EE34 state of biodiversity Flashcards
CBD?
convention of biodiversity in 2002
LPI?
living planet index
combo of time series (more than 10,000) and more than 2000 vertebrate species
LPI 2010 report-= decline in biodiversity
LPI 2010 report results
Biodiveristy varies amongst regions
temperate = +29%
tropical= -70%
Incomewise
high=little change
medium=decline
low= 1 has become 0.6-0.2
vertebrate pops down by 30% since 1970
Georgina mace
key in developing criteria for listing species in IUCN redlist. Emphasis on actual DATA not previous expert nominations
Russel Lande
worked with georgina mace and was specialist for stochastic populations
what is the IUCN red list?
most comprehensive information source on the global conservation status of animal, fungi and plant species and their links to livelihoods
categories of IUCN red list
Extinct EX Extinct in the wild EW Critically endangered CR Endangered EN vulnerable VU Near threatened NT Least concern LC Data deficient DD Not evaluated NE
does least concern mean the population isn’t at risk of population crash?
NO. ie passenger pigeon demonstrates large populations can still go extinct
what are the 5 criteria?
1) population reduction
2) geographic range (extent of occurence or area of occupancy)
3) small population size AND decline
4) very small/restricted population
5) quantative analysis (PVA)
PVA?
population variability analysis
what has PVA already been done on?
birds,mammals and amphibians
-full analysis to use numbers to put into multiple categories A-D
can be used to plot density of threatened species and see if those with high level of threat are in similar areas (they are not)
what is population reduction based on?
quantitive values (cut off points)
does area of occupancy show true range?
no, often underestimates
what percentage of birds mammals and amphibians are threatened with extinction?
11% birds
25% mammals
33% amphibians
how many species are critically endangered?
3900 ish
what are the pros of the red list?
less data required easy to understand quick practical overprotective rather than under can be converted to probabilities
Practical was meant to show red list= repeatable and robust
what are the cons of the red list?
makes assumptions
not accurate
requires SOME data
species focus- (doesn’t account for large variation between species ie demes)
thresholds- just above/below are put in different categories
RLI?
red list index
if groups have been assessed MULTIPLE TIMES
calculate % species up and % down then express as a RLI (steepness of slope shows rate of change)
this enables you to see overall decline or increase
what does the RLI enable you to do?
see overall decline or increase
what is there a current huge lack of info for?
insects
tropical plants
did we reach 2010 target “to significantly reduce the rate of biodiversity loss by 2010”?
No.
rate of decline is STEADY at best
pressures are worsening
conservation attempts are increasing though
why is a sampled red list useful?
if you pick a species from a represented taxa to get info tyou can produce a helpful representative measure on a species we know little about
how do you produce something for the sample red list?
apply red list criteria to a representative sample of species
sample proportionally to biodiversity
choose species at random