Andorra Biodiversity: Habitats and Conservation

Andorra’s biodiversity is concentrated along steep gradients. In a short distance, the terrain rises from densely occupied valley floors through farms, river corridors and forests to alpine grasslands, wetlands, rock and permanent snow. Temperature, rainfall, exposure and land use change with elevation, creating many ecological niches in a small country. The result is not one continuous wilderness but a patchwork of natural and human-shaped habitats whose condition depends on how they connect.

Knowing how much biodiversity Andorra has is less straightforward than quoting a species total. Checklists record what has been identified; Red Lists assess extinction risk under a defined method; monitoring schemes look for change over time; habitat maps classify the places where species live. Each answers a different question. The evidence assembled so far is substantial, but it is uneven across taxonomic groups, locations and years. Understanding those limits is part of understanding the conservation task.

A small territory with a steep ecological gradient

Andorra lies in the eastern Pyrenees, where elevation and topography create marked differences over short horizontal distances. Valley bottoms receive settlements, roads, agriculture and river corridors; slopes carry varied forest and scrub; higher ground opens into subalpine and alpine grasslands, wetlands, lakes and exposed rock. Climate, soil, water regime and human use all help determine which plant communities and animal species can persist at a particular site. A valley aspect can receive different sunlight from the opposite slope, while altitude changes growing conditions and snow cover. The national-scale pattern is therefore a mosaic, not a simple belt of vegetation arranged neatly by height.^S001^S002

The National Biodiversity Strategy of Andorra (ENBA), approved by the Government in 2016, described up to 180 habitat types based on the country’s 2012 digital habitat map. The categories included wetlands such as lakes, rivers and peatlands; shrublands and heath; meadows and pasture; forests ranging from Scots pine and mountain pine to oak and mixed woodland; rocky ground and scree; and agricultural land. The same strategy stressed that farmland can hold distinctive biodiversity shaped through human activity. The 180 figure belongs to that cited mapping work and classification, not to a current census of separate ecosystems or species.^S001

The contrast between alpine habitat and a worked valley floor can be misleading if human influence is treated only as a pressure. Some habitats are shaped by grazing, mowing, cultivation or the maintenance of open land. A pasture is not simply a forest that has not yet grown; its plant communities and the species using it can depend on how it is managed. At the same time, abandonment or intensification can change those conditions. Conservation planning must identify which ecological values are present, what processes maintain them and whether intervention is needed. “Natural” and “human-made” are not reliable synonyms for low and high biodiversity.^S001^S003

Rivers and wetlands make the gradient particularly visible. Water connects a slope to a stream and a stream to a downstream valley, carrying sediment, nutrients and organisms across habitat boundaries. A wet meadow may depend on a seasonal water table; a riverbank supports a different set of plants from a dry slope beside it. Ponds, lakes and peatlands can function as small but specialised habitat units. The ecological value of such a place is not measured only by its surface area: hydrological connections, surrounding land use and the species able to reach it can influence its condition.^S001^S004

This patchwork also explains why a countrywide map and a local field study do different jobs. A national map can locate broad habitat types and support land-use decisions. A specialist survey can look for a threatened plant, a breeding amphibian or a community of insects in a particular season. Neither scale replaces the other. National overviews can hide small sites and local variation; isolated site studies do not necessarily reveal how common a habitat is across Andorra or whether it is changing over time.

Habitat maps are classifications made for a purpose and at a chosen scale. The Government’s catalogue describes the 2002 Andorran habitat map as a digital product at 1:25,000, with a legend of around one hundred units based on plant communities. The later ENBA cites a 2012 digital map and says up to 180 habitat types can be distinguished. These figures should not be combined as if they came from a single unchanged classification. They refer to different map editions or analytical contexts, and the available catalogue entries do not provide a complete crosswalk between categories. A revised legend can split one broad unit into several classes without any physical habitat having appeared overnight.^S001^S002

Scale determines what a map can resolve. A small wetland or narrow strip of riverbank may be ecologically important yet too small to appear as a distinct polygon on a national map. A mapped polygon can include internal variation that the map’s minimum unit does not show. A class is also a model of ecological similarity, not a guarantee that every part of the polygon has identical plants, water conditions or conservation status. For land-use screening, the map may identify where a closer survey is needed; it cannot replace field inspection before a decision affecting a specific plot.^S001^S002

Updating a map requires repeated classification and a way to distinguish real land-cover change from methodological change. If a later edition uses sharper imagery, a new field key or a different minimum mapping unit, the apparent boundary of a habitat can shift even where ecological conditions have changed little. Conversely, a stable mapped boundary does not prove that quality is stable: species composition, hydrology, grazing intensity or invasive plants may change inside the same outline. A robust comparison needs harmonised classes and supporting observations, not just two coloured maps placed side by side.^S001^S003

What a species total can—and cannot—say

Species totals are useful only when their scope is clear. The ENBA cited historical national data that included 1,584 plant species, 745 fungi, 382 lichens, 237 bryophytes, 134 molluscs, 988 arthropods and 238 vertebrates. These numbers reflected the studies and lists available to the strategy, including a 2002 breeding-bird atlas reference; they were not a single survey carried out in one year with identical effort for every group. Some groups had more complete catalogues than others, and taxonomic knowledge continues to change.^S001

A later flora page maintained by the Government and Andorra Research + Innovation (AR+I) explains that the 2008 vascular-plant checklist began with 1,537 native and well-established non-native taxa, while noting that the presence of nearly 200 additional plants could not then be confirmed. Subsequent publications have added to the checklist. The same page says that the list and the flora Red List are periodically updated through an agreed protocol. AR+I’s project information describes a 2021 Andorran Flora Master List with 1,604 taxa, of which 309 were in threatened categories and 54 were Data Deficient. These are version-specific figures, not a timeless final total.^S005^S006

Several details matter when comparing those numbers. A taxon may mean a species or another recognised unit such as a subspecies. “Native”, “introduced”, “established” and “observed” are not the same status. A checklist can include species present in the country without implying that each has been recently surveyed at every potential site. A later report may add a taxon because a botanist identified it, because its classification changed, or because a previously uncertain record was verified. Without the edition date and definitions, a difference between totals cannot be interpreted as a population increase or decline.^S005^S006

The Global Biodiversity Information Facility’s Andorran node provides a complementary kind of evidence: occurrence records supplied by Andorran and foreign institutions. AR+I coordinates the national node, whose purpose includes gathering and standardising data for global access. The portal says records can be searched and downloaded, subject to the Creative Commons licence shown for each dataset. An occurrence record documents a report at a place and time; it is not, on its own, proof that a population remains there today or that the species is common across the country.^S007

That distinction is essential for maps. A dense cluster of records can reflect popular or accessible survey locations, institutional collections or a long-running monitoring route. An empty grid square could mean that the species is absent, that suitable habitat is absent, or that no one has surveyed it adequately. Presence-only data are valuable for distribution and research, but estimating change or absence usually requires a sampling design that accounts for effort, detectability and timing. A database search is a starting point for inquiry, not a complete national census.^S007

Red Lists are assessments, not headcounts

A Red List addresses a different question from a checklist. It evaluates the risk of extinction for a taxon within a defined region using criteria and evidence. Andorra’s Government describes its flora Red List as an assessment of vascular-plant taxa at national or regional scale using IUCN categories and criteria. The starting list published in 2008 contained 346 taxa; the Government says additions, exclusions and category changes can follow as knowledge is updated. That means categories need to be read with their assessment date and scope, not treated as permanent labels.^S005

The 2021 flora master-list figures distinguish threatened categories—Critically Endangered, Endangered and Vulnerable—from Data Deficient. Data Deficient does not mean that a species is safe, nor does it mean that decline has been demonstrated. It means the available information is insufficient to assess extinction risk confidently. Conversely, a threatened category is an assessment of risk under the chosen criteria; it is not a direct count of remaining individuals or a prediction that extinction is certain.^S006

Andorra’s assessments are updated through a process involving the Group for the Study of Flora and Vegetation (GEFVA). The Government describes GEFVA as a body formed in 2010 to pool knowledge, coordinate research and improve national information on flora and vegetation. Its protocol allows relevant information from professional and amateur botanists to be submitted for potential inclusion in the checklist or reassessment of a Red List category, with public communication of updates. This pathway shows why a checklist is maintained rather than closed once and for all.^S005

Taxonomic coverage also varies across fauna. A 2023 national assessment record for Andorra appears in the National Red List Database under the title Llista vermella de la fauna vertebrada d’Andorra — Validació de l’actualització parcial 2023. The database lists assessment results for several taxa, including insects, showing that its displayed records should be checked by taxonomic group and assessment rather than assumed to be a uniform vertebrate-only table based on the title. This source is useful for tracing dated assessments; it is not a substitute for the underlying full report or a single all-species status summary.^S008

Conservation categories can change for more than one reason. New survey records can expand the known range. A taxonomic revision can split one species into several or combine names. Improved trend data can change the calculated risk. A threat may intensify or recede. A category shift, by itself, therefore does not prove that nature improved or worsened since the previous edition. Readers need the assessment dates, methods and supporting population or range evidence before interpreting a change as a biological trend.^S005^S008

How researchers build a picture of change

Long-term monitoring is designed to compare observations made repeatedly under a consistent method. AR+I’s common-bird programme, SOCA, began in 2011 and uses transects surveyed with help from volunteer ornithologists. Its goals include measuring population trends for common birds, assessing conservation status and contributing Andorran observations to European monitoring. The project page reports 25 active transects in 2025, 72 per cent of them classified as priority routes. It also states that trends for most Andorran species are not yet statistically significant. That candid limitation matters: a monitoring scheme can be active and valuable before it can support a strong trend claim for every species.^S009

The strength of a repeated transect is comparability. Observers return to a defined route and follow a method, so the resulting series can help separate persistent change from an isolated sighting. But the programme monitors common birds and represents only part of the country and its habitats. It cannot be treated as a census of all birds, still less as a measure of every taxonomic group. Rare or elusive species may need targeted surveys, while trends in plants, insects or amphibians call for different sampling methods.^S009

AR+I’s Butterfly Monitoring Scheme, BMSAnd, began in 2004. Its stated purpose is to monitor diversity and population change in diurnal butterflies, which can be sensitive to environmental changes and act as bioindicators. A project page describes transects involving volunteers and natural areas; its earlier map and route count were explicitly dated to 2021. The value is in the time series and consistent methodology, not in treating butterflies as a stand-in for all biodiversity. A change in butterfly observations may prompt questions about climate or land use, but it does not identify a cause without further analysis.^S010

In 2026 AR+I began a project to improve knowledge of Andorra’s wild bees. The April 2026 announcement describes surveys in three periods—early spring, late spring and midsummer—using entomological nets and coloured traps across sectors chosen to cover suitable bee habitats. The initial field day in the south led researchers to anticipate identification of more than thirty species, but that was a preliminary expectation, not the completed national checklist. The project’s design illustrates the sampling problem: many pollinating species are small, seasonal and poorly known, so survey timing and habitat coverage influence what can be recorded.^S011

Amphibians provide another example of an evidence gap being addressed. AR+I launched the Common Amphibian Monitoring of Andorra (SACA) in 2026, adapting a method developed by the Catalan Herpetological Society. The programme is intended to estimate population trends over the medium and long term using citizen participation. An earlier announcement described six known amphibian species in the country and explained that a large data gap motivated monitoring beyond the boundaries of natural parks. The project is new; it should not yet be presented as a trend series with settled results.^S012

At a broader scale, AR+I’s Sentinel Observatory applies long-term monitoring along an altitudinal gradient in the Sorteny Valley. The project draws on standardised approaches to study mountain ecosystem dynamics, with permanent plots separated along the gradient. Elevation provides a way to examine how ecological communities vary with conditions over short distances, while repeated measurements can help track change through time. A single site cannot represent all Andorran habitats, but a carefully selected observatory can reveal processes that a one-off national map would miss.^S013

These programmes are complementary because organisms and habitats respond on different schedules. Butterflies can be counted during suitable flying periods; plants may need surveys at flowering or fruiting stages; amphibians are detected during breeding activity; lake ecology requires water and sediment methods; birds can be monitored on transects. No single calendar or sampling instrument captures all this life. National understanding comes from combining inventories, repeat monitoring, habitat mapping, targeted research and transparent records of where evidence is weak.^S001^S009^S010^S011^S012^S013

Habitats, connectivity and conservation priorities

Habitat classification provides a bridge between species records and land management. The Government’s 2018 assessment of ecosystems of interest describes ecological risk as involving external influences, internal biological processes, historical legacies and spatial context. These factors can interact: a habitat may be exposed to a changing climate, altered by land use and isolated from similar areas, while its species also face competition or other biological pressures. Ecosystem condition is therefore more complex to evaluate than simply counting threatened species within a boundary.^S003

The national strategy called for better knowledge of habitats and species, identification of pressures, and attention to ecological connectivity. It also noted that some areas of high interest for biodiversity and connectivity lacked a formal conservation and management status when the strategy was written. That observation is historical and should not be used as a current inventory of unprotected land. It does, however, explain why conservation planning cannot focus only on designated parks: movement between habitats and the condition of land beyond park boundaries also matter.^S001

Connectivity describes how organisms, water or ecological processes move through a landscape. A continuous river corridor can link habitats from headwaters to the valley floor; a group of wetlands can provide a sequence of breeding or feeding sites; woodland patches can be separated by roads or development. A habitat map can help planners see these relationships, but the actual connection depends on species and process. A route useful to a flying bird may not help an amphibian crossing a road. “Connected” must therefore be specified, not used as a generic synonym for green space.^S001^S002

The 2018 ecosystem assessment uses a Red List of Ecosystems approach to identify ecosystems at risk of collapse. This is not the same as a species Red List. Species assessments focus on extinction risk for a taxon; ecosystem assessments examine the risk that an ecosystem loses its characteristic biota or moves outside its natural variation in composition, structure or function. A habitat can be under pressure even if some constituent species are still present. Equally, an ecosystem category does not reveal the status of every species within it.^S003

Priorities depend on the conservation objective. Protecting a rare plant’s known site may require preventing a specific disturbance. Maintaining a pasture-associated community may need continued grazing at an appropriate intensity. Restoring a wetland may involve water flow and drainage, not simply fencing its perimeter. Maintaining forest structure may require decisions about succession, deadwood or fire risk. A map can flag where attention is warranted; field evidence and site-specific planning determine what action is suitable.^S001^S003

Pressures: land use, climate and introduced species

Andorra’s ENBA identified occupation of valley floors and likely climate-change effects among pressures on biodiversity. It also described agriculture and livestock raising, forests and tourism as sectors that need to account for biodiversity in their policies. The strategy’s own period was 2016–2024, and the Government page confirms approval in 2016. The plan remains useful as a dated account of priorities and intended activities, but its time horizon has passed. No replacement national strategy was verified for this report, so the ENBA is not presented as a current post-2024 action plan.^S001^S014

Land use can affect biodiversity through direct habitat loss, fragmentation or change in management. Development concentrated in valley bottoms can press on river corridors and natural or semi-natural land where transport and settlement are already clustered. Infrastructure can divide habitat or alter drainage. On the other hand, traditional agriculture and grazing can maintain open habitats that would otherwise change through succession. The effect depends on where, how and at what intensity land is used; a national statement that all development or all grazing is harmful would erase important ecological differences.^S001

Climate change affects species and habitats through several pathways. Temperature and precipitation patterns influence snow cover, growing seasons, water availability and the conditions under which species can persist. Mountain organisms may shift their distributions upslope, but a high-elevation specialist can encounter a shrinking area of suitable terrain. Changes in timing can also disrupt ecological relationships, such as flowering and pollinator activity. These are established mechanisms of mountain ecology; the size and timing of effects at a particular Andorran site require local observations or appropriately scaled modelling.^S001^S015

Introduced species are another management concern. The Government defines exotic species as those brought by people, intentionally or accidentally, into a territory outside their native range; an exotic species is considered invasive when it can threaten ecosystems, habitats, native species or biodiversity. Law 7/2019 directs administrations to prevent introduction and spread where scientific evidence shows potential harm. The Government’s current invasive-species page lists examples recorded in Andorra, including red-bellied minnows, Asian hornets, Vietnamese pigs, Senecio inaequidens, buddleia and Himalayan balsam. The list and regulations should be checked in their current version before giving practical control advice.^S004^S016

The red-bellied minnow example is relevant to mountain lakes because the Government says it was introduced as fishing bait and is present in many of the country’s lakes. This is a specific recorded pathway, not evidence that every lake is invaded or that every native species has declined there. Introduced fish can alter aquatic communities, but determining local effects requires site-specific ecological data. The example connects everyday behaviour to a possible landscape-scale pathway: moving organisms between water bodies can create conservation problems that are difficult to reverse.^S004

Invasive-species control requires more than naming a plant or animal. A manager needs to know where it occurs, whether it is spreading, which habitat is affected, what removal method is appropriate and whether control itself could cause harm. The Government lists control plans for several species, including Asian hornet and invasive plants. Such plans show that action is being organised, but their existence alone does not demonstrate that an invasive population has been eradicated or that risk has ended.^S004

Policy and the changing evidence base

Andorra’s National Biodiversity Strategy was approved on 19 October 2016 under the country’s commitments to the Convention on Biological Diversity. Its published horizon ran from 2016 to 2024. The document set five strategic objectives and 24 activities, including improving inventories, filling research gaps, identifying factors affecting biodiversity, managing invasive species, strengthening protected-area planning and integrating biodiversity into sectoral policies. These were policy commitments and planned actions, not a guarantee that every activity was completed by the end date.^S001^S014

The strategy’s age makes the newer monitoring work important but does not, on its own, show that policy implementation has been evaluated. Bird transects, butterfly monitoring, amphibian surveys, bee inventories, flora updates and habitat assessments add to the evidence base in different ways. A full progress audit would need official implementation reports, current biodiversity indicators and a successor strategy or updated national targets. The article’s evidence review did not establish that complete audit, so no claim is made that Andorra has met all of the ENBA’s objectives or that the strategy has been replaced.^S001^S005^S009^S010^S011^S012

Legal protections operate alongside research. Law 7/2019 establishes general duties and instruments, while regulations may govern protected fauna, flora and invasive species. The Government has reported updates to protected-fauna rules based on Red List assessments, technical analysis and review by the relevant biodiversity commission. This illustrates a link between scientific assessment and regulation: a status evaluation can inform which species receive legal protection or management attention. The legal list, however, is not identical to the full scientific checklist or Red List.^S017

Public databases extend access to knowledge but need careful reading. GBIF-Andorra shares records under dataset-specific licences and metadata. The flora page provides checklist and Red List files and explains its update protocol. Monitoring schemes publish dated reports. Together they allow researchers, land managers and interested readers to trace observations and methods. Before reusing a record, the user should inspect its source, date, location precision, identification confidence, licence and sampling context. A dot on a map is evidence of a record, not a ready-made conclusion about conservation status.^S005^S007^S009

Reading the evidence responsibly

Three questions help prevent common mistakes. First, what is the unit: a species, taxon, habitat class, observation, transect count or mapped area? Second, what is the period: one field season, a dated checklist, a multi-year series or a strategy’s planning horizon? Third, what is the geography: a point location, one valley, a park, the whole country or the wider Pyrenees? Results that differ in any of these respects should not be merged into one national trend without analysis.^S001^S005^S007

The same care applies to ecological interpretation. A species appearing in more records might reflect increased observer activity or better data sharing. Fewer observations could result from weather or survey effort rather than population decline. A habitat map showing a class across the country does not tell whether that habitat is in good condition. A threatened category signals assessed risk, not necessarily an immediate local disappearance. Strong conclusions require a method that matches the question and evidence capable of distinguishing the plausible explanations.^S003^S005^S007^S009

For a traveller or resident, the practical lesson is modest but meaningful. Andorra’s nature is not confined to famous parks or dramatic high peaks. Wetlands, pasture, woodland, river edges and managed land can all support important ecological values. Observation should be low-impact: keep to authorised paths where required, avoid disturbing wildlife, do not release bait or plants, and follow current restrictions from the relevant park or authority. These actions do not replace public policy, but they avoid adding preventable pressure to sensitive places.^S004^S018

Andorra has developed a broad base of habitat mapping, species lists, Red List assessments, databases and long-term monitoring. The picture remains more detailed for some taxa and places than others, and the published national strategy’s 2024 horizon makes policy follow-through a live question. The most useful account is therefore neither a single headline species count nor a claim that monitoring has solved the problem. It is a layered view: habitats create the setting, inventories record what is known, assessments evaluate risk, repeated observations measure change, and management decisions respond to that evidence while acknowledging what is still uncertain.