Plant population dynamics research has a long history, and data collection rates have increased through time. The inclusion of this information in databases enables researchers to investigate the drivers of demographic patterns globally and study life history evolution. Studies aiming to generalise demographic patterns rely on data being derived from a representative sample of populations. However, the data are likely to be biased, both in terms of the species and ecoregions investigated and in how the original studies were conducted. Matrix population models (MPMs) are a widely-used tool in plant demography, so an assessment of publications that have used MPMs is a convenient way to assess the distribution of plant demographic knowledge. We assessed bias in this knowledge using data from the COMPADRE Plant Matrix Database, which contains MPMs for almost 800 plant species. We show that tree species and tropical ecoregions are under-represented, while herbaceous perennials and temperate ecoregions are over-represented. In addition, there is a positive association between the number of studies per country and the wealth of the country. Furthermore, we found a strong tendency towards low spatiotemporal replication: More than 50% of the studies were conducted over fewer than 4 years, and only 17% of the studies have replication across >3 sites. This limited spatiotemporal coverage means that the data may not be representative of the environmental conditions experienced by the species. Synthesis: The biases and knowledge gaps we identify are a challenge for the progress of theory and limit the usefulness of current data for determining patterns that would be useful for conservation decisions, such as determining general responses to climate change. We urge researchers to close these knowledge gaps with novel data collection.
|Publication date||26. Apr 2021|
|Number of pages||46|
|Publication status||Published - 26. Apr 2021|