The latter phenomenon can be temporary or permanent, depending on whether the environmental degradation that leads to the loss is reversible through ecological restoration / ecological resilience or effectively permanent (e.g. through land loss). Global extinction has so far been proven to be irreversible.
Even though permanent global species loss is a more dramatic phenomenon than regional changes in species composition, even minor changes from a healthy stable state can have dramatic influence on the food web and the food chain insofar as reductions in only one species can adversely affect the entire chain (coextinction), leading to an overall reduction in biodiversity, possible alternative stable states of an ecosystem notwithstanding. Ecological effects of biodiversity are usually counteracted by its loss. Reduced biodiversity in particular leads to reduced ecosystem services and eventually poses an immediate danger for food security, also for humankind.
Locally bounded loss rates can be measured using species richness and its variation over time. Raw counts may not be as ecologically relevant as relative or absolute abundances. Taking into account the relative frequencies, a considerable number of biodiversity indexes has been developed. Besides richness, evenness and heterogeneity are considered to be the main dimensions along which diversity can be measured.
As with all diversity measures, it is essential to accurately classify the spatial and temporal scope of the observation. "Definitions tend to become less precise as the complexity of the subject increases and the associated spatial and temporal scales widen. Biodiversity itself is not a single concept but can be split up into various scales (e.g. ecosystem diversity vs. habitat diversity or even biodiversity vs. habitat d.) or different subcategories (e.g. phylogenetic diversity, species diversity, genetic diversity, nucleotide diversity). The question of net loss in confined regions is often a matter of debate but longer observation times are generally thought to be beneficial to loss estimates.