Monitoring Coastal Changes in Saltmarshes: Maps or Aerial Photography?

A key aspect of the recent changes in planning legislation is that development should be preferred provided it is sustainable. On coastal margins there is an additional issue of the potential impact of sea-level change on increased development. Development behind existing coastal defences, both human and natural, seems to be encouraged within the new planning legislation as it makes use of investment already sunk into defence and implies that any future sea-level rise will be accompanied by increasing investment these defences.

Given the importance of these defences it is essential to know how the coastline has responded in the past. The rate of past change is an important indicator as to how dynamic the coast is and how it is likely to respond to increasing sea-level. A long-term, over 50 or 100 or more years, view of rates of change often makes use of historic maps to establish baselines from which change is measured. A recent paper by Brian Baily and myself ( I made the coffee again!) published in the Journal of Coastal Conservation looks at how maps have been used as sources of evidence of coastal change in the Solent (specifically Lymington, Beaulieu River, Calshot Spit, Eling, Portsmouth Harbour, Langstone Harbour and Pagham Harbour). The paper  is entitled Assessing historical saltmarsh change; an investigation into the reliability of historical saltmarsh mapping using contemporaneous aerial photography and cartographic data, a long title but a very accurate description of what the paper does. The paper assesses how these maps have been used to identify and quantify changes in saltmarsh, an important coastal ecosystem and a natural protective barrier. In current terms a key ecosystem service. Importantly, the locations and rates of change these maps suggest are compared to the rates of change that an analysis of aerial photography provides. 

The mapping of saltmarsh is full of problems that limit the reliability of the changes measured. Surveyors in the mid-nineteenth century, for example, did not have a clear and specific set of instructions about what to map above the low water mark. Inconsistencies in the accuracy and precision of saltmarsh identification and mapping are bound to arise when a surveyor was confronted with the practical and often hazardous task of trying to get into a saltmarsh and survey it. Similarly, this ecosystem was not viewed as a particularly valuable resource in the nineteenth and early twentieth centuries and so the incentive to get into the mud and accurately survey was not really there. In addition, the growth of Spartina spp. in this period would have made the identification and mapping of this ecosystem tricky at best in some locations. Often saltmarshes were only indicated by some symbol on a map covering a vaguely defined area of land – not the best baseline from which to accurately measure changes.

The aerial photography of the same areas provided a useful control from which to assess the accuracy of the location and rates of change in saltmarshes derived from maps. The aerial photography shows that large areas of saltmarsh were excluded from the OS maps – so major losses of a valuable coastal ecosystem can not be quantified. These areas seem to be the newer salt marshes and so areas that are likely to have provided coastal protection in the recent past as development has occurred in these coastal regions. Given that salt marshes can change in extent rapidly this suggests that analysis of rates of change in this important protective coastal ecosystem needs to be gauged against the accurate data provided by aerial photography which is only available from the early twentieth century onwards rather than from the potentially more inaccurate figures provided by historic mapping in the nineteenth century.

Understanding Daily Air Quality

Atmospheric pollution is a continuing environmental problem across the globe. Within the UK data on historic pollution levels as well as current pollution levels can be found at DEFRA Air Quality site. A great store of information and one that you can download data from.

Wonderful as this source is for research, atmospheric pollution is not a problem that has past or is under tight control on a global scale. The locations of UK data reflects the monitoring networks set up in the 1960s by Warren Springs Laboratory, largely in response to the Clean Air Act (1956) and the need to monitor levels of pollutants to ensure that standards were being meet. The early monitors tended to be instruments such as the sulphur dioxide bubbler (so old I couldn't fidn a photo of it on the Web!). Air was pumped into the machine at a known rate and the atmosphere reacted with the liquid as it bubbled through the machine. After day the flask with the liquid in was removed and replaced with another flask of liquid. The liquid from the previous day was analysed using titration techniques (reacting the sulphur dioxide with another chemical to get a colour change and a reaction product that could be accurately measured) to determine the levels of sulphur dioxide (once the various calibrations and calculations had been done). I know this because I used an old bubbler in my thesis to monitor sulphur dioxide levels on the roof of the Geography Department at UCL, London. It was educational, but was a pain to have to process daily, particularly as it was just self-taught me undertaking the titration much to the amusement of colleagues in the lab. Passive monitors such as nitration tubes (they just sit there and the pollutants react with them) were also used, but still needed chemical post-processing to obtain a result.

By the time I finished my thesis in 1989, real-time monitoring of pollutants, or at least hourly averaged and then 15-minute averaged values, were becoming more usual and replaced the daily averaged data. This is great for monitoring levels virtually continuously and for identifying specific pollution episodes but how much information is there and how can you interpret it? Air quality standards have varying monitoring levels for different pollutants and even the same pollutant can have different exceedence values. Sulphur dioxide levels in the UK, for example, should not exceed 266 mirocgrams/m3 more than 35 times per year if measured as averaged 15-minute concentrations. If measured as 1 hourly means then 350 micrograms/m3 should not be exceeded more than 24 times per year. If measured as a 24 hour average then 125 micrograms/m3 should not be exceeded more than 3 times a year. So the limits change with the monitoring period and the type of equipment being used to monitor pollution levels. This variation may begin to get confusing if you try to communicate it to too many different end-users.

A simplified version, the Daily Air Quality Index, recommended by the Committee on the Medical Effects of Air Pollutants (COMEAP) uses an index and banding system with bands numbered1-10 and colour coded for the level of severity of atmospheric pollution. The scale mirrors established ones for pollen and sunburn so draws on existing public understanding of colour and levels. Bands 1-3 are green, so represent low atmospheric pollution levels, 4-6 are shades of orange and represent moderate atmospheric pollution levels, 7-9 are darkening shades of red ending up at brown and represent high atmospheric pollution levels, whilst the oddly coloured purple band of 10 represents very high levels of atmospheric pollution. The index itself combines the highest concentrations for a site or region of five key pollutants: nitrogen dioxide, sulphur dioxide, ozone, PM2.5 and PM10.

The DAQI may be a useful tool for communicating information about the general level of pollution as it relates to human health but does the simplicity mask complexity that disaggregated data would not? The relative contribution of the five pollutants to the index can be gauged by the information on each at the DEFRA website. PM2.5 and PM10 uses 24-hour running mean concentrations and have specific threshold levels fro each band, whilst sulphur dioxide is measured as 15-minute averaged concentrations and, again, has threshold values for each band. The index itself, though, hides if all or just one or two of the pollutants push the DAQI into a band. The index misses other pollutants that could impact upon human health, even though these may be monitored such as benzene. The cocktail of pollutants used to create the index also reflects a specific context, the UK, would the cocktail of significant pollutants vary with other contexts? The cocktail and the monitoring intervals are not necessarily ‘natural’ ones – they have been developed from monitoring set up for other purposes such as regulatory requirements. The index is squeezed out of what exists.


The DAQI is a very, very useful tool, but it reflects an attempt to communicate complex and huge volumes of information in a simplified manner that, the makers believe, will be of use to specific end-users. Once data is compressed and simplified you are bound to loss some of the information contained in its variations and detail. The index you develop for targeted end-users will, of necessity, exclude a lot of the information you have collected and it is just useful, for the end-users in particular, to be aware of this.

Beijing Atmospheric Pollution now online

On 23rd January the Beijing Municipal Environmental Monitoring Centre began to release atmospheric pollution data online (see this site but a knowledge of Chinese helps in navigating and understanding the data http://translate.google.co.uk/translate?hl=en&sl=zh-CN&u=http://zx.bjmemc.com.cn/&ei=EBVjT5LAIKXS0QWE-JT1AQ&sa=X&oi=translate&ct=result&resnum=5&ved=0CH4Q7gEwBA&prev=/search%3Fq%3DBeijing%2BMunicipal%2BEnvironmental%2BMonitoring%2BCenter%26hl%3Den%26biw%3D1707%26bih%3D1121%26prmd%3Dimvns ).





The hourly data had previously only been available for laboratory use (http://www.china.org.cn/environment/2012-01/06/content_24337033.htm) but the release of the data seems to be a response to public concern over air quality and the mismatch between government statistics and public perception of air quality. Some of this perception may have resulted from the release of atmospheric pollution data by the US Embassy from a rooftop monitoring station(http://ecocentric.blogs.time.com/2012/01/22/political-pollution-how-bad-air-equals-social-unrest-in-china/ for report and http://www.nytimes.com/2011/12/07/world/asia/beijing-journal-anger-grows-over-air-pollution-in-china.html?_r=1 for a discussion of the halting of the tweets in July 2009 and http://twitter.com/#!/beijingair for the tweet). The mismatch between official announcements about good quality air and the tweet caused some friction between officials and the embassy.



So what are we to make of the release of this data? Firstly, it is handy to know Chinese to interpret the site but then again the site is not aimed at an English speaking foreigner but at the Chinese inhabitants of Beijing so this is a fairly lame criticism. Secondly, the data release may be a political decision but at least the data is out there and can now be assessed by the public and by other scientists around the world working on air pollution – surely a good thing in its own right. Thirdly, should the data be questioned? The US embassy site seems to have taken on the role of arbitrator in assessing the data quality (at least in Western press releases). The US embassy is just one site with monitoring equipment at a specific height (not necessarily standardized to the height of the Chinese monitoring stations) so any spatial variation in air quality would not be picked up by data from one site. Even asking the question about data reliability is political. It suggests that the Chinese data will somehow be affected by the political powers that be (as if the US act of monitoring pollution isn’t political as well?!) Details of where the monitoring sites are located, the accuracy and standardization of the monitoring equipment, etc are reasonable scientific questions to ask both of Chinese pollution data and the pollution data of any monitoring network wherever it is. Such questioning ensures comparability of datasets. By releasing the data the Chinese scientists and authorities are putting themselves within this scientific debate. Criticising a dataset does not mean the data set is wrong; questioning and clarification and refinement to ensure compatibility is merely part of the scientific process.