Cabot Institute blog

Find out more about us at

Monday, 13 February 2017

Contemporary Eco-Cities: An improvement on previous work?

History offers up many grand ideas for how urban planning and design can be used to improve cities and society to be more sustainable and liveable. These ideas include early urban reforms by David Dale, Robert Owen, and Titus Salt, Benjamin Ward Richardson’s City of Health, Ebenezer Howard’s Garden City and Le Corbusier’s Contemporary City, amongst others. The eco-city, as an idea that initially developed out of the 1960s grassroots environmental movement, should be considered within this long tradition of ideas for the betterment of cities. The construction of ‘eco-cities’ in the recent decade has also been controversial. The aim of this article is to introduce the ideas and practices of contemporary eco-cities and to discuss the extent to which they can be regarded as an improvement on the work of previous urban reformers.

Eco-city concept and development stages of eco-cities

The term ‘eco-city’ was first coined in 1987 by US-based eco-city pioneer Richard Register as ‘an urban environmental system in which input (of resources) and output (of waste) are minimised’. As the eco-city concept began to be used more extensively, it became more broadly defined, ergo, there is no single accepted definition in the literature. Recognising and permeating ideas of nature in the city into the concept of sustainable cities should be considered essential in setting a framework for eco-cities. This idea can be seen as a continuation of planning trends which attempted to reconcile the nature-city relationship, beginning with Howard’s Garden City movement. The approach is also novel, in that it represented a wide range of factors which coalesce around sustainability in the contemporary context of global climate change, environmental degradation and environmental politics.

There are three stages in the development of eco-city. Early initiatives in the 1970s and 1980s were locally-oriented, bottom-up approaches which aimed to improve environmental conditions, with only few examples of practical projects. The period of 1990s to early 2000s marked the emerging stage of eco-city development, when balanced sustainable development was adopted as a principal objective. At this stage, national and municipal governments started to develop eco-cities and eco-towns, most of which were redevelopments or expansions of existing towns and cities as demonstration projects. Successful examples include eco-towns in Japan and small-scale ones in Europe. Since the mid-2000s, a number of ambitious built-from-scratch eco-city masterplans have emerged, aiming to make entire towns and cities highly sustainable. These top-down proposals are occurring predominantly in Asia, including the highly publicised Dongtan Eco-City, Sino-Singapore Tianjin Eco-City and Caofeidian Eco-City in China and the Masdar City which was proposed as a zero-carbon city in the UAE. The latter two stages of eco-cities are the foci in the following discussion, considering the development scale and significance.

General visions and planning of contemporary eco-cities

Contemporary eco-cities and eco-towns are, in general, an improvement on previous ideas and works in terms of their visions, planning and objectives.

Almost all smaller eco-town plans and holistic eco-city blueprints in recent years are emerging as reflections of and responses to the global context of climate change and environmental degradation, with emphasis on cutting-edge technology, clean energy, and circular economy to achieve sustainable living. This indicates that the finiteness of resources as well as the ecosystem itself has been more extensively recognised. This is an improvement from earlier works. Earlier works targeted urban problems including sanitation and personal health in the case of ‘Hygeia, City of Health’ by Richardson in 1876, and poor working and living conditions in the case of Saltaire and New Lanark, but hardly realised the limits of the environment. An example was Howard’s claim in his 1902 book Garden Cities of Tomorrow: ‘…to a more noble use of its infinite treasures. The earth for all practical purposes may be regarded as abiding forever’. Today’s improvement leads to the consideration of intergenerational ecological justice, being an essential principle of the development of eco-cities.

Moreover, there are three forms of eco-city projects identified: new developments, expansions of urban areas, and retrofits where existing cities adopt eco-city principles. Unlike earlier visionaries such as Ebenezer Howard and Le Corbusier who tended to reject the idea of gradual improvements to the conditions of existing cities in favour of comprehensive transformations of the urban environment, the planning of contemporary eco-cities/towns includes various forms of urban development and redevelopment. Whilst the ambitiously proposed built-from-scratch eco-cities in China and the UAE belong to the first type, most examples of smaller-scale ones in Japan, Germany, Scandinavia, and the UK are expansions of urban areas and retrofits, which is an improvement in terms of development forms.

Implementations and outcomes of contemporary eco-cities

The implementations and outcomes of contemporary eco-cities at early stage of the development in the 1990s and early 2000s are an improvement upon early urban reformers in many aspects, while those at current development stage may vary in light of different contexts. It is doubtful that the prospects of these newly-built eco-cities might be an improvement.

At the early stage, contemporary eco-cities and eco-towns were mainly developed in Japan and central and northern Europe, exemplifying the eco-cities in Germany. During this stage, a distinctive characteristic of small-to-moderate-scale developments is that they gave equal weighting to the environmental, economic and social aspects of their design and integrated them well.

The Eco-Town Program was launched in Japan in 1997 through national initiatives and municipal redevelopment planning in order to deal with waste management issues, industrial pollution and to stimulate new industry development during a time of economic stagnation. The Eco-Town concept in Japan originally focused on Industrial Symbiosis—the iconic 3R application of Industrial Ecology which concerns ‘the productivity and environmental impacts of resources in industrial societies’ (McManus 2005). The theory then extended to Urban Symbiosis to become part of the Eco-City concept, focusing on overall urban planning and urban ecosystems, civil society and greening of cities. The implementation of this program and the successful development of 26 eco-towns, which focused mainly on circular economy and environmentally conscious planning, highlight three major aspects of improvement on earlier urban reformers. First, the government put in place a comprehensive legal framework for becoming a recycling-based society. This action provides a legislative foundation, imposes a sense of duty, and reduces the risk for further transition and development in industries and society, which is a practical improvement on the ideas of earlier reformers which were often hard to carry out (in a large scale) without legislative basis. Second, the program emphasised a combination of initiatives from public sector, business sector and, especially, civil society. There were a number of citizen activities emerging during that period and engaging in the program, which stands in contrast to places like Saltaire that were largely paternal. Third, the program was carried out mainly in the form of redevelopment and retrofitting projects, which faced less trouble than those early built-from-scratch ideas because it could take advantages of existing social capital.

There are also two smaller-scale eco-town examples in Freiburg, Germany: Vauban and Rieselfeld. These two district-scale projects are expansion of the city in response to the increasing population. They were carefully planned and developed with foci on public transport, dense but diverse housing, and environmental buildings. A major improvement on earlier works includes the emphasis on public transport, as these two districts were designed to minimise car dependency, given that 35% of residents in Vauban have abstained from driving (Beatley 2012). Le Corbusier’s fetishisation of the automobile in his La Ville Radieuse (1967) is of little relevance considering today’s traffic congestion, the high energy consumption, and air quality concerns associated with automobiles. These two examples successfully demonstrate how public transport can contribute to urban and environmental sustainability.

At the current stage (since mid-2000s), contemporary eco-cities are primarily proposed to be built from scratch, on a grand scale. They would represent highly sustainable, experimental flagships, most of which are promoted by the Chinese national and municipal governments including examples such as the Masdar City in the UAE. The characteristic of these projects is they are predominantly entrepreneurial cities and tend to prioritise economic concerns over environmental ones. In terms of implementation and outcomes, these new cities sometimes fail to demonstrate improvements on previous works, with uncertainty remaining considering that many of them are still under construction.

Due to rapid population influx, emissions pressure and pollution issues, the Chinese state government encourages local governments to experiment the ‘eco-city’ as a flagship project for new technologies and ‘sustainable’ economic and ecological urban development. Highly publicised projects include Dongtan Eco-City on Chongming Island, Shanghai, Sino-Singapore Tianjin Eco-city and Caofeidian Eco-City. The first project, Dongtan Eco-City, has already been stalled due to issues of land quotas. The subsequent Caofeidian Eco-City, claiming to be a renewable energy city, has only completed a few buildings and failed to attract residents, suffering from huge debt and being indefinitely postponed. The only project that has come close to completion is the Sino-Singapore Tianjin Eco-city, which has not been developed without issues and is still of little improvement upon earlier urban reformers. It is being developed according to 26 sustainability indicators (Fig. 1), demonstrating the progressive practices within the Chinese context, although many of them are taken for granted in the West. There are many problems in the construction. The city claimed to promote green transport but the implementation of numerous highways are still the dominant transport structure in the city, accompanied by high-rise residential buildings, which is a strong resemblance of the type of city Le Corbusier imagined for ‘the contemporary city’. There is even less open green space compared with Le Corbusier’s ideas. Additionally, this city has only managed to attract 6000 residents thus far—far less than its objective. There is also hardly any inclusion of social sustainability, where there should be relevant viable attempts, as Caprotti states (2015), ‘the view of sustainability which is concerned purely with a city’s environmental footprint, or with its economic success is severely limited’.

26 Key performance indicators to measure success (Sino-Singapore Tianjin Eco-City Investment and Development Co.,Ltd 2015)
In the case of Masdar City, the so-called ‘carbon-neutral city’ (recently modified to be ‘low- carbon city’) initiatives are just a part in making Abu Dhabi a leader in the industry of renewable energy technologies, which does not include many real actions with reconciliation of the nature-human relationship. An exception to the Asian ambitions of the eco-city is the Eco-town plan in the UK, which may constitute an improvement on earlier works. Those eco-towns proposed by the North West Bicester, which are an expansion of existing urban areas, aim for affordable housing and promoting social justice.

In conclusion, the contemporary eco-cities, from the early emerging stage of the1990s till today, are in general, an improvement upon the work of earlier urban reformers in terms of their ideas and planning. Whilst the early-stage developments such as Eco-towns in Japan demonstrate an improvement in terms of practical implementations and existing outcomes, brand-new built-from-scratch eco-cities may not be sustainable in reality in light of different contexts.  They tend to prioritise economic goals instead of environmental concerns. Whether these newly built eco-cities will be an improvement on those of earlier reformers remains uncertain. The developments which have begun, however, provide lessons for future urban developments which can be introduced to improve future designs and the redevelopment of existing cities.

Blog by Cabot Institute Masters Research Fellow Shiyao (Silvia) Liu.

Further reading

Beatley, T., 2012. Green cities of Europe: global lessons on green urbanism, Washington DC, Island Press.
Caprotti, F., 2015, Eco-Cities and the Transition to Low Carbon Economies, Palgrave.
Howard, E., 1902, Garden Cities of Tomorrow, London: Swan Sonnenschein.
Le Corbusier, 1967, The Radiant City (La Ville Radieuse): Elements of a doctrine of urbanism to be used as the basis of our machine-age civilization, New York, The Orion Press.
McManus, P., 2005, Vortex Cities to Sustainable Cities: Australia’s Urban Challenge, UNSW Press, Sydney.
Register, R., 1987, Ecocity Berkeley: Building Cities for a Healthy Future. Berkeley, CA: North Atlantic Books.
Richardson, B.W., 1876, Hygeia, A city of health. MacMillan & Co., London.
Sino-Singapore Tianjin Eco-City Investment and Development Co.,Ltd. 2015, Tianjin Eco City website,

Wednesday, 8 February 2017

Model uncertainties in multispecies ecological models

We live in an increasingly uncertain world.  Therefore, when we model environmental processes of interest, it is vital to account for the inherent uncertainties in our analyses and ensure that this information is communicated to relevant parties.  Whilst the use of complex statistical models to estimate quantities of interest is becoming increasingly common in environmental sciences, one aspect of uncertainty that is frequently overlooked is that of model uncertainty.  Much of the research I conduct considers this additional aspect of uncertainty quantification; that is not just uncertainty in the quantities of interest, but also in the models that we use to estimate them.

An example of this is in a paper recently published in Ecology and Evolution (Swallow et al., 2016), which looks at how different species of birds that we commonly see in our gardens respond to the same environmental factors (or covariates).  Some of the species have declined rapidly over the past 40 years, whilst others have remained stable or even increased in number.  Possible drivers of these changes that have been suggested include increases in predators, changes in climate and availability of natural food sources.  Statistically speaking, we try to understand and quantify changes in observed numbers of birds by relating them to changes in measured environmental quantities that the birds will be subjected to, such as numbers of predators, weather variables, habitat quality etc.  Most previous analyses have modelled each of the species observed at many different geographical locations (or monitoring sites) independently of each other, and estimated the quantities of interest completely separately, despite the fact that all these species share the same environment and are subject to the same external influences.  So how do we go about accounting for the fact that similar species may share similar population drivers?

This essentially constitutes a model uncertainty problem – that is, which parameters should be shared across which species in our statistical model and which parameters should be distinct?

If we were to consider two different species and use two different environmental factors to explain changes in those species, say habitat type and average monthly temperature, there are four possible models to consider.  That is,

Habitat type
No parameters

This can easily be extended to a higher number of species and covariates.

There is also inevitably going to be some aspects of variability shown by some of the species that we cannot account for through the quantities we have measured.  We account for this using site-specific random effects, which explain variability that is linked to a specific monitoring site, but which is not accounted for by the environmental covariates in the model.  Again, we would usually assume this is a single quantity representing the discrepancy between what we have accounted for using our measured covariates and what is ‘left over’.  Following on from work of previous authors (Lahoz-Monfort et al., 2011), we again split this unexplained variation into two – unexplained variation that is common to all species and unexplained variation that is specific to a single species.  The ratio of these two quantities can give us a good idea of what measurements we may be missing.  Is it additional environmental factors that are wide-ranging in their effects or is it something relating to the specific ecology of an individual species?

In the paper, we apply our method to a large dataset spanning nearly 40 years, collected as part of the British Trust for Ornithology’s Garden Bird Feeding Survey.  We selected two groups of similar species commonly found in UK gardens during the winter.  For ecological reasons, we would expect the species within the two groups to show similar traits, so they act as ideal study species for detecting synchrony in responses to environmental factors.  Whilst most the results were consistent with those from single-species models (e.g. Swallow et al., 2015), studying the species at an ecosystem level also highlighted some additional relationships that it would be impossible to study under more simplistic models.  The results highlight that there is unsurprisingly a large degree of synchrony across many of these species, and that they share many of the traits and drivers of population change.  The synchronies observed in the results corresponded to both significant positive or negative relationships with covariates, as well as those species that collectively show no strong relationship with a given environmental factor.  There is, however, more to the story and some of the species showed strong differences in how they respond to external factors.  Highlighting these differences may offer important information on how best to halt or reverse population declines.

The results from our analyses showed the importance of considering model uncertainty in statistical analyses of this type, and that by incorporating relevant uncertainties, we can improve our understanding of the environmental processes of interest.  Incorporating more data into the analysis will help in further constraining common or shared parameters and reduce uncertainties in them.  It also allows us to guide and improve future data collection procedures if we can gain a better understanding of what is currently missing from our model.

Blog written by Dr Ben Swallow, a Postdoctoral Research Associate, studying Ecological and environmental statistics in the School of Chemistry.


Lahoz-Monfort, J. J., Morgan, B. J. T., Harris, M. P., Wanless, S., & Freeman, S. N. (2011). A capture-recapture model for exploring multi-species synchrony in survival. Methods in Ecology and Evolution, 2(1), 116–124.

Swallow, B., Buckland, S. T., King, R. and Toms, M. P. (2015). Bayesian hierarchical modelling of continuous non-negative longitudinal data with a spike at zero: An application to a study of birds visiting gardens in winter. Biometrical Journal, 58(2), 357–371

Swallow, B., King, R., Buckland, S. T. and Toms, M. P. (2016). Identifying multispecies synchrony in response to environmental covariates. Ecology and Evolution, 6(23), 8515–8525

Figure 1. Blue tits show a highly synchronous response with great tits, and to a lesser degree coal tits, to their surrounding environment.

Figure 2. Male house sparrow feeding on fat balls.  Whilst they show some synchrony in their response to environmental factors, they appear to be subject to a differing ecology to the other two species they were compared with.

Challenges of generating solar power in the Atacama Desert

My name is Jack Atkinson-Willes and I am a recent graduate from the University of Bristol’s Engineering Design course. In 2016 I was given the unique opportunity to work in Chile with the renewable energy consultancy 350renewables on a Solar PV research project. In this blog I am going to discuss how this came about and share some of the experiences I have had since arriving!

First of all, how did this come about? Due to the uniquely flexible nature of the Engineering Design course I was able to develop my understanding of the renewable energy industry, a sector I had always had a keen interest in, by selecting modules that related to this topic and furthering this through industry work experience. In 2013, the university helped me secure a 12 month placement with Atkins Energy based near Filton, and while this largely centred around the nuclear industry it was an excellent introduction into how an engineering consultancy works and what goes into development of a utility-scale energy project.

In 2015 I built on this experience with a 3 month placement as a research assistant in Swansea University’s Marine Energy Research Group (MERG). I spent this time working on the EU-funded MARIBE project, which aimed to bring down the costs of emerging offshore industries (such as tidal and wave power) by combining them with established industries (such as shipping). This built on the experience I had gained through research projects I had done as part of my course at Bristol, and allowed me to familiarise myself further with renewable energy technology.

Keen to use my first years after graduation to learn other languages and travel, but also start building a career in renewables, I realised that the best way to combine the two was to start looking at countries overseas that had the greatest renewable energy potential. Given that I had just started taking an open unit in Spanish, Latin America was, naturally, the first place I looked; and I quickly found that I needn’t look much further! Latin American was the fastest growing region in the world for renewable energy in 2015, and this was during a year when global investment in renewables soared to record levels, adding an extra 147GW of capacity. (That’s more than double the UK demand!)

So, eager to find out more about the opportunities to work there, I discussed my interest with Dr. Paul Harper. He very kindly put me in touch with Patricia Darez, general manager of 350renewables, a renewable energy consultancy based in Santiago. As luck would have it, they were looking to expand their new business and take someone on for an upcoming research project. Given my previous experience in both an engineering consultancy and research projects I was fortunate enough to be offered a chance to join them out in Chile. Of course I jumped at the opportunity!

1 - Santiago, Chile (the smog in this photo being at an unusually low level)
Fast-forward by 8 months and I am tentatively stepping off the plane into a new country and a new life, eager to get started with my new job. Santiago was certainly a big change to Bristol, being about 10 times the size, but to wake up every day with the Andes mountains looming over the skyline was simply incredible. The greatest personal challenge by far has been learning Spanish, largely because the Chilean version of Spanish is the approximate equivalent to a thick Glaswegian accent in English. So for my (at best) GSCE level Spanish it was quite a while before I felt I could converse with any of the locals (and even now I spend almost all my time nodding and smiling politely whilst my mind tries to rapidly think of a response that would allow the conversation to continue without the other person realising I haven’t a clue what they’re saying!) But it has taught me to be patient with my progress, and little by little I can see myself improving.

Fortunately for me though, I was able to work in English, and before long I was getting to grips with the research project that I had travelled all this way for! But before I go into the details of the project, first a little background on why Chile has been such a success story for Solar.

2 - There's a lot of empty space in the Atacama

The Atacama desert ranges from the pacific ocean to the high plains of the Andes, reaching heights of more than 6000m in places. It is the driest location on the planet (outside of the poles) where in some places there hasn’t been a single drop of rain since records began. This combined with the high altitude results in an unparalleled solar resource that often exceeds 2800 kWh/m2 (Below are two maps comparing South America to the UK, and one can see that even the places of highest solar insolation in the UK wouldn’t even appear on the scale for South America!)

3 - Two maps comparing the solar resource of Latin America to the UK. If you think about the number of solar parks in the UK that exist, and are profitable, just imagine the potential in Latin America!
The majority of the Atacama lies within Chile’s northern regions, and because of this there has been a huge rush over the past 3 years to install utility-scale Solar PV projects there. Additionally, Chile has seen an unprecedented period of economic growth and political stability since the 1990’s, in part due to the very same Atacama regions which are mineral-rich. The mines used to extract this wealth are energy-hungry, and as Chile has a lack of natural fossil fuel resources, making use of the plentiful solar resource beating down on the desert planes surrounding these remote sites made perfect economic sense. This is added to the need for energy in the rapidly-growing cities further south, in particular the capital Santiago, where almost a third of the Chilean population live. From 2010 to 2015, the total installed capacity of PV worldwide went from 40GW to 227GW, a rapid increase largely due to decreasing PV module manufacture costs. As the cost of installation dropped, investors began to search for locations with the greatest resources, and so Chile became a natural place to invest for energy developers.

However, as large scale projects began generating power, new challenges began to emerge. New plants were underperforming and thus not taking full advantage of the powerful solar resource. This underperformance could be down to a whole range of factors; faulty installation, PV panels experiencing a drop in performance due to the extremely high UV radiation (known as degradation). But the main culprits are likely to be two factors; curtailment and soiling.

Firstly, curtailment. Chile is a deceptively large country, which from top to bottom is more than 4000km long (roughly the distance from London to Baghdad). Because of this, instead on having one large national grid, it is split into four smaller ones. The central grid (in blue, which is connected to the power-hungry capital of Santiago) offered a better price of energy than the northern grid (in green) supplying the more sparsely populated Atacama regions. This lead to a large number of plants being installed as far into the Atacama desert as possible, and therefore as far north as possible, whilst still being connected to the more profitable central grid.

4 - A map of the central (blue) and northern (green) grids in Northern Chile. Major PV plants are shown with red dots
This lead to a situation where the low number of cables and connections that existed connecting these areas with the cities further south suddenly became overloaded with huge quantities of power. When these cables reach capacity, the grid operators (CDEC-SIC -, with no-where to store this energy, simply have no other option but to limit (or curtail) the clean, emission-free energy coming out of these PV plants. This is bad news for the plant operators as it limits their income, and bad news for the environment as fossil fuels still need to be burned further south to make up for the energy lost.

The solution to this is to simply build more cables, a task easier said than done in a country of this size and in an area so hostile. This takes a long time, and so until the start of 2018 when a new connection between the northern and central grids will be made, operators have little choice but to busy themselves by improving plant performance as much as possible in preparation for a time when generation is once again unlimited.

This leads me onto soiling. Soiling is a phenomenon that occurs when wind kicks up sand and dust from the surrounding environment and this lands on the PV panels. This may seem relatively harmless but in Saudi Arabia is has been found to be responsible for as much as a 30% loss in plant performance. Chile, however, is still a very new market and so the effects of soiling here are not as well understood. What we do know for sure is that it affects some sites much more than others – the image below being taken by the 350renewables team at an existing Chilean site.

5 - The extent of soiling in the Atacama. One can appreaciate the need for an occasional clean!
These panels can be cleaned, but this becomes somewhat more complicated when you consider that some of these plants have more than 200,000 panels on one site. Cleaning then becomes a balance between the cost of cleaning, the means of cleaning (water being a scarce commodity in the desert) and the added energy that will be gained by removing the effects of soiling.

This is what the research project that I am taking part in hopes to establish. Sponsored by CORFO, a government corporation that promotes economic growth in Chile, and working with the University of Santiago, 350renewables hopes to establish how soiling effects vary across the Atacama and which cleaning schedules are best suited to maximising generation. There are 10 utility scale projects currently taking part, providing generation data and cleaning schedules. My role within this project has thus far been to inspect, clean and process all the incoming data and transfer this to our in-house tools for analysis. In the future (as my spanish improves) this will move onto liaising with the individual maintenance teams at each site to ensure that cleaning schedules are adhered to.

My most notable challenge thus far was presenting some of our initial findings at the Solar Asset Management Latin America (SAM LATAM - conference in September. Considering I had only been in the country for just over a month, it was a lot to learn in not very much time! My presentation discussed the underperformance of Chilean PV plants and the potential causes for this, examining some of the publicly available generation data over the past few years. It was certainly terrifying, but getting the opportunity to share a stage with a plethora of CEOs, managers and directors from the Chilean solar energy industry was a fantastic opportunity.
6 - I felt like an impostor amongst all the Directors and Managers
A few weeks prior to this we had also gone to the Intersolar South America conference ( in São Paulo, Brasil, where Patricia was speaking. This was another fantastic opportunity to meet other people from the industry (although somewhat limited by my non-existent Portuguese abilities) and I was lucky enough to have some time to explore the city for a few days thereafter.

7 - São Paulo, Brazil
In addition to São Paulo, I have been able to find the time to travel elsewhere in Chile during my time here, including down to Puerto Varas in the south with its peaceful lakes nestled at the feet of imposing active volcanoes (including the Calbuco volcano, which erupted in spectacular fashion in early 2015: Being further south the countryside is much more green, and with a significant German influence from several waves of immigration in the 1800s.

8 - Me in front of the incredible Osorno Volcano near Puerto Varas
9 - Puerto Varas

By far my favourite though was the astounding Atacama desert. As beautiful as it is vast. The high altitude making for astounding blue skies contrast against the red rocks of the surrounding volcanic plains. It is also one of the best stargazing spots on the planet, and the location of the famous A.L.M.A. observatory, which hopes to provide insight on star birth during the early universe and detailed images of local star and planet formations (

10 - Me in the Atacama. The second photo being what can only be described as a dust tornado. To call the Atacama inhospitable would be taking it lightly
11 -Valle de la luna, an incredible formation of jagged peaks jutting out of the desert plains. Certainly a highlight.
In the new year the soiling project really gets underway, and by the end of 2017 we hope to have some findings that will provide some insights into the phenomenon that is soiling. Personally, it has been a great adventure so far, the language skills I have developed and the experience of living in another culture, as opposed to merely passing through as a tourist, has been very rewarding. I still have a long way to go, and hope to post an update to this blog in the future, but for now a Happy New Year from Chile!

Wednesday, 1 February 2017

The 95th percentile

The 95th percentile is a way of describing, in a single value, a surprisingly large outcome for any quantity which can vary.  As in ‘surprisingly large, but not astonishingly large’.

For example, heights vary across people.  Consider adult UK women, who have a mean height of about 5’4’’ with a standard deviation of about 3’’. A woman who is 5’7’’ inches would be tall, and one who is 5’9’’ would be surprisingly tall.  5’9’’ is the 95th percentile for adult UK women.  The thought experiment involves lining every adult UK woman up by height, from shortest to tallest, and walking along the line until you have passed 95% of all women, and then stopping.  The height of the woman you are standing in front of is the 95th percentile of heights for adult UK women.

The formal definition of the 95th percentile is in terms of a probability distribution.  Probabilities describe beliefs about uncertain quantities.  It is a very deep question about what they represent, which I will not get into!  I recommend Ian Hacking, ‘An introduction to probability and inductive logic’ (CUP, 2001), if you would like to know more.  If H represents the height of someone selected at random from the population of adult UK women, then H is uncertain, and the 95th percentile of H is 5’9’’.  Lest you think this is obvious and contradicts my point about probabilities being mysterious, let me point out the difficulty of defining the notion ‘selected at random’ without reference to probability, which would be tautological.

So the formal interpretation of the 95th percentile is only accessible after a philosophical discussion about what a probability distribution represents.  In many contexts the philosophy does not really matter, because the 95th percentile is not really a precise quantity, but a conventional label representing the qualitative property ‘surprisingly large, but not astonishingly large’.  If someone is insisting that only the 95th percentile will do, then they are advertising their willingness to have a long discussion about philosophy.

Blog post by Prof. Jonathan Rougier, Professor of Statistical Science.
First blog in the series here.
Second blog in series here.