Canadian species index
Access PDF (858 kB)
Animal wildlife is highly valued by Canadians and it is one of the most visible and well-studied aspects of biodiversity. The Canadian species index shows whether monitored species tend to have increasing or decreasing population sizes. This, in turn, provides an integrated measure of the condition of our environment.
- Between 1970 and 2014, vertebrate populations have, on average, declined by about 10%.
- Freshwater species trends varied over time and by 2014 showed little net change.
- Terrestrial species have declined on average, and by 2014 were about 10% below the 1970 baseline, mainly due to declines in mammal populations.
- Marine species generally increased in the 1970s and then declined. Trends vary among groups of species. The 2 largest species groups are fish, which showed a downward trend after 1975, and birds, which generally increased. Short-term trends should be interpreted with caution.
Canadian species index, 1970 to 2014 (1970 = 0)
Overall, the trend for monitored vertebrate species in Canada shows a gradual decline. Three sub-indices are also shown. The marine sub-index shows an initial increase and then declines. The freshwater sub-index shows some fluctuation, but ends near baseline values. The terrestrial sub-index shows a general decline.
Data for this chart
|Year||National index||Terrestrial index||Freshwater index||Marine index||Number of species|
Download data file (Excel/CSV; 2.63 kB)
Note: Trends are calculated based on the proportional change in population size for monitored vertebrate species. All species are weighted equally, such that a species that doubled in population would be balanced out by a species that declined by half. Results are plotted on a ratio (logarithmic) scale; 0.06 represents about a 15% increase from the baseline and -0.06 represents about a 13% decline.
Source: Zoological Society of London, 2017.
The freshwater trend is a balance between increases in waterfowl, which have benefitted from initiatives such as the North American Waterfowl Management Plan, and generally declining population trends of other freshwater species.
The sharp decrease in marine species after 2011 occurred when birds, mammals and fish were all declining at the same time. This short-term trend should be interpreted with caution.
The national trend is the average of the trends for all vertebrates; that is, it is the average rate of change across species. Data are not available for all species, and existing data do not always cover the geographic range of each species or the whole time period. The index should be interpreted keeping these limitations in mind.
Bird and mammal species are well represented. Many fish species are included, but the proportion of fish species included is low because in Canada there are more species of fish than there are species of all other vertebrates combined.
Although averaging the trends of many species provides an overall picture, it is important to look for smaller-scale patterns that may have been masked. For example, if we consider taxonomic groups within the marine index, quite different patterns can be seen for birds and fish. By 2014, the marine bird index had increased to an average of 175% of the 1970 levels, while the fish index had declined to around 70% of the 1970 levels.Footnote1
Canadian species index for marine birds and fish, 1970 to 2014 (1970 = 0)
The marine bird index shows a mainly upward trend beginning in 1970. The marine fish index increases until 1975 and then shows a mainly decreasing trend.
Data for this chart
|Year||Marine bird index||Number of bird species||Marine fish index||Number of fish species||Marine mammal index||Number of mammal species|
Download data file (Excel/CSV; 2.35 kB)
Note: Trends are calculated based on the proportional change in population size for monitored vertebrate species in the selected groups. Results are plotted on a ratio (logarithmic) scale.
Source: Zoological Society of London, 2017.
In addition to reflecting environmental trends, indices may reflect changes in data availability. This complicates the interpretation of short-term patterns. For example, a rapid change in the freshwater index beginning in 1992 occurred at a time when data availability increased. Changes in the trajectory may be due to genuine environmental change or may occur because species with diverging patterns have been added to the index. Often, these effects are both present.
The Canadian species index is broadly similar to the Living Planet Index. At the global scale, the Living Planet Index shows a strong decline in vertebrate populations in terrestrial, marine and, especially, freshwater biomes. The Living Planet Index for Canada (2017) uses the same methods as the Canadian Species Index and reports different sub-indices.
About the indicator
About the indicator
What the indicator measures
The Canadian species index represents the average proportional change in the sizes of Canadian vertebrate species' populations. The index is a "trend of trends", rather than a measure of changes in the total number of animals: each species, whether it is common or rare, has the same effect on the index. The index reports general trends rather than progress towards desired levels. The national index contains data for 900 of the nearly 2000 native vertebrate species in Canada.
Why this indicator is important
Animal wildlife populations depend on the availability of suitable habitat. This indicator shows trends in animal populations and is a good proxy measure of overall trends in biodiversity and ecosystem health. This indicator also supports the measurement of progress towards the following 2016–2019 Federal Sustainable Development Strategy long-term goal: All species have healthy and viable populations.
Species at risk population trends follows changes in the population size and distribution of species at risk that are managed under the Species at Risk Act.
Status of wild species in Canada reports extinction risks across a broad set of species.
Trends in Canada's migratory bird populations tracks population size changes in birds that migrate different distances.
Data sources and methods
Data on changes in the size of vertebrate populations are gathered from a variety of sources and collated in the Living Planet Database by the Zoological Society of London. Sources include the peer-reviewed scientific literature, government reports, and reliable online databases. Examples of important sources for Canada include the North American Breeding Bird Survey and the Fisheries and Oceans Canada Library.
Data include counts of individuals, as well as proxy measurements such as indices of abundance, spawning density, or detection rates of individuals. Each record is also tagged with geographical and ecological information to allow for further analysis. Together, these records form the data set used to calculate the index.
At least some information for 900 (50%) of the 1794 regularly occurring native vertebrate speciesFootnote2 found in Canada has been captured in the data set. Birds are the best represented species group.
The index has been calculated for the period 1970 to 2014, the time period where sufficient data exist for credible estimates.
The trend in the population size of each species is estimated using all the information available for that particular species. This may include just one measurement, or a combination of measurements from different populations of the same species. These trends are averaged across all species to generate the Canadian species index.
Data collection and tagging
For use in the index, the time series drawn from Canadian data contained in the Zoological Society of London database must meet all of the following criteria:
- contain data for at least 2 points in time since 1970
- have been collected for a defined population using comparable methods across years
- use units of population size or a reliable proxy, such as spawning biomass or density
- have a referenced and traceable source
Each time series is referred to as a "population" although they may not be populations in the biological sense of the term.
Population data have been gathered from the literature and by performing online searches and by contacting experts. Birds have been monitored at the national level for decades and high-quality data are readily available for this species group. Fewer data are available for other species groups. To help address the imbalance in the data available for the different species groups, targeted searches were carried out for under-represented groups. Searches were also conducted to locate data for under-represented regions.
Each record is tagged with contextual information such as geographical region, species group, and habitat type, for example. Data tags allow a subset of the database to be extracted for targeted analysis. Information for these tags is drawn from the original data source if possible; however, additional reference material is also used. Species that occur in more than one ecosystem type (terrestrial, freshwater or marine) are tagged as belonging to the ecosystem in which they were observed and on which they rely for at least part of their life cycle. For example, a time series containing counts of salmon spawning in rivers would be considered freshwater, while one containing observations at sea would be considered marine.
Data for the overall index were restricted to vertebrate species that regularly occur in Canada. Classification was based on Wild Species 2005 and 2010 reports. Species assessed as "invasive" or "accidental" were excluded.
Increasing population sizes are generally interpreted as a sign of environmental improvement. However, a few bird species are known to have a population size that is above acceptable bounds (see Population status of Canada's migratory birds), and for these species, an increase in population is a negative outcome. Three species, Snow Goose (both subspecies), Ross's Goose and Canada Goose,Footnote3 have been excluded from the index for this reason.
In a few cases (6 time series, 3 species),Footnote4 data could not be clearly assigned to a particular species because of changes in taxonomy, and therefore these time series were excluded.
For each population, a record of abundance over years is created. Modelling is used to reduce the effect of random variations and measurement noise. For time series containing at least 6 data points, trends were modelled using Generalised Additive Modelling. For shorter time series, and for any series that could not be modelled with Generalised Additive Modelling, a linear regression model was used. For time series with only 2 data values, this is equivalent to a straight line connecting the 2 points. Time series are not extrapolated beyond the start and end date of observations.
In some years and for some time series, a zero value has been recorded. In a few cases, this may be due to a local extinction, but more often, it is because animals are not observed. A failure to observe animals may be because there are few animals to observe, which is a genuine signal of low numbers. It could also mean the animals were simply not detected. This might happen, for example, if unusual weather conditions made movement patterns unpredictable, in which case a zero would represent a missing value. For the purposes of the indicator, zeros have been treated as missing values, resulting in a conservative estimate of change.
Calculation of the index
Trends within a time series
For each time series, proportional change dt is calculated for each year for which data exist, as follows:
dt = log10(Nt /N(t-1))
Nt = modelled population size estimate in year t
N(t-1) = modelled population size estimate in year t-1
- For species with more than one time series, the average proportional change (lambda, λ) is calculated for each year across all time series (including all subspecies) for that species. Formally, for species i in year t:
λi,t = average proportional change for species i in year t
di,j,t = proportional change for time series j, for species i in year t
m = number of time series for species i in year t
For a species with only one time series,
- The overall annual change is calculated as the average lambda across all species with data for that time step. In other words, the index for 2014 is the average for all species with population estimates in 2013 and 2014. Species are weighted equally, regardless of data availability.
- The index for a particular year is the sum of annual changes since 1970.
Percentage changes are calculated by taking the antilog of the index.
Sub-indices are calculated using the same methodology, but for a selected subset of species.
Assessment of uncertainty
The degree of variability within the species-level lambdas for a given year provides an indication of whether trends are similar across the species included in the index. A narrow interval means that most species are changing by similar proportions, while a wide interval means that there is a wide range of patterns. Because indexed species are not a random or representative selection of the species in the environment, this can only be a partial assessment of uncertainty. The uncertainty due to a non-representative sample of species cannot be measured.
Distribution of species-level lambda values, 1970 to 2014
The dots show the average annual lambda across all species; vertical bars show the standard deviation of average annual lambda across all species.
Four charts (national top left, terrestrial top right, freshwater bottom left and marine bottom right) show the average and standard deviation of the annual lambda across all species. For the national index, the average annual lambda fluctuates around zero over time, and is generally slightly negative. The standard deviation is roughly the same for each year. For the terrestrial species index, the average annual lambda fluctuates over time, and is generally slightly negative. The standard deviation is similar across years. For the freshwater species index, the average annual lambda fluctuates over time; more years are slightly negative until around 2000, when values are more often positive. The standard deviation is higher for the second half of this time series. For the marine species index, the average annual lambda is generally positive until 1975, after which it is mainly negative until the early 1990s. Values are mainly slightly positive for about a decade, and then are mainly negative, becoming strongly negative in the last 3 years of the series. The standard deviation is generally higher than for other ecosystem types.
Data for this chart
|Year||National index, average lambda||National index, standard deviation||National index, number of species||Terrestrial index, average lambda||Terrestrial index, standard deviation||Terrestrial index, number of species|
|Year||Freshwater index, average lambda||Freshwater index, standard deviation||Freshwater index, number of species||Marine index, average lambda||Marine index, standard deviation||Marine index, number of species|
Note: n/a = not applicable.
Download data file (Excel/CSV; 5.26 kB)
Source: Zoological Society of London, 2017.
Caveats and limitations
The Canadian species index has been developed from the Living Planet Index, originally conceived by the World Wildlife Fund and now developed in partnership with the Zoological Society of London. The index is based on a peer-reviewed methodFootnote5 that can integrate many types of population measurements. The methodology for the Canadian species index has been improved and revised, and results should not be directly compared to the Living Planet Index.
The index uses previously collected data, and is therefore biased towards certain species. These include species that are easy to observe, species that are managed for human use or for conservation and species with aesthetic appeal. Birds are well represented, but most other vertebrate groups are not. Some species are represented by data that come from a local study involving a small part of the total population. While there is considerable uncertainty in the trends for these species, combining data for many species leads to more interpretable results.
The index is descriptive. Because the underlying data have been collected for other purposes, the set of species contained in the index has unknown sampling biases. For this reason, it does not meet the requirement for randomized sampling that is necessary for traditional statistical hypothesis testing, and changes in the index cannot be said to be statistically significant. Trends in the index do provide an indication of trends in the environment, and can be used to identify where additional analysis or information is required.
Averaging trends across all populations within each species can obscure important variability between subspecies, varieties, or geographic regions. Averaging trends across species may also obscure important information. Analysis of different parts of the data set can help uncover these patterns.
Population size measurements always include some uncertainty, because not every individual animal can be found and counted at every sampling interval. The effect of uncertainty in measurement cannot be separated from genuine changes in population sizes. Random variability may lead to a few more or less individuals being counted. If this variability leads to a large proportional change, as is the case when the average number of individuals found is small, the resulting uncertainty in the index can be large. However, measurement uncertainty does average out over longer time series and over species. For this reason, interpretation of small subsets of data must be done with an understanding of the context of the biology of the species that are included and the strengths and weaknesses of the monitoring protocols.
Only vertebrate species are included in the index, because they are the only group with sufficient population-level data. Invertebrates and plants tend to be monitored using area of occurrence, a type of data not readily integrated into the index.
Canadian Endangered Species Conservation Council (2006) Wild Species 2005: The General Status of Species in Canada. National General Status Working Group. Retrieved in August 2017.
Canadian Endangered Species Conservation Council (2011) Wild Species 2010: The General Status of Species in Canada. National General Status Working Group. Retrieved in August 2017.
Canadian Endangered Species Conservation Council (2017) Wild Species 2015: The General Status of Species in Canada. National General Status Working Group. Retrieved in August 2017.
Collen B et al. (2009) Monitoring Change in Vertebrate Abundance: the Living Planet Index. Conservation Biology 23(2): 317-327.
Report a problem or mistake on this page
- Date modified: