Biological Conservation

Description

Biology Flashcards on Biological Conservation, created by Tara McShane on 28/02/2017.
Tara McShane
Flashcards by Tara McShane, updated more than 1 year ago
Tara McShane
Created by Tara McShane almost 8 years ago
6
0

Resource summary

Question Answer
Lecture 1 Biodiversity, conventions, targets and data
Biodiversity is in decline worldwide The living plant index holds over 10,000 population trends for more than 3,000 species of fish, amphibians, reptiles, birds and mammals. The current living planet index reveals a global decline of 52% between 1970 and 2010. This means that animal populations are roughly half the size they were 40 years ago. The way that it works is that the average abundance of all those species were given the value of 1 in 1970 and if the averge goes up the index rises above 1 and if it goes down then the index goes below 1. The has been a decline in the living planet index from 1970 onwards - has declines by more than a half. Animal populations on avergae roughly half the size they were 40 years ago.
Aims to understand To understand the problem of species extinction in terms of; Rates Causes Consequences
Outlie of lecture Biodiversity conventions and targets Sources and uses of biodiversity data Past, present and future biodiversity loss
1992 Biodiversity defenition Coined at the Rio Earth Summit - Attended by reps of UN. Convention on Biological Diversity 1992 Biodiversity went global. 'The variety among living organisms from all sources including terrestrial, marine and other aquatic ecosystems and the ecological complexities of which they are part; this includes diversity within species between species and of ecosystems'. As importantly as that, there was a convention that most members of the UN siged up to. It commite them to recognising the importance of biodiversity and trying to conserve it.
Since then a number of related conventions have arisen. There is now several different conventions Convention on biological diversity Convention on trade in endangered species of wild fauna and flora (people take individuals from the wild and sell them - this has had an adverse effect on populations) Convention on the conservation of migratory species of wild animals International treaty on plant genetic reources for food and agriculture ( recognising many crops have wild relatives and the biodiversity of those wild relatives are important and should be conserved) Conservation on wetlands (protection on habits that are unusually threatened)
Following on from those, the way that the industry works (conservation has become an industry) - follwing the summit in 1992 there are annual conferences of the parties (COP) meetings. These are the same people that attended the origional earth summit (members of the UN). Used to see how we are doing and if anything needs to be done better. Since 1992 there has been a hole series of the COPs. 2002 was the 6th COP at the Hague. This was the first time to set targets. What was the target that they set? ''to achieve by 2010 a significant reduction of the current rate of biodiversity loss at the global, regonal and national level as a contribution to poverty alleviation and to the benefit of all life on earth''. 2010 targets have been missed (and were probably unachievable in any case) even though 2010 was UN International year of biodiversity. Here in 2002 for the first time they were asking why is biodiversity worth conserving and the answer is that biodiversity is useful to us - humanist thinking. We value biodiversity because it can help us to alleviate humanmpoverty.
In 2011 they had nother get together (COP) this time at a town called Aichi in Japan. Recognised that they missed the last target so they set new targets. These were known as the Aichi targets, which are to be met by 2020. Now have a series of goals. 2010 was the year of biodiversity. 2010-2020 is the UN decade of biodiversity. There is now much more of a focus on sustainabe use and benefits - have realised now more than ever that biodiversity is good for us because we can expoit it: Goal A - address the underlying causes of biodiversity loss by mainstreaming biodiversity across government and society. Goal B - Reduce the direct pressures on biodiversty and promote sustainable use Goal C - Imporve the status of biodiversity by safeguarding ecosystems, species and genetic diversity Goal D - Enhance the benefits to all from biodiversity and ecosystem services Goal E- Enhance implementation through participatory planning, knowledge management and capacity building. We have moved away from the idea that we are protecting biodiversity because its something we like towards protecting it because we can exploit it.
What is the Nagoya Protocol (adoptedat Nagoya Japan in OCT 2010) Full title: Nagoya Protocol on Access to Genetic Resources and the Fair and Equitable Sharing of Benefits Arising from their Utilization (ABS). Promotes fair and equitable sharing benefits arising from the utilisation of genetic resources. Therby contributing to the conservation and sustainable use of biodiversity. This protocol is about the use and exploitation of biodiversity.
Sources of biodiversity data needed to meet targets. How do we know what the state of the world is in terms of biodiversity? Living plantet index (LPI) gives indication of how we know. But what about where this data comes from. There are two types of data. 1. On the distribution and abundance of species and 2. On the vital rates of those species.
Distribution and abundance 1. Numerous recording schemes, often locally based but interestingly collated nationally 2. Best schemes often involve repeated annual surveys along fixed transects e.g BTO/RSPB/JNCC Breeding Bird Survey (BBS), Butterfly Conservation Monitoring Scheme (UKBMS) 3. Countires richest in biodiversity often have the smallest capacity to collect and collate data. Data on distribution and abundance are by far the moe common types of data. There are all sorts of recording schemes for different fauna and flora in different areas. The best ones involve repeated surveys along fixed transects which are repeated annually. The problem with it si that the countries richest in biodiversity have the smallest capacity to collect and collate these data.
Souces of biodiversity data on vital rates is much scarcer and harder to come by Much more patchy but good for some taxa (especially birds) and some countries E.g BTO nest record scheme has over 600 volunteers which monitor over 30,000 nests each year to record breeding success BTO constant effort site scheme (CES); the same nests, locations and time periods at regular intervals through the breeding season at 120 sites in the UK and Ireland; this provides data on adult and juvenilles and prodctivity and adult survival rates. JNCC Seabird Monitoring Programme; samples breeding numbers and success of selective species at selected sites around the country throughout the UK e.g adult survival at key sites Long term studies in universities and research institutes e.g centre fo ecology and hydrology
Uses of biodiversity data One use - LPI comes from this source of information. There are reports e.g annual british brd survey report which comes online every year. There is the IUCN (international union for conservation of data) who produce lists of threatened species. America does things its own way - has its own endagered species act listings. Can be used to predict what might go on in the future. This data can be used to show annual population trends (the above examples) or for predicting future changes
IUCN red list Contains lots of information about species. Says how endangered they are. Roughly how many species. Varies from species to species depeding how well known they are.
What is the effectiveness of listing under the US endagered species act Taylor et al (2005; Bioscience 55: 360-367) examined population trends for 1,095 species listed as threatened or endangered under the act. Species with protected habitat were more than twice as likely to be increasing and less than half as likely to be decreasing as those without. Species with a dedicated recovery plan were also more likely to be increasing and less likely to be decreasing. Overall, the proportion of species improving increased, and the proportion declining decreased, with increasing time listed. The longer a species had been listed as threatened or endangered the higher the likelihood that the populaton is now increasing. Looks like up until now in the US, the listing of species under this act has been beneficical.
The other thing we can do with this sort of data is? Make predictions, sometimes powerful predictions. As climate changes species will either move or they will evolve or they will die. Looking at the fossil record, the most likely thing is that they will move if they can but how far will they move and how far will they have to move in order to keep their distribution within their suitable climate. If there is data on the distribution and abundance on species and have data on the climate and how it varies across the earth then you can construct a climate envelope for that species, which is a combination of climatic conditions in places where that species occurs. We have several different models now that predict climate change for the future based on global circulation based on carbon dioxide and other conditions. Can predict how distribution of species would need to change.
E.g impacts of climate change on European birds (huntley et al 2008 PLoS One) combine EBCC census data with climate data. Huntley has produced whole atlas of birds. Example species of bird that breeds in britain but isnt particularily comon (grasshopper warbler). Huntley combined data on distribution and abundance with climatic data to work out what the climate envelope of grasshopper warbler is and then used this to predict where the climate envelope will move to in the future. See paper for graph. Predited the envelope will shift north due to warming. Remains only in mountanous areas in central europe and goes from being central eropean to a very much norhtern european species. Thats assuming that its able to reeach those areas and that there are suitable habitats in those areas. These predictions can be done for many different species.
Predicting future changes continued All species are predicted to alter their distribution in the future Some species are predicted to overlap only slighly in the future with their current range (y-axis). Most species are predicted to have smaller ranges in the future than they do now. Winners will have a broader range in the future, some species predicted to grow due to climate change. If the future range doesnt overlap with current range we need to assume these species can move somewhere completely knew and hope there are habitats availible. The more the overlap between current and future range will be the safest.
Thomas et al 2004 Nature Used a simular approach for a range of taxa in sample regions covering 20% of the earths terrestrial surface. Used mid-range climate warming scenarios, they predicted that by 2050, 15-37% of species will be commited to extinction as a result of reductions in geographical range size. In some cases, suitable climate change will be out of reach e.g if yourre a lizzard and suitable climate is across the sea, either need t evolve or will go extinct. Climate change is happening too rapidly to evolve.
Extinction Natural phenomenon More than 90% of species that have ever existed have gone extinct but pattern of extinction there have been 5 major extinctio events in the history on earth. Geological eras are defined to some extent by these extinction events. A way to remeber the geological eras and events is camels often sit down carefully, perhaps their joints creak. At the end of the permian 95% of spcies on earth lost.
There are grwoing suggestions that we are currently enering a 6th major extinction event Survey carried out in 1998 by the american museum of natural history www.mysterium.com. 70% of biologists believed that in the next 30 years up to 20% of species would go extinct. 30% believed that up to half would go extinct. Yet only 1,200 species are officially recorded to have gone extinct since 1600 (IUCN red list). There somewhere around 12 and 20 million species of eukaryotes on earth. So is talk of 6th mass extinction justified?
Lecture 2 Extinction rates
Widespread concern that we are entering a sixth global mass extinction event. But in the past 500 years, only 1200 species have been recorded as going extinct. Reasons for the mismatch between recorded recent extnctions and predicted future extinctions 1. Least important but worth knowing about: Time-lags in recording and designating species as extinct. There is a 50 year rule. How can you tell when a species is extinct? No species can be declared extinct until the species havent been seen and efforts have been made to see them for at least 50 years. There are a number of species that are essentially waiting to be declared officially extinct even though we are pretty sure they are extict. e.g Yangtze river dolphin or baiji are listed by the IUCN as critically endangered, but have probably been extinct since 2007. There are also others in waiting e.g christmas island pipistrelle (2012). Its not unknown for species that were throught to be extinct to reappear. Some species reapear after more than 50 years (50 years rule isnt failsafe) But this time lag is not enough to account for size of dispaity between the recorded and predicted extinctions.
2. Unrecorded extinctions Some species may have gone extinct without even being described. Problem with this is that if they have never been described then we dont know how many. However we can have a look at historical records and make inferences. David Steadman did this in 1995 in Science where he used subfossil evidence (old bones from archaelogical sites) not old enough to become full fossils yet. He looked at sites on Pacific islands, dug up old bones, took them away to figure out what species they are. He worked out how many were a species that are still around and how many arent around anymore. Results suggest that polynesians eliminated about 1,800 species of endemic island birds as they colonised the pacific islands over the last 2 millenia. These islands had only been colonised in the past couple of thousand years. That is a 20% worldwide reduction in the number of species of birds. This suggests the current global extinction concern does have historical basis.
No obvious decrease in rate on those pacific islands up to late 20th century e.g new zealand: extinction rate after arrival of Polynesians very simular to that after arrival of Europeans New zealand is one of the pacific islands, it is quite isolated and wasnt colonised by people until 900AD from polynesian records, this is around when they first arrived. We know from subfossil remains that when they first set foot on new zealand there were 125 species of land and freshwater birds so excluding sea birds. By 1769 when captin cook arrived, there were only 89 of those species left and in 1993 when the data for this paper were collected there were only 81 species left so 36 species went extinct between the arrival of the first polynesians and the arrival of first europeans. A further 8 species in the 200 odd years since the arrival of the first europeans. The more species that are present or availible to go extinct, therefore need to look at the rate of extinction per species per century.
We shouldnt think this is a phenomenon that is restricted to the pacific islands, there have been simular losses closer to home Lonesome George. Before his death in 2012 in the Galapagos islands, Lonesome George became famous as the last remaining Geochelone abingdoni on Earth. Giant tortoises once also inhabited the canary islands, as did many other species now extinct. Extinctions followed very simular pattern to those in the Pacific islands. Both the first (pre-European) settlement of the islands <2000 years ago (Gaunches from North Africa) and European settlement the 15th century. Dune shearwater Puffins holeae went extinct following the arrival of Gaunches and then the second species of shearwater (lava shearwater) P. olsonii went extinct following arrival of the europeans. High rates of species extinctions following human settlement of the islands
Not just islands that are affected Over the last 12000 years a high proportion of herbivores have been lost over continental mainland over that period. There is a debate over the causes of those extinctions- thought to be a combination of climate change and hunting by people.
Third reason for mismatch. 3. Extinction rates strongly depend on the interval over which they're calculated Extinction rates over the past 500 years are not comparable with mass extinction events that occurred over periods of up to 1 million years. So maybe there are good reasons to be concerned.
How can we predict future extinction rates? There are three main approaches. 1. Species area estimates Based on established relationships between land area and species richness. By reversing this species-area accumulation curve, you can extrapolate backwards to calculate expected species loss from loss and degredation of key habitats - particularly tropical forests. Can generally establish a relationship between land area and species richness. Can then apply this to conservation by saying the more habitat availible, the more species there will be in it. Can then reverse that relationship to a situation where we are losing habitat and we can say if land is lost and suitable habits are lost then we will lose species. By knowing how much habitat we will lose, we can predict how many species we are going to lose. If this is done across different habitats, we can come up wth estimated extinction rates of 5-30% of all species on earth per decade. However hgiher estimates from the 1980s clearly arent true
How reliable is this method generally? Not very reliable: species-area relationships always overestimate extinction rates from habitat loss. This was pointed out in 2011 in a paper in nature by He and Hubbell. They pointed out that we use the idea that as we lose land we will lose species and this is used to predict species losses. Then go along to for example, an area of a forest that has been lost by half and we expect species should have been lost by the same proportion. However the species are still there. This is due to extinction debt. This means they will go extinct, they are commited to it but they havent yet become extinct. He and Hubbell pointed out that this extinction debt is largely a sampling artefact.
Why is extinction debt largely a sampling artefact? The area needed to remove the last individual of a species (extinction) is much larger than the area needed to encounter the first individual. Hence, extinctions through habitat loss need much greater loss of habitat than previously thought. Habitat loss is a real and growing threat but impacts on rate of extinction.
Mendenhall et al 2014 questioned why these models are always used on islands. We are acting as if islands represent the whole world. We should use countryside biogeography, recognising the importance of matrix. Countryside can support the movement of species wheras the sea cant.
2. Estimation from empirical data on actual extinctions Can use fossil evidence which suggests 'normal' background extinction rate of approx 0.25 extinctions per million species-years (before humans even evolved and in the absence of mass extinction events). This can be done to look at how long species persist as fossils from when you first see them to when you last see them. Can look at how many species disappear. Have to calculate this in species years. The more species there are as fossils, the more of them are able to go extinct and because it is a slow thing, the normal unit is extinctions per million species-years. E.g 5,500 living species of mammal, so if they were going extinct at the background rate we would predict that there should be 1 mammalian extinction every 700 years (rate of 0.5 million species-years).
Stuart Pimm et al, 2006 suggested that globally, recent extinctions of birds are 100 times higher than this, and predicted that they will rise to 1500 times higher by the end of the 21st century. This is an alarming prediction This estimate is based heavily on the data that we saw earlier in the lecture. 20% of all bird species were wiped out over 200 years. That has happened now, those species have gone. There is no reason to assume that that rate of extinction will continue and be passed onto mainland continental areas. We have seen in New Zealand that rates are slowing down there but that is not to say that other islands e.g pacific islands are able to predict what will happen on continental areas. So this is a controversial prediction.
3. Estimation from Red Lists of threatened species Most compelling way of predicting future extinction rates. IUCN list species and put them in categories from percently ok, to vulnerable, to threatened, to endangered, to critically endangered to extinct. Barnosky et al 2011 Nature looked at data from Red list and what is happening with time. We have different groups of organism, mammals, birds, reptiles, rayfinned fishes, invertebrates etc. The white figures are % of species that have gone extinct in the last 500 years. Can see there is no real reason or concern just looking at those. 1% of mammals, 1% birds, 1% reptiles, 1% amphibians and so on. Gets scary when looking at the proportions of species that are currently threatened, endangered or critically endangered. 22% mammals, over half all known species of gastropods, 2/3 of all species of cycads etc. Really scary if you look at how those proportions have changed with time, looks like species go from being perfectly ok to vulnerable and then threatened until extinction.
How long will it take us until 75% of species have gone extinct if the rate at which species move along this conveyor belt continues. 75% is what we consider as a mass extinction. This is a big IF. It may be that these predictive extinctions of threatened species may be unduly pessimistic. e.g if conservation efforts targeted at endangered species are successful (see taylor et al. 2005 in previous lecture)
Or they could prove unduly optomistic e.g if additional factors have additive effects (e.g rapid climate change) see Thomas et al 2004 in previous lecture - this could push things to extinction even faster. e.g Batrachochytrium dendrobatidis (Bd frog fungus affecting skin- spreading across amphibians, causes skin to become hardened, stops frogs breathing because they breathe through skin. New phenomenon) and they die causing global amphibian extinctions). And because most unknown species have small ranges and are likely to be threatened, so current estimates of proportions of threatened species are probably too low.
Hawaiian honeycreepers case study Group of birds called hawaiian honey creepers so the same way as in the galapogas, finches radiated after they had arrived and filled a lot of empty niches and formed a whole suite of new species, same thing happened in hawaii so some finch or other turned up sometime in the deep geological past and ancestors raidiated out, evolved into new species to fill vacant niches, gave rise to a whole range of species of honeycreepers (birds that walk around in the undergrowth. Some feed on nectar and plants, insects, seeds and some feed on blood. They peck at seabirds, make wounds and drink blood from wounds.
In 1827 Culex pipens introduced onto Maui by British sailors from a whaler ship (The wellington) was returning from the sub antarctic from an unsuccessful whaling trip. Common practice was that they had barrels on board that contained fresh water that the crew could drink. Over the voyage the water would go off due to bacteria. Whenever they could, when they passed land they would stop off and empty the foul water and fill up with clean water. The wellington did this at Maui (Hawaiian islands). By doing so they introduced larva of mosquitoes. Suitable conditions for mosquito larvae to develop. This simple event introduced mosquito to the island. Mosquitos are vectors of diseases such as avian malaria and pox. Having the vector inst a problem if you don't have the disease, mosquito larva don't have the plasmodium in them, they need to feed on an infected animal for them self to get infected and pass the disease on. Maui was visited repeatedly, ships would also have livestock on board, riddled with avian malaria and pox. Mosquitoes would feed on the birds on the ships and then go back to the island and feed on the local wildlife, introducing avian malaria and pox into them.
What happened to the birds on Hawaii? All lowland endemics (24 species: 80% of endemics) now extinct but some species are still present in the uplands. Malarial parasite has minimum temperature requirement to complete development within vector. The warming climate may be extending upper elevation of both mosquitoes and malarial parasites into uplands.
Generally over the last 12000 years since the and of the placeothene, weve seen mass extinction of mega herbivores on continents, seen 20% of birds dissappearing from pacific islands and probably simular rates of extinction on other islands areound the globe. We have predictions possibly as low as 24o years for us to lose 75% of species even from continental landmasses - this might be unduly pessimistic or optomistic. Overall it looks like following the end of the last series of ice ages we are now an era which we can define geologically, officially we are currently in the holoscene (era following the last ice age). Unofficially we are now in a whole new geological era called the Anthropocene era. Earth is dominanted by the presence of homosapiens. This was first coined by Crutzen in 2002 in a paper in Nature in recognition of these geological scale changes produced by human activity. Half of the lands surface has been transformed by people. Most of the wolds major rivers have now been dammed or diverted. If they havent there are plans that there will be. More nitrogen produced by fertiliser plants than is fixed naturally by all terrestrial eco systems. Fisheries have removed more than a third of primary production of the oceans costal waters and continue to do so annually. Humans use more than half of the worlds readily accessible fresh water run off for their own uses and we have had a significant impact on the global climate and even if we stopped burning fossil fuels today, those changes in atmospheric composition and ocean chemistry would persist for arguable for many million years to come.
Even if humans disappeared tomorrow - someone who comes along to the earth in a million years, when the earth has been reduced to an a layer of stuff it will have evidence of our activities. Term hasn't yet been formally recognized. To be formally recognized it has to be voted for by the international Stratigraphy commission in 2017. They didn't vote for it last year but they might this year.
Lecture 3 Small populations: the problem of rarity
Rarity is a perfectly natural phenomenon. All natural communities contain some rare species. Most species have an average abundance. However there are species with a naturally low abundance. Some species have only 1 individual. Should be surprised that some species are naturally rare. This issue is that distribution and abundance have a positive relationship.
Speckled wood butterfly have been spreading northward over the last 20 or so years. Known from data collected from butterfly monitoring scheme. People walk along transects throughout the UK and count number of individuals seen along transect. If a species occurs at all 42 sites it is clearly a widespread species. Conversely a species that only occurs at one site clearly has a narrow distribution. Species that are widely distributed are also abundant where they occur. Those that have narrow distribution have a low abundance even where you find them. This creates problems for those species.
Distribution and abundance are related Rare species often have restricted ranges and species with restricted ranges are often rare even where they do occur. They therefore face a double whammy. Not only are they rare in terms of distribution but they also rare in terms of their abundance. Places them at risk of extinction.
This is supported by fossil evidence e.g bivalve and gastropod molluscs from the late Cretaceous. They fossilise well because they have got hard shells. Good group to look at in terms of fossils. Losts of species in terms of bivalve and gastropod mollucs in the fossil record. Particularity species rich in the late Cretaceous. A group hit by the affects of the asteroid, they didn't go extinct the way that dinosaurs did. But within the cretaceous before the asteroid hit, species were still coming and going. Can work out how widely distributed a species is and how long it has persisted in the fossil record. Species with a wide distribution when they were first encountered in the fossil record, hung around for a long time in the fossil record and vice versa. Might be an issue here to do with sampling, but even counting for this, it still seems to be species with restricted ranges had a higher likelihood of going extinct more quickly from the fossil record. This implies being rare is dangerous.
Consequences of relationship between distribution and abundance If we have a rare species and we do something that makes it even rarer (e.g we hunt it or we introduce a predator or new competitor species), leading to a reduction in population size. Because population size and distribution are related, we can expect that if we reduce population size then we will reduce its distribution. This occurs without any changes in habitats. Similarly if we do something that reduced the range of the species maybe through habitat loss, we would predict that the abundance will decrease even in that remaining protected area.
Why does this occur? If we induce a reduction in population size, why does it induce a reduction in the range? This occurs because species are not evenly distributed through their ranges. What ever occurs on the outer edge is no good for species. As you move in towards the core of the distribution, conditions improve and the ideal conditions are somewhere within the core of distribution. Not necessarily the geographical dead center. If conditions improve then survival is likely to be higher and the reproductive rate is likely to be higher. A combination of survival and reproduction results in little r. r=instantaneous rate of population growth will reach its maximum somewhere within the core of the distribution where conditions are as good as they ever get (r=0). Moving away, conditions aren't ideal but they are still good (r>0) and population in this area tends to expand. Move even further away (r<0) because conditions arent good enough for pop growth to be higher than 0. In this range, individuals either dying too quick or reproducing too slow for population to persist.
So how does the outer edge persist? It is being maintained by migration from the source area. When r>0 the population is tending to increase and getting more crowded so individuals move away from the high density area into this margin habitat. If its a territorial species it will be those individuals that cant maintain territory. This population around the margin is called a sink population and the population nearer to the core is called a source population.
If humans then come along with shotgun or whatever and start killing individuals in source area and reduce population size, there is now enough territory to go around. Individuals no longer need to migrate outwards in order to find a territory. Cut off the movement of individuals moving from source to sink area. The distribution then collapses back into the source area.
How can reducing the range result in a decrease in abundance even in protected areas? Its because reductions in the range usually occur as a wave of habitat loss or invasion. In the real world, we don't usually act on the source population e.g parachute into the center of a rain forest and start logging there. We start at the edge because we need to get big machines in. Reduction in range usually occurs as a wave of habitat loss, it pushes population towards the edge of the former range where conditions are poor. Once the main source area has been deforested and then suddenly we decide to protect the remaining area, it could be that we are protecting mainly sink habitat where climatic, habitat etc conditions are too poor to sustain populations of certain species. Ends up with population size within the remaining habitat falling.
Case study: New Zealand Takahe Porphyrio hochstetten. New Zealand mainly 2 islands. On the south island there is a species of bird the Takahe which was once abundant on the south island. Had a related but separate species on the North island. The habitat that it liked was boggy, swampy ground in the lowlands. Following the arrival of humans, particularily europeans, swamps throughout the lowlands were drained for agriculture and whole range predators introduced. They are ground dwelling so badly affected by predators. This meant that lowland populations were wiped out. This species was thought to be extinct by 1898. 50 years later they were rediscovered in the mountains in 1948 but they had a marginal habitat with low breeding success so much so that despite strenous efforts to conserve the remaining population, it continued to decline. The mountains of south islands are sink habitats. Conservation required catching individuals and moving them. New zealand has lots of offshore island that remain predator free. They have been moved to wet boggy lowland habitats where it isn't so cold that chicks die of hypothermia on predator free islands.
Other examples include: Species of duck called madagascar pochard (endemic to madagascar) was heading extinct due to human development draining the pollution of ponds and lakes. 1 remaining population was on the lake far north of madagascar in an area with very low human population density. Problem is it was a steep sided lake. Ducks can either dive down to surface and feed at the bottom or mallards stick head under water and feed in shallow water (mallards only feed in shallow water). One lake where this species was found to remain was steep sided meaning there was very little remaining foraging habitat. Conservation required moving them into shallow sides lakes.
Red kite in the UK Population in mid wales clinging on for years despite any effort to get them expand, they couldn't because breeding success and survival weren't high enough. Reason why red kites are now common and abundant in the UK is because individuals from Europe were deliberately introduced into core habitat.
How do rare species go extinct? Two types of forces push them towards extinction: Determinisic: will inevitably have an affect e.g habitat loss due to deforestation or hunting. Stochastic: Might have an affect but then again might not - its a question of chance
Deterministic processes occur when? Something is taken away e.g habitat loss or when something nasty is added e.g predator, competitor, pollutant or hunting by humans.
What are Stochastic processes? Extrinsic: characteristics of the environment a population lives in e.g probability a forest might burn down, probability of flood or infectious disease. Intrinsic: characteristics of the population itself
Intrinsic stochastic process Demographic uncertainty: Natural variation e.g birth and death rates or offspring sex ratio. In large populations, differences between individuals balance out: in small populations, they can produce erratic variation at population level.
Intrinsic stochastic process Genetic uncertainty: a) genetic drift. Loss of alleles by chance: rare alleles may fail by chance to be passed onto the next generation. The smaller the population, the more chance a rare allele is lost b) Inbreeding. Probability of inbreeding increases as numbers decline.
Very rough rule of thumb Need 50 individuals to avoid inbreeding and need 500 to maintain long term adaptability. These numbers are for effective population size that's contributing to the next generation. Doesn't include individuals that aren't breeding so there is always less than total population size.
How does all of this come together to effect the likelihood that a species will go extinct? Deterministic and stochastic process tend to interact in a positive feedback loop (extinction vortex). Most extinctions start with a dterministic event (e.g hunting, habitat loss, introduction of predator) that generally reduces population size. Small populations are then vulnerable to effects of stochastic processes. These can push the population closer and closer to extinction. They can do that even if the deterministic forces are removed.
Case study of how small populations go extinct: Heath Hen When extinct from mainland North America. In 1908 the last reaming refuge popualtion on Martha's Vineyard protected by 1600 acre reserve - no more hunting or habitat destruction. 1915 -2000 birds in refuge population. It was hoped this was enough that there wouldn't be a problem.
In 1916 there were very stong gales, fire destroyed breeding habitat, harsh winter, heavy protection by Goshawks, population reduced to 150; inbreeding depression. 1920 - outbreak of poultry disease introduced accidentally from mainland. 1927 only 13 birds left, 11 of them male 1932 last bird died. Catastrophic and environmental stochasticity reduced population to low levels, demographic stochasticity helped finish it off.
Extinction vortex can be avoided: Mauritius kestrel Falco punctatus Endemic to Mauritius, confined to low land forest. Breeds on cliff edges, feeds by ambushing geckos in trees. Starting in 1800 the kestrels were subjected to deterministic processes e.g loss of habitat, pesticide poisining, degredation of habitat (people introduced privet which is the stuff found in hedges around Leeds and strawberry guava creating thick foliage which stops these species from foraging effectively) shooting, predation by introduced monkeys, rats and indian mongoose.
What happebned as a result of all these factors? By 1973, the population was reduced to two breeding pairs plus 2-3 non-breeders. ICBP established a rescue programme. One pair taken into captivity but the female died - world population down to 6. The remaining pair reared three offspring in 1974 but then were killed by cyclone Gervais (Stochastic event). In 1979 appointed new personnel called Carl Jones. He knew if you take kestrel eggs at the right time then females can re lay in the wild especially if you give them supplementary food. Fed the remaining breeding pairs and took eggs into captivity: females relaid. In 1981 three chicks were reared in captivity, they were all male (demographic uncertainty). Next year three more including a female. He used this to establish a captive breeding population. Wild birds also successful, helped by supplementary food. Population now more than 500 birds - very successful.
Carl 'There are no hopeless cases. Only expensive cases and people who give up hope.' Mauritius kestrel was chronically rare and inbred (population never more than >300 birds). So this avoided problems of genetic uncertainty. It would be a much bigger problem for species that are not naturally rare.
Lecture 4 Consequences of loss of biodiversity
Increasing recognition of ecosystem services provided by biodiversity e.g provisioning (food), agricultural, supporting services and regulating services.
Climatic regulation - one of the biggest services. Can see the effects of losses of biodiversity on climate by looking at the destruction of tropical forests. Destruction of tropical forests accounts for one quarter of all anthropogenic carbon emissions. This is 3 billion tonnes of carbon a year. The Kyoto protocol called for international carbon market (REDD - reducing emissions from deforestation and degradation). This is the idea that we should be paying big countries that have a lot of forests - to not destroy their forests. The Financing for REDD was enshrined in Warsaw Framework for REDD-plus (COP 19, Nov 2013). The plus is essentially reducing emissions from deforestation and degradation and also reducing losses of biodiversity.
Another large scale thing biodiversity does is hydrological regulation. So as well as regulating the climate it also regulates the water cycle. In the tropics, evapotranspiration contains atmospheric moisture in air passing over extensive vegetation (rainforests). Air that passes over extensive vegetation in preceding few days produces at least twice as much rainfall. Amazonian deforestation predicted to reduce precipitation by up to 21% by 2050, due to less efficient moisture recycling - important for farmers that need rainfall for growing crops.
Pollination 2/3 of food crops benefit from pollination and those services are worth about £440 million a year in the UK. Globally worth something like $160 billion, which is about 10% of the total agricultural food value. Importance of biodiversity of pollinators in the UK is recognised by the Insect pollinator initiative
But how much biodiversity do we need? OR what is the relationship between species richness and ecosystem functioning? One of the indicators often given to hgihglight the importance of pollinatiors - in California have almond crops which are insect pollinated grown in monoculture, there's a shortage of pollinators so end up with whole convoys of trucks rolling up to almond groves when they are in flower and people unpacking crates of bees, so that bees can pollinate. But they are all one species of bee.
Would it be more efficient to have fewer species - if we only had the species that pollinate most efficiently? This is what the Californian almond industry does.
So, how do we understand how the relationship between species richness and ecosystem functioning? One way is to experimentally set up systems that have different numbers of species and see how they function. A great system for doing this is grasslands because grass grows quickly and can set up grassland plots relatively easily so you can seed an area of land of grassland with different species richness and can see how well it functions. There are a number of predictions we might have to what the relationship might look like.
There are a number of predictions we might have to what the relationship might look like. 1) As we increase species richness the system performs better but only up to a point. After this point adding more species doesn't add much in terms of functionality. 2) Bit like popping rivets out of a sheet of metal. At first have redundancy so not much happens, then get to critical threshold to which redundancy drops very quickly but to a new stable level, lower level of ecosystem functioning but it is stable. 3) Might be that there is no predicatble relationship between species richness and ecosystem functioning (idiosyncrasy). Need evidence to be able to which of these patterns real experimental ecosystems actually conform to.
Data support positive relationship, with some evidence of redundancy. Cardinale, Sankaran et al 2006. In each case the X axis shows an increase in species richness and Y shows some measure of ecosystem functioning. On first one we have plant species richness in out experimental grassland and total plant cover. It goes up to a point and then tends to level off.
If rather than plant cover, we look at net primary productivity, we find a similar relationship. It goes up at first with increasing number of species and then levels off.
If we look at the biomass of microbes within the soil, the more species of plants we have, the more species of microbes we have but notice there are only 3 data points. So the straight line is simply the shortest distance between 3 data points which isn't convincing in understanding overall relationship.
This one looks at the number of mycorrhizal fungal species in the soil in these experimental grasslands in relation to shoot biomass. Looks at mycorrhiza in soil and biomass of green plants above the soil and the more species of mycorrhiza there are in the soil, the better the plants are growing but only up to a point.
Looking at CO2 flux, this is a measure of how much energy the system is capturing through photosynthesis and then is using for respiration. The more species there are, the greater flux there is but there is so much noise around the data. So not very informative.
Might interpret this one as a decrease in functioning but it isn't. We have the number of consumer species (herbivores) and the biomass of plants. So the increase in ecosystem functioning here is an increase in herbivory and an increase in herbivory would lead to a reduction in the biomass of palnts. So this is also an increase in ecosystem functioning, there is no evidence of ecosystem redundancy here.
In at least half of the above graphs, there is strong evidence of redundancy, why? One only has 3 data points, one is so noisy we cant see a relationship.
Why should ecosystem functioning increase as species richness increases? Cardinale et al 2007 produced paper on the possible mechanisms. 1. Sampling 1. Sampling: individual species differ in their contributions to ecosystem functioning. Some don't contribute very much and some always contribute a lot. If you have more species then more likely by chance to have present one or more of those species that contribute a lot. So as species richness increases, there is a higher chance of increasing species that contribute a lot so ecosystem functioning is likely to increase.
2. Species complimentary Species differ in their resource use. More species means resources are used more thoroughly e.g plant cover. The reason why plant cover increases as you increase plant species richness, might be because when there are more species present, more likely to have one species that grows rapidly and has massive leaves and covers a lot of ground. Or it might be that the more species present, the more species there are making use of different micro environments, so don't get a system dominated by one species but resources present are being used more efficiently.
3. Positive interactions More species leads to more mutualistic interactions there are, this could drive this relationship. What Cardinale suggested was that the main effects are sampling and complementarity. He said that this third possibility didn't seen to be born out by the evidence. Mainly the first 2 which drive the relationship.
If we are interested for biodiversity for what it does for us and what it will provide, then does this redundancy mean that there are lots of spare species and we can afford to lose maybe lots of species without paying any real penalty in terms of ecosystem goods, service and functioning? Possibly. But: We don't have enough evidence yet. 1) We need insurance. A pool of species that can buffer a system against environmental stress. In other words, some redundancy can be a good thing. 2) We need a longer view. Consider not only short term relationships but also long term consequences.
All data were collected over short period of time - single growing system. Not representative of real world. Also range of species on X axis of graphs are unrealistic, many systems have much more species than what is shown here.
Loss of Pleistocene megafauna ( huge herbivores used to trundle around that disappeared after the end of the Pleistocene era). If we look at parts of world where those species used to occur,. There is a recurrent pattern. Many species that have fruit which are too tough to be dispersed by any living species. One is called Cassia grandis (massive beam with thich shell to pods), another is Dioclea megacarpa. Nothing on earth that can disperse these seeds. They relied on megafaunal species e.g glyptodonts, mastodons) for dispersal. They are still found along rivers, so if they have the chance to soften up, they can germinate along rivers, if they happen to drop in river they soften and then disperse on river bank. Location is restricted to river banks.
There are other examples. Species redundancy might be a feature of ecosystems, so we can lose species without much happening. But we all know examples of where loss of species can have catastrophic effects on ecosystem functioning. One good example that is often cited is Barro Colorado Island (Terborgh, 1988) Small island in Gatun Lake, Panama: formed when river Chagres dammed to provide water for Panama Canal. When that river had been formed, what had been hill tops in the valley, became islands in a lake. When islands were isolated they had several large predators e.g pumas, jaguars. They soon disappeared because islands were too small to support them. If these species were redundant it wouldn't have had much effect. Instead it did and resulted in a 2-10 fold increase in medium sided predators and omnivores e.g cotimundies, pacas, monkeys, peccaries. This meant by 1970s 45 species of bird had disappeared including all ground nesting species e.g ground cuckoos, wood quail, ant thrushes. Probably due to insupportable high nest predation by the inflated populations of medium sized predators.
A similar thing happened in Venezuela Set of islands created by building damn for hydroelectric scheme. Small islands less than hectare in size. After short period of time there are no predators of vertebrates left meaning that densities of seed predators (rodents, howler monkeys, iguanas, leaf cutter ants) and herbivores are 10-100 times greater than on the nearby mainland. This has meant that densities of seedlings and saplings of canopy trees have been severely reduced. This is preventing the forest from regenerating. Due to loss of predators, over time the loss will lead to loss of the entire forest.
Example of a totally different system. In Alaksa there was overhunting of sea otters. Hunted extensivley for fur, almost driven to extinction which reduced what was a complex marine ecosystem of kelp forest. Kelp that grows in the pacific is different to the UK, it is huge and complex with many species. Removal of one species (sea otters) reduced this complex ecosystem to a 2 trophic layer ecosystem to simple sea urchins grazing on algae. Relaxation of hunting pressure has restored the system in some places. The problem is predation by orcas: normally prey on much larger seals and sea lions, but these they have declined: due to a decline in fish stocks: due to impacts of climate change on primary production. So it is a bottom-up effect. The lack of seals and sea lions means orcas have learned to prey on these instead.
Finally plant and pollinators Biesmeijer et al (2006) study of bees and hoverflies in Britain and the Netherlands revealed causal connection between extinction of pollinator species and the plants that they pollinate. Last year Bill was co author on paper in nature called historical nectar assessment which records the fall and rise of Britain in bloom looking at causal links of biodiversity of plants and pollinators. Take a read of these.
Two opposing sets of evidence. Rather than think of ecosystem functioning lets look at an aspect of their function e.g they stability. Does the stability of ecosystems vary with species richness? Are species rich ecosystems in some way more vulnerable? 1. Species rich communities are more stable: Complex food webs with more cross-links enables losses to be absorbed with less impact on remaining species - this view was prevalent up until 1970's 2. Species rich communities are less stable: With more species, each species is less abundant, so more at risk from extinction processes. Consequences then propagate more widely through multiple trophic connections: prevalent view during 1970's-1990's. When we reduce pop size, species are at danger of being dragged into extinction vortex. But if there is an environment with a set amount of resources, if you add more species then resources available per species must decline and so the population size of species decline. More species = smaller populations = greater risk of extinction.
The simultaneous occurrence of these two opposing views depending on how you think about it was called what? May's paradox. Both types of though seem perfectly logical. Which one is correct?
Why was it called Mays paradox? Bob May was prominent in solving the dilemma. He realized that whether you think species rich systems are more or less stable, depends on what you mean by stability. He said you have to divide stability into two concepts. 1. Species-rich communities have lower constancy: they are more likely to lose species 2. Species-rich communities have higher resilience: They are more likely to continue functioning normally despite lower constancy.
Evidence to support this idea comes from experimental grasslands Tilman et al (2006). Seed grasslands with different species richness. This time they allowed the system to run for a number of years. They looked at plant cover, net primary productivity and variation from year to year in plots with different species richness.
What they found supports what we expect from this division of stability into constancy and resilience. What were the results? Plots with higher species richness had not only higher productivity but also less variable productivity over a 10 year period. At the same time, species rich plots varied much more in terms of relative abundance of each species. Stable biomass output was maintained not because of steady populations but despite a more unstable dynamic system when there were more species present. So having more species is a bad thing if you want constancy but a good thing if you want resilience.
Why in terms of the behavior of food webs, complex systems should be more stable. The idea that species rich systems are less stable was prevalent up until 1990's. We have no gone back to a view similar to 1970's. Even if we look at constancy and the stability of population sizes we might expect that some species rich ecosystems will be more stable than non species rich ecosystems. This is because of the stabilizing effect of weak trophic interactions. For years people look at ecosystems and communities and trophic interactions and have been recognizing that in some cases there are strong trophic interactions and it has been assumed that what is most important in the functioning of the food web is the strength of these trophic interactions. It has been recognised over the years that there are also weak trophic interactions and these are assumed to be unimportant. Its only recently people have realised how wrong that thinking is and how important these weak trophic interactions are in ecosystem constancy and resilence.
Example to explain this: Coast of britain. Mussles grow in mussel beds around thew low tide mark, they are preyed on by herring gulls and dog welks. Both prey on common mussels. If just had these two strong tropic interactions, we would expect Lotka-Volterra cycles. Herring gull eats mussels, mussel abundance decline, because mussel abundance declines, herring gulls have less food so their abundance goes down which means mussels can increase again. The same would be true of dog welks. Three populations always cycling and never reach constancy. In reality there is a weak trophic interaction in this system. Insertion of a weak trophic link stabilizes all three populations: if mussles start to decline, gulls switch diet. to eat the dog welks. Mussels recover without decline in gulls and whelks decline less than before.
Conclusion Species-rich communities may support less stable populations but more stable ecosystem processes But: Many natural communities are already impoverished Many of the most threatened species are key species (e.g large carnivores) whose loss would have the most far-reaching consequences. Recent data suggest that weak trophic interactions (which usually appear unimportant) may in fact play a vital role in stabilizing food webs, and species-rich food webs probably contain more weak interactions
Show full summary Hide full summary

Similar

GCSE Biology AQA
isabellabeaumont
Cells And Cell Techniques - Flashcards (AQA AS-Level Biology)
Henry Kitchen
Cell Structure
megan.radcliffe16
Exchange surfaces and breathing
megan.radcliffe16
BIOLOGY HL DEFINITIONS IB
Luisa Mandacaru
Key Biology Definitions/Terms
courtneypitt4119
Biology B1.1 - Genes
raffia.khalid99
The Circulatory System
Johnny Hammer
Biology Unit 1a - GCSE - AQA
RosettaStoneDecoded
Plant Structure and Photosynthesis
mckenziedev
AS Biology Unit 1
lilli.atkin