to the phenomenon of interest and be sufficiently comprehensive both temporally and spatially.

However, indicators are rarely perfect, which must be taken into account when drawing conclusions. For this reason, it’s more reliable to examine several indicators related to the same phenomenon, but which are based on different datasets.

What does the ‘status’ of an indicator mean?

For most indicators, the defined “Status” describes the current state of the indicator in relation to a reference value. Most often, when determining the status of an indicator, the average values of the indicator over the most recent five years are examined and related to this reference value. The reference value, in turn, can be determined mainly in three different ways:

These definitions conceptually differ from each other significantly, and depending on the definition method, very different status assessments for the same indicator might emerge. The latter, in particular, is very different from the first two. For each indicator in Luonnontila, there is an effort to clearly express the reference point according to which the status has been defined for that particular indicator.

However, defining the status of an indicator is challenging in many cases because, for many features or measures, we might not have reliable reference assessments available based on any of the above principles. For some indicators, the status might need to be left undefined due to missing reference values or too short time series.

What do the classifications used to describe the status mean?

The states of the indicators have been classified on a scale of ‘very good‘, ‘good‘, ‘satisfactory‘, ‘poor‘, and ‘very poor‘. The absolute value of the status that leads to each category depends on the scale in which the indicator values vary. Currently, there are two different classification criteria in use. One has been developed for indicators varying from zero to infinity, and the other is for indicators that vary from zero to one or depict a percentage share of the fulfillment of a certain condition.

Defining the status for indicators varying from zero to infinity

Many indicators inevitably have positive values, but their upper limit of variation is not theoretically restricted. The classification of such indicators considers the possibility that the status can be significantly better than the reference value.

  • Very Good: The indicator value has improved by more than 20% relative to the reference value.
  • Good: The indicator value has deteriorated by no more than 5% or improved by up to 20% relative to the reference value.
  • Satisfactory: The indicator value has deteriorated between 20% and 5% relative to the reference value.
  • Poor: The indicator value has deteriorated between 20% and 50% relative to the reference value.
  • Very Poor: The indicator value has deteriorated by more than 50% relative to the reference value.

Examples of indicators following this classification are based on indices depicting average species populations derived from systematic species monitoring, such as the mire bird indicator.”

Defining the status for indicators varying between zero and one or depicting a percentage share

Many indicators depict a share of a certain set. For such indicators, it is essential to consider that the indicator is strictly limited between zero and one (or a hundred percent).

  • Very Good: 95-100% meets the desired condition.
  • Good: 80-95% meets the desired condition.
  • Satisfactory: 65-80% meets the desired condition.
  • Poor: 50-65% meets the desired condition.
  • Very Poor: less than 50% meets the desired condition.

Examples of indicators following this classification are those that describe the protection level of species mentioned in the annexes of the EU’s Habitats Directive (for instance, the indicator of the protection level of species under the Habitats Directive).

Calculating status relative to the reference value

In addition to defining the reference values and classification criteria, calculating the status estimate poses its own challenges. In the simplest situations, the status estimate is absolute if the indicator value accurately represents the set it aims to depict. For example, the percentage of species at a favorable conservation status under the EU Habitats Directive is accurate in the sense that all relevant species under the Directive’s annexes have been included in the percentage calculation, and the indicator is not considered to represent a broader imaginary set of species.

However, indicators are most often based on some form of sampling and the idea that the indicator represents a larger whole than the measured set. In these cases, there is uncertainty associated with the indicator value, which must be considered when defining the status. Besides the indicator value, there is often also uncertainty associated with the reference value, which is also frequently based on some form of sampling that is believed to best represent the reference status.

When there’s uncertainty associated with the indicator value and/or the reference value, the status is estimated in two stages. In the first stage, it’s determined whether the indicator value is with high probability better, worse, or the same as the reference value. In the second stage, the minimum deviation from the reference value is determined.

Stages of status calculation when there is uncertainty

In the first stage, it is checked whether the status of the indicator is better or worse than the reference value. This is done using the 90% credible interval of the statistical model estimate used for status determination. If the 90% credible interval includes zero, the status cannot be confidently assumed to clearly deviate from the reference value. If the credible interval is very wide on both sides of zero (e.g., spanning several status classes), the status may be left undetermined due to high uncertainty.

If, in addition, the 90% credible interval spans two different status classes (e.g., good and satisfactory), the status can be determined based on which status class the majority of the posterior distribution of the estimate lies in. However, uncertainty related to this classification is always noted in the context of the indicator.

In the second stage, the minimum deviation of the status with 90% probability is determined. If in the first stage the conclusion is that the status has deteriorated compared to the reference value, the minimum deterioration with 90% probability is estimated from the distribution (the ninetieth percentile of the posterior distribution). Conversely, if in the first stage the conclusion is that the status has improved compared to the reference value, the minimum improvement with 90% probability is estimated from the distribution (the tenth percentile of the posterior distribution).

What does the development of an indicator mean?

The development of an indicator describes the direction and the rate at which this particular indicator has on average evolved since the year 2000. The rate of development is expressed as the average annual change. The average annual change can be stated relative to the value of the previous year (percentage change from the previous value of the indicator) or as an absolute annual change.

What do the classifications used in describing the development mean?

The development of indicators is classified on a scale of rapidly improving, improving, stable, deteriorating, and rapidly deteriorating. Which calculated value for development leads to which category depends on the scale on which the indicator values vary. So far, there are two different sets of classification criteria. One has been developed for indicators varying from zero to infinity and the other for indicators varying from zero to one or depicting a percentage share of some condition being met.

Classification of development for indicators varying from zero to infinity

Many indicators inevitably have positive values, but at least theoretically, their upper limit of variation is not restricted.

  • Rapidly improving: The annual change in the indicator’s value is on average more than 5% in a favorable direction.
  • Improving: The annual change in the indicator’s value is on average 0-5% in a favorable direction.
  • Stable: There’s no clear annual change in the indicator’s value.
  • Deteriorating: The annual change in the indicator’s value is 0-5% in an unfavorable direction.
  • Rapidly deteriorating: The annual change in the indicator’s value is more than 5% in an unfavorable direction.

Examples of indicators following this type of development classification are based on indices depicting average species populations produced from systematic species monitoring (e.g., the indicator for peatland breeding birds).

However, it’s important to note that equivalent annual percentage changes for better and worse lead to different results on an absolute scale. For instance, a 2% annual population growth over 20 years leads to almost a 50% overall growth in population, but a 2% annual population decline over the same period results in only about a 33% decline in the population.

Nevertheless, the annual percentage change is ecologically intuitively, as especially in indicators related to population development, its interpretation is closely linked to changes in individual-level reproductive and/or survival potential and population growth rate.

Classification of development for indicators varying between zero and one or depicting a percentage portion

Many indicators depict a portion of a certain set. The values of such indicators are strictly confined between zero and one (or a hundred percent), which must be considered in statistical models.

  • Rapidly improving: The favorable portion of the indicator increases by more than 5% annually.
  • Improving: The favorable portion of the indicator increases by 0-5% annually.
  • Stable: There’s no clear annual change in the favorable portion of the indicator.
  • Deteriorating: The favorable portion of the indicator decreases by 0-5% annually.
  • Rapidly deteriorating: The favorable portion of the indicator decreases by more than 5% annually.

Examples of indicators following this type of classification are indicators depicting the conservation level of species mentioned in the annexes of the EU Habitats Directive (e.g., the indicator for the conservation status of species in the Habitats Directive).

As in the case above, when dealing with percentage changes, one must keep in mind that equivalent percentage changes for better and worse result in different amounts of change on an absolute scale.

Furthermore, for indicators with values varying within a restricted range, there’s a distinct feature: if the state of the indicator is nearing its maximum (e.g., 100%), it can no longer develop rapidly in a favorable direction. Therefore, such indicators cannot be in an excellent state and develop rapidly towards a better state at the same time.

Calculating Development

The assessment of development is almost always carried out with the help of statistical models. From the fitted statistical model, the direction and rate of the recent development (starting from the beginning of the 2000s) are assessed. Depending on the indicator, the estimate obtained from the statistical model might be further converted into an estimate of the annual percentage change.

Since development is assessed using a statistical model, there is always uncertainty associated with the estimate, and this must be taken into account when classifying the state. As with the state assessment, the assessment of development is also carried out in at least two stages. In the first stage, it is determined whether the indicator has developed in an improving or deteriorating direction, or if the value has remained stable. In the second stage, if the model suggests that there seems to be some kind of development trend in the indicator, the minimum change in the state is determined. In some cases, the estimate describing the development might need to be converted into an estimate for the annual percentage change.

Calculating development when the indicator varies between zero and infinity

In the first stage, it is checked whether the statistical model supports the idea that the indicator has developed in any direction. This is done using the 90% credible interval of the estimate from the statistical model used to determine the direction of development. If the 90% credible interval for the annual development estimate includes zero, it cannot be confidently stated that the indicator has developed in any direction and is interpreted as stable. On the other hand, if the credible interval for the development direction is very wide on both sides of zero, the development direction may also be left undetermined due to high uncertainty. This can especially occur with short time series data or data collected with a sparse sample.

In the second stage, the minimum deviation of the state is determined with 90% probability. If, in the first stage, it has been concluded that the indicator has deteriorated since the 2000s, the minimum annual deterioration with a 90% probability is assessed (based on the 90th percentile of the posterior distribution). On the other hand, if in the first stage it has been concluded that the indicator has improved, the minimum annual improvement with a 90% probability is assessed (based on the 10th percentile of the posterior distribution), which is compared to the development classification criteria described above.

Calculating development when the indicator varies between zero and one or represents a percentage share

As with indicators that vary between zero and infinity, when modeling shares, it is first checked whether the statistical model supports the idea that the indicator has developed in a particular direction. This is done using the 90% credible interval of the estimate from the statistical model used to determine the direction of development. If the 90% credible interval for the annual development estimate includes zero, it cannot be confidently stated that the indicator has developed in any direction and is interpreted as stable. Due to high uncertainty, the direction of development can also be left undetermined if, for example, the data are insufficient (short time series or sparsely sampled data) for assessing the development direction.

In the second stage, the minimum deviation of the state is determined with a 90% probability. If, in the first stage, it has been concluded that the indicator has deteriorated since the 2000s, the average minimum annual deterioration with a 90% probability is assessed (based on the 90th percentile of the posterior distribution). On the other hand, if in the first stage it has been concluded that the indicator has improved, the minimum annual improvement with a 90% probability is assessed (based on the 10th percentile of the posterior distribution).

When modeling shares, a third stage is introduced in which the model’s estimate depicting minimum deterioration or improvement is converted to an estimate of the average annual percentage change.