Conservation biology asks how biological knowledge can help maintain and recover living diversity. It connects evidence about populations, habitats, genetic variation and ecological relationships with interventions whose effects must be tested. The gold standard is a credible account of what is declining, why it is declining, what can be changed and how we will know whether an action helped.
Imagine a wetland where a familiar bird becomes harder to find. Fewer sightings could mean fewer birds, changed movement, different survey conditions or several processes together. Planting vegetation may be a useful response, but only if the relevant habitat relationship and other pressures are understood. Concern begins the investigation; it does not supply the diagnosis.
The wider Conservation root includes natural and cultural conservation and the responsibilities of caring for valued places and objects. This subject root focuses on the biological science of persistence and recovery. Ecology, evolution, genetics, systematics and biogeography remain the specialist foundations. Their connections make conservation biology a practical science, rather than a list of appealing projects.
All numerical cases in this guide are invented educational examples. They are not assessments of a named species or instructions for releasing, moving or treating wildlife. The educational purpose follows The Gold Standard of Education: learn a mechanism, inspect the evidence and use the idea carefully when the context changes.
Find your next question
Begin with the route that fits your purpose. Read sequentially for the full argument, or use the chapter index to return to a worked example, an evidence check or a connected guide.
Full chapter index · Conservation Biology: mechanism library · How X Works
Chapter index
Define diversity and persistence · Chapters 1–4
Follow populations and habitats · Chapters 5–8
Diagnose threats and evidence · Chapters 9–12
Evaluate actions and decisions · Chapters 13–15
Learn, check and continue · Chapters 16–20
CHAPTER 1 OF 20 · Define diversity and persistence
1. Define what persistence would mean in this case
Conserving biodiversity can involve genetic diversity, species, populations, ecological communities and processes. A project needs a clear target because success at one level does not automatically establish success at every other level. More individuals of one species can coexist with losses elsewhere in the community.
Begin by describing the entity, region and period. A local breeding population, a species across its global range and a functioning wetland are different targets. The same count can have different meanings for each. A bird observed once at a site is not evidence that a breeding population has become established there.
Persistence also includes the conditions that allow continuation. An adult population may remain visible for years while recruitment has failed. A planted community may appear healthy while depending on indefinite replacement. Short-term presence is useful evidence, but the project should identify the processes needed for long-term maintenance.
For a fictional pond project, a target might be successful breeding by a native amphibian over several suitable seasons, with the habitat conditions required for its life cycle. That is more testable than “make the pond more natural.” It still requires locally appropriate expertise and permissions before any intervention.
Ask what observation would change the team’s conclusion. If every possible outcome is described as success, the target is too vague. A useful conservation objective gives the investigation somewhere to go when results disappoint. It supports both an encouraging finding and an honest revision of the plan, without reducing living systems to a single attractive photograph.
CHAPTER 2 OF 20 · Define diversity and persistence
2. Biodiversity cannot be represented by one number alone
Species richness counts the species recorded within a defined scope. Abundance concerns how many individuals or another relevant measure are present. Community composition asks which species occur. Genetic diversity concerns variation within and among populations. These quantities connect, but they answer different questions.
Imagine two fictional plots containing ten plant species each. In one, a single species accounts for most individuals; in the other, abundance is distributed more evenly. Their richness is identical, but their composition and relative abundance differ. An equal species count does not establish equivalent ecological condition.
Now replace several native specialists with widespread generalists while maintaining the same total richness. The unchanged number can conceal a significant biological change. Interpreting its importance requires the conservation target and the organisms’ roles and status, not a rule that a larger number is always better.
Sampling scope matters. A larger surveyed area or more visits can detect more species even if the underlying community has not changed. Compare like with like or use an analysis that addresses the difference. Counting extra observations is not automatically evidence of recovery.
For learning, ask students to create two small datasets with the same richness and different abundance patterns. Then ask which additional information is needed to judge a restoration claim. They should request identity, sampling method, target and time, rather than reach immediately for a preferred index. The gold standard uses a measure because it serves the biological question, and preserves the aspects of diversity that the chosen summary leaves out.
CHAPTER 3 OF 20 · Define diversity and persistence
3. Population change is a budget of individuals
For a defined population and interval, the number at the end depends on the starting number, births, deaths, immigration and emigration. The relationship is simple enough to teach early and powerful enough to reveal why a local count alone does not identify the mechanism of change.
In an invented population, 100 individuals are present initially. Twenty births occur, fifteen individuals die, ten arrive and twenty leave. The final population is 100 + 20 − 15 + 10 − 20 = 95. Reproduction occurred, yet the local population declined. The balance of all terms matters.
If the same site receives enough immigrants, its abundance can remain stable despite local reproduction being insufficient to replace losses. Conversely, a productive site can export individuals. Ecological source and sink concepts help investigate these relationships, but their identification requires demographic evidence rather than a favourable impression of habitat.
Age structure adds another layer. Two populations of 100 can have different prospects if one includes many individuals likely to breed and the other consists mainly of old adults with little recruitment. A single total compresses those differences.
For a changed-context check, remove immigration from the example and calculate again. Then ask what observation could distinguish low reproduction from high emigration. The answer should involve different monitoring methods or data, not more confidence in the same count. Conservation biology becomes useful when the arithmetic leads to a better biological question about which process needs attention.
CHAPTER 4 OF 20 · Define diversity and persistence
4. Small populations face variation that averages can hide
Births and deaths are events affecting individuals. In a small population, chance differences in those events can substantially change the total. Environmental variation can also affect many individuals together. A model based only on an average annual growth rate can conceal those risks.
Suppose a fictional species produces, on average, enough recruits to replace adults under a simplified model. That average does not mean every small population will replace itself every year. Several unlucky outcomes can leave too few breeders, while a drought can reduce reproduction across many nests simultaneously.
Distinguish demographic variation from environmental variation. The former concerns the realised individual events under given conditions; the latter changes the conditions experienced by the population. Their effects depend on life history, population size and relationships among sites.
A deterministic classroom projection of 100 individuals growing by 5% each year gives 105 after one year and 110.25 after two. The fractional result is an expected model quantity, not a claim that a quarter of an individual exists. Real outcomes are discrete and variable. The useful interpretation is the direction and scale implied by the assumptions.
Population viability analysis extends this reasoning through explicit models and scenarios, but its outputs depend on inputs, structure and uncertainty. A precise extinction probability from an under-supported model is not automatically a strong result. The gold standard asks which vital rates, threats and correlations were included and whether the conclusions remain useful when uncertain values are varied.
CHAPTER 5 OF 20 · Follow populations and habitats
5. Genetic variation is connected with population history
Genetic diversity can influence the capacity of populations to respond to changing conditions, while genetic drift, inbreeding and selection shape variation. The number of individuals counted in the field is not always the number that effectively contributes genetic material to later generations.
An effective population size summarises particular genetic processes under a model. It can differ substantially from census size because of unequal breeding success, sex ratios, population fluctuations and other features. Treating the two quantities as interchangeable can distort a conservation argument.
For a deliberately simple illustration, imagine a census of 100 adults but only a small subset produces most offspring. The next generation’s ancestry is concentrated more narrowly than the census alone suggests. The example establishes why breeding structure matters; it does not provide a universal conversion from 100 counted animals to a particular effective size.
Genetic rescue or movement among populations can sometimes be considered, but mixing is not automatically beneficial. Disease, local adaptation, ecological compatibility and the history of the populations require specialist assessment. A school explanation of diversity is not authority to move organisms between habitats.
Connect this chapter with the Genetics and Evolution roots. Students should learn to distinguish evidence of variation from an assumption that every difference improves fitness. Ask which population, genetic marker and timescale a claim concerns. Then ask which demographic and ecological observations would help interpret it. Conservation decisions improve when genetic evidence is combined with the living context instead of being treated as an isolated molecular verdict.
CHAPTER 6 OF 20 · Follow populations and habitats
6. Habitat quality depends on the organism and its life cycle
A habitat supplies conditions and resources for particular organisms. “Green space” is a useful broad description, but it does not tell us whether a species can feed, breed, shelter and disperse there. Different life stages can require different conditions.
A fictional butterfly might use one plant resource as an adult and a different host plant as a caterpillar. Increasing flowers alone may attract adults without supporting recruitment. The incomplete intervention is not repaired by adding more of the same adult resource while ignoring the missing life stage.
Similarly, a pond can provide water yet lack suitable surrounding terrestrial habitat for an organism that spends part of its life outside it. A patch can look visually untidy and still contain important structure. A manicured appearance is not a biological quality score.
Ask how habitat quality is being inferred. Occupancy, breeding success, survival and resource availability provide different evidence. Animals can occur in a site that does not support long-term persistence, particularly if they are attracted to cues that no longer reliably indicate good conditions.
For learning, give students a fictional life-cycle description and three habitat plans. Ask which requirement each plan supplies and which remains unaddressed. Then change the focal species. A feature that helps one organism can hinder another. The answer should become more specific, rather than a competition to name the most universally natural-looking design. Professional planning must integrate the relevant species, community, landscape and human context before choosing interventions.
CHAPTER 7 OF 20 · Follow populations and habitats
7. Fragmentation changes area, edges and movement
Habitat loss reduces the amount available. Fragmentation changes its arrangement. The two often occur together, but separating them helps identify mechanisms. A landscape with the same total habitat area can have different patch sizes, edge exposure and connections.
For an invented geometry exercise, a single square habitat has area 16 square kilometres and perimeter 16 kilometres. Divide the same total area into sixteen separate one-square-kilometre squares and their combined perimeter becomes 64 kilometres. This simple shape comparison does not calculate ecological harm, but it shows why equal area can create very different amounts of edge.
Edge conditions can differ from interior conditions through light, temperature, wind, disturbance or interactions among organisms. Their significance depends on the species and setting. More perimeter is not automatically bad for every organism, and the geometric exercise should not be turned into a universal biological coefficient.
Movement depends on the landscape between patches, not only on straight-line distance. A road, crop field, stream or built area can be more or less permeable to different organisms. A corridor should therefore be evaluated through actual movement and demographic outcomes, rather than approved because it resembles a line joining green shapes on a map.
Ask students to propose two landscapes with equal habitat area and explain a testable difference for a specified species. A strong answer identifies the mechanism, such as access to breeding sites, and the observation required to test it. That makes fragmentation a reasoning problem rather than a slogan about smaller patches.
CHAPTER 8 OF 20 · Follow populations and habitats
8. Metapopulations connect local loss with landscape persistence
Some species occupy networks of habitat patches where local populations can disappear and patches can be recolonised. Persistence then depends on both local conditions and movement among patches. A local extinction is serious evidence, but its meaning for the wider network depends on the system.
The connected metapopulation guide develops this mechanism. It helps explain why protecting one apparently successful patch may be insufficient if the network supplying colonists collapses, or why a temporarily empty patch can remain important future habitat.
In a toy occupancy budget, ten patches are occupied at the start. Two occupied patches lose their local populations and three previously empty patches are colonised. Occupancy rises to eleven, provided enough suitable empty patches existed. The arithmetic says nothing yet about the numbers of individuals or the quality of each patch.
Now introduce a shared disturbance affecting all patches simultaneously. Treating local losses as independent would underestimate the possibility of a network-wide failure. Spatial correlation matters, as do dispersal distances and the condition of the intervening landscape.
For a transfer check, ask whether the same model suits a long-lived tree population with very slow colonisation or a species whose individuals routinely use several patches within a day. The framework may require adaptation or replacement. The gold standard preserves the biological scale of the terms. A convenient patch model should not override the organism’s life history simply because it produces a neat diagram and an easily calculated answer.
CHAPTER 9 OF 20 · Diagnose threats and evidence
9. Diagnose threats through mechanisms, rather than labels alone
Habitat change, overexploitation, invasive species, pollution and climate change can influence biodiversity through different pathways. Listing a pressure is a beginning. A useful diagnosis explains how it affects survival, reproduction, movement or another process relevant to the target.
The IUCN distinction between threats and biological stresses is helpful. A threatening process can produce a degraded condition such as reduced reproductive success or habitat fragmentation. Recording only the symptom can make an intervention too indirect; recording only the broad pressure can leave the action unspecified.
Imagine a fictional breeding colony with poor recruitment. Possible explanations include food shortage, disturbance during breeding, altered habitat, contaminants or inadequate detection of young. Each predicts different evidence. More nest boxes will not solve every possibility simply because the colony breeds in cavities.
Threats can interact. Habitat changes may affect access to food while a new disturbance increases the cost of reaching it. Conversely, an apparent association can arise because both a threat indicator and a population trend respond to another factor. The mechanism needs testing, not just a plausible story.
A practical reasoning table can record the proposed threat, affected process, expected observation, alternative explanation and feasible evidence. This keeps the diagnosis open to correction. It also makes limited resources easier to use: the next survey or analysis should distinguish possibilities that would lead to different actions. Collecting additional data without a decision question can consume effort while leaving the central uncertainty unchanged.
CHAPTER 10 OF 20 · Diagnose threats and evidence
10. Observation is filtered through detection
An organism can be present without being detected. Detection depends on behaviour, season, observer, equipment and survey conditions. A decline in observations can reflect abundance, detectability or both. Conservation monitoring therefore needs methods suited to the organisms and the question.
In an invented example, a site contains 100 individuals. A survey detects about half under one set of conditions, producing roughly 50 detections. Under another set, the same population has an approximate detection fraction of 0.3 and produces around 30. Treating that difference as a measured 40% population decline would confuse the observation process with the population.
The simple multiplication is an expectation for teaching, not a full estimator. Real monitoring can require repeated visits, individual identification, sampling design or model-based approaches that separate occurrence and detection under explicit assumptions. Merely dividing by a guessed detection fraction does not solve the problem.
Survey effort must be documented. More camera nights or a different time of day can change detections. Automated recording and identification tools add coverage but also need validation, including false detections and missed organisms. A large volume of machine-labelled records is not automatically an accurate population census.
For practice, give learners identical survey counts collected with different efforts. Ask what comparison can safely be made and what information is missing. The best answer may be that a population trend cannot yet be estimated from those records. That is a useful scientific conclusion when it identifies a better monitoring design.
CHAPTER 11 OF 20 · Diagnose threats and evidence
11. Extinction-risk categories are assessments with defined scope
The IUCN Red List uses categories and criteria to assess extinction risk. Vulnerable, Endangered and Critically Endangered are collectively described as threatened categories. The criteria consider evidence such as population change, geographic range and population size under specified conditions. An assessment is more structured than a judgement based on whether a species seems rare locally.
Global and regional assessments have different scopes. A species can be common in one locality while facing risk across its wider range, or scarce locally without being globally threatened. Check the assessment date, geographic scope, taxonomic identity and supporting rationale before summarising its status.
Data Deficient does not mean safe. It indicates that available information is insufficient for the relevant assessment. Conversely, an absence of a threatened classification does not establish that no local conservation problem exists. The category must be used for the question it was designed to inform.
Do not assign a category from one toy population calculation in this article. Applying the actual criteria requires the current official guidance, definitions and evidence. The educational task here is to understand why scope, trend, uncertainty and identity matter.
Ask a student to compare two fictional summaries: one states a local count with no date; the other gives a dated assessment across a defined range and explains the criterion. The second is more interpretable, even if its conclusion is less dramatic. The gold standard preserves that supporting structure when communicating risk to a reader who will not inspect every technical document.
CHAPTER 12 OF 20 · Diagnose threats and evidence
12. Interventions need a causal question and a comparison
An intervention is intended to change an outcome. Observing improvement afterwards does not, by itself, show how much improvement the action caused. Weather, regional population change, survey effort and other influences may also have changed. Evaluation asks what would plausibly have happened without the intervention.
For an invented project, a managed site increases from 40 to 60 recorded breeding pairs, while a comparison site increases from 30 to 45 over the same period. The changes are +20 and +15. A simple difference-in-differences calculation gives +5, rather than attributing all 20 additional pairs to management.
That calculation is not automatically a valid causal estimate. The comparison must be appropriate, the observation methods consistent and the assumptions defensible. If the sites were already following different trajectories, the simple interpretation may fail. Several baseline observations and replicated sites can strengthen the design.
The Conservation Evidence project at the University of Cambridge summarises studies of conservation actions. Its role is to help readers examine what has been tested, with which outcomes and in which settings. An action supported elsewhere still requires attention to local conditions and implementation.
For a learning exercise, change the comparison site’s improvement from 15 to 25. The same managed-site increase now corresponds to a relative change of −5 under the toy comparison. The managed site improved in absolute terms, yet did less well than the proposed counterfactual. This distinction is central to evaluating projects honestly without treating every favourable before-and-after photograph as proof of effectiveness.
CHAPTER 13 OF 20 · Evaluate actions and decisions
13. Restoration must rebuild processes as well as appearances
Restoration can involve recovering habitat conditions and ecological processes, but its goals need definition. Plant survival, species composition, hydrology, recruitment and ecological interactions can provide different evidence of progress. A visually green site is not automatically a restored ecosystem.
For a fictional planting project, 1,000 seedlings are installed and 700 survive the first year. That is 70% first-year survival among the planted seedlings, assuming the count is complete and the cohort is defined consistently. It does not establish 70% ecosystem recovery or demonstrate that the surviving plants will reproduce.
The next questions concern causes and continuation. Are the plants appropriate to the site? Do water conditions support them? Are naturally recruiting individuals present? Are important habitat features returning? Does management remove a pressure or merely compensate for it temporarily?
Reference conditions help establish a target, but historical conditions may not be fully recoverable, and future conditions can differ. A defensible plan states what it is trying to recover and which constraints require adaptation. This makes change in the plan explainable rather than an embarrassment to conceal.
Students can compare a project reporting the number of seedlings planted with one reporting survival, recruitment and habitat outcomes over several periods. Neither activity nor a single outcome supplies the whole story. The stronger report connects the action to the process needed for persistence and explains what remains uncertain. This is how conservation biology turns enthusiasm into a learning cycle.
CHAPTER 14 OF 20 · Evaluate actions and decisions
14. Decisions combine evidence, feasibility and human responsibilities
Scientific evidence can clarify likely consequences, uncertainties and mechanisms. Choosing among actions also involves values, rights, feasibility, costs and responsibilities. A biological model does not, by itself, decide whose land use or livelihood should change.
Projects need the knowledge and participation of people who know and use the landscape, alongside relevant scientific and regulatory expertise. Local and Indigenous knowledge can contribute observations and understanding that a short external survey misses. Its use requires respectful relationships and appropriate consent and recognition.
Trade-offs should be described concretely. An intervention favouring one species may alter conditions for another. Restricting access can affect livelihoods or cultural practices. Delaying action can allow losses to continue, while acting under uncertainty can introduce other risks. The useful question is which option best addresses the defined objectives under the actual constraints.
In a fictional comparison, one action has stronger evidence of benefit but cannot be maintained, while another has more modest evidence and a feasible long-term management arrangement. The decision needs both scientific and implementation reasoning. Choosing the largest predicted benefit without checking delivery can produce a plan that exists only on paper.
The broader Conservation root develops the responsibilities of care and significance across natural and cultural settings. This biological root contributes testable mechanisms and outcome evidence to that wider decision. Readers should leave with a clearer division of labour: evidence informs judgement, and accountable judgement must still explain the objectives, people and consequences involved.
CHAPTER 15 OF 20 · Evaluate actions and decisions
15. Build an adaptive monitoring plan before the first action
A useful plan identifies the target, starting condition, proposed mechanism, action, expected outcome and monitoring method. It also states when the evidence will be reviewed and which findings could justify changing course. Monitoring is part of the intervention, rather than a decorative report written after the budget is spent.
Choose indicators that connect with the objective. If the goal concerns successful recruitment, adult sightings alone may not suffice. If the action changes a threat, monitor both that threat and the biological response where feasible. A lag between the two can be biologically reasonable and should be anticipated.
Set a review period suited to the organism and process. A tree community and a seasonal insect population need different timescales. Requiring a rapid visible outcome can encourage measurement of an easy activity while overlooking the slower process that actually matters.
Adaptive management does not mean changing everything whenever a result is disappointing. Record the reason for each change so that the project remains interpretable. If methods and actions change simultaneously without documentation, later outcomes become difficult to learn from.
For a student exercise, create a one-page plan for an invented habitat using supplied observations. Require one measurable objective, one alternative explanation for the trend, one appropriate comparison and one decision rule for review. The exercise should remain a reasoning task; actual wildlife work requires the relevant permissions, safeguards and specialist supervision. The gold standard builds the habit of learning from action without pretending that a classroom plan supplies all the authority to act.
CHAPTER 16 OF 20 · Learn, check and continue
16. Learn through ecological questions and increasingly independent checks
Young learners can start with observation and respectful recording: distinguish organisms, note habitat features and compare conditions without disturbing wildlife. The teaching goal is to ask a clearer question, rather than collect the largest number of specimens or produce an impressive species list.
At secondary level, connect population budgets, ratios, graphs, sampling and causal reasoning. Compare an observation count with an estimated population. Ask what a percentage uses as its denominator. Change the survey effort or the comparison site’s outcome and require a revised conclusion.
Advanced learning adds ecology, evolutionary biology, genetics, statistics, geography, economics and policy. Professional conservation biology also requires field methods, data management, ethical practice and collaboration. A broad interest in wildlife is valuable, but the daily work includes careful records, imperfect data and decisions that may remain contested.
Teach from an example to a guided task and then to a changed context. A learner who can evaluate a bird-nest intervention should next consider a plant-restoration case and identify which reasoning transfers. Population persistence, a counterfactual and detection matter in both, while the organism’s life history and the measured outcome differ.
Progress becomes visible when a student separates a target from an activity, questions detection and proposes evidence that could disconfirm a preferred explanation. An error log can distinguish biological misunderstandings from sampling, arithmetic, scope and causal mistakes. That gives revision a purpose and makes conservation science more demanding and more hopeful than repeating general statements about protecting nature.
CHAPTER 17 OF 20 · Learn, check and continue
17. Independent checks with explained answers
| Question | Useful evidence | Interpretation check |
|---|---|---|
| Is abundance changing? | Consistent surveys and detection analysis. | Separate population from observation process. |
| Is habitat supporting persistence? | Relevant survival and recruitment observations. | Occurrence alone may be incomplete. |
| Did the action help? | A defensible comparison and outcome. | Before-and-after change can have other causes. |
| Is the species recovering? | Evidence at the stated geographic and temporal scope. | Local improvement is not automatically global recovery. |
Check one: a population begins at 80, with 15 births, 10 deaths, 5 immigrants and 12 emigrants. It ends at 78. The decline is two individuals even though births exceed deaths, because net movement is outward. A correct answer accounts for every term over the same interval.
Check two: a survey records 24 organisms with an approximate detection fraction of 0.4 in a toy model. Its expected-count relationship corresponds to 60 present. This is a model exercise, not permission to estimate a real population by dividing by an unverified detection guess.
Check three: a managed site’s outcome improves by 12 units and an appropriate comparison’s improves by 8. The simple relative improvement is 4 units. Causal interpretation still depends on the comparison and assumptions; the arithmetic does not validate the design.
Check four: a habitat project records the same richness before and after, but several specialist species have been replaced by widespread species. The unchanged richness does not establish unchanged conservation condition. Composition and the target must be examined.
Check five: a species has a local increase in one well-monitored patch. This is encouraging local evidence, not proof of recovery across the entire species range. A strong report states what improved and what wider monitoring is needed. Each check preserves the link among numbers, biological meaning and the scope of the conclusion.
Is conservation biology the same as caring about nature?
Concern supplies motivation. Conservation biology adds testable explanations, measurements and evaluation of outcomes. It works with ecology and evolution to investigate which conditions support persistence and whether particular actions help.
Is a protected area automatically effective?
Legal protection is an important action in many settings, but effectiveness depends on threats, implementation, ecological connections and the outcome being measured. A boundary on a map does not by itself establish population recovery or functioning habitat.
Does a species count tell us whether a habitat is healthy?
It provides one piece of evidence under a defined sampling method. Identity, abundance, recruitment, interactions and the conservation objective can change the interpretation. A count should be explained with its scope rather than used as a universal condition score.
Why can a successful project still have uncertainty?
Living systems vary, monitoring is incomplete and the no-action outcome cannot always be observed directly. A strong project narrows the important uncertainties and reports them clearly. It does not need to claim complete certainty to support useful action.
What is the most useful everyday lesson?
Separate intention, activity and outcome. Ask what process an action changes and which comparison shows its effect. This habit is useful whenever a worthwhile goal is being pursued with limited resources and evidence that needs careful interpretation.
Choose a route by the question you want to answer. These connected guides develop the specialist mechanisms and adjacent subjects; they are companions to this explanation rather than substitutes for checking a particular real-world case.
Build the biological foundations
How Science Works | Conservation Biology — Biodiversity, Extinction Risk, Fragmentation, Recovery and Evidence for Persistence — Explore the existing full subject library for persistence, extinction risk and recovery.
How Science Works | Ecology — Populations, Communities, Ecosystems and Changing Environments — Understand populations, communities and ecological interactions.
How Science Works | Evolutionary Biology — Variation, Selection, Drift, Adaptation and Common Descent — Connect selection, adaptation and evolutionary history with changing conditions.
How Science Works | Genetics — Inheritance, Variation, Genes, Genomes and Causal Evidence — Examine genetic variation, inheritance and population processes.
Follow identity and landscape connections
How Science Works | Systematics & Taxonomy — Naming Life, Species Concepts, Phylogenies and the Evidence for Biological Relationships — Keep taxonomic identity and evolutionary relationships clear when comparing records.
How Science Works | How Habitat Patches Stay Connected Through Local Populations — Explore local extinction, recolonisation and connected habitat patches.
Continue across Gold Standard subjects
The Gold Standard of Conservation provides the connected conservation perspective.
The Gold Standard of Ecology provides the connected ecology perspective.
The Gold Standard of Evolution provides the connected evolution perspective.
The Gold Standard of Genetics provides the connected genetics perspective.
The Gold Standard of Biogeography provides the connected biogeography perspective.
The Gold Standard of Systematics and Taxonomy provides the connected systematics and taxonomy perspective.
The Gold Standard of Wildlife provides the connected wildlife perspective.
The Gold Standard of Environmental Science provides the connected environmental science perspective.
For the wider collection, return to How X Works and How Science Works.
These primary institutional and research sources were checked on 21 October 2026. They support the scientific framing and specialist references; the numerical cases, teaching sequence and interpretation checklists are original educational examples.
IUCN Red List: categories and criteria — The official risk-assessment framework; apply current definitions and guidance to real assessments.
IUCN Red List: supporting information — The documented evidence and geographic scope that support a classification.
IUCN Red List: threats and stresses — Distinguish a threatening process from the biological condition it produces.
University of Cambridge: Conservation Evidence — The project’s method of summarising tested conservation actions.
Conservation Evidence: evaluating counterfactuals — Why before-and-after improvement alone does not establish an action’s causal effect.
IPBES: Global Assessment — The wider assessment context for biodiversity and ecosystem relationships.
Return to the source appropriate to the specific measurement, method and context before using a classroom model to make a professional decision. Where the article states a conditional example, its conclusion depends on the assumptions explicitly given with it.
Contents · Previous chapter · Continue into the subject mechanisms
Continue into the subject guides
Compare vertebrate evidence: amphibian and reptile monitoring clarifies imperfect detection, while primatology connects behavioural observation with conservation survey assumptions.
The Gold Standard of Herpetology · The Gold Standard of Primatology