UKAirQuality.net    Before and After    Rogues Gallery    Bad NO2    Free Health Check    Free Calibration Spreadsheet     Costs    Achievements    Profile    Geoff.Broughton@aqdm.co.uk

Geoff Broughton can help you collect and process your ambient air quality data. Most air quality data processing is routine. Scaling raw measurements into concentrations using the latest calibrations and eliminating bad data requires the right software or spreadsheet. This will produce a good dataset fit for your Review and Assessment. Problems may surface if you are brave enough to put your data on a website or show timeseries charts in your published reports. The data should advertise your work and not become a potential source of embarrassment! Our UK air quality data is shown on a real-time and historic air quality website http://www.UKAirQuality.net

A good looking final dataset is far more important than how well you ran the project!

Data management costs are small in comparison to the operation of your monitoring site. Keeping on top of data issues is no more expensive than allowing peformance to slide. Here are some good reasons to contact me.  I can process your data better, faster and cheaper. You can't afford to employ an expert full time in these difficult times. I've processed the data for 1,300 sites over the last 27 years at AEA, NETCEN and WSL. I've solved data problems that you may never experience or know how to diagnose. I can review your final data and show you what you need to correct.

Test yourself with the Before and After and the Rogues Gallery and see if you can do better. You need my help if you think nothing is wrong.

Here is a simple everyday problem. How would you process these NO2 concentrations? Delete the negatives? Replace the negatives with 0 ppb? Add 5 ppb to the NO2? Decrease the NO sensitivity by 2%? Replace the charcoal scrubber? Replace the three-way valve? Replace your ESU? Recertify your poisoned NO cylinder? Get this wrong and the problem is never fixed and your reports become visibly suspect. No data is better than wrong data.

How can the NO2 measurements go wrong? Click here is see a simple worked example and the dramatic effect of typical instrument drift.   A tiny 1% error can easily overestimate the NO2 concentrations by 10%! How many AQMAs have been declared or not declared based on suspect data? How often have the NO2 diffusion tube bias adjustment or the AQ models been doubted? You need an expert to check, select, smooth and apply the best calibration scalings to your measurements.

Day-to-day operations require you to identify the problems. Final data ratification requires you to understand the problems. Day-to-day data collection and fault identification can be learnt by a training course and some practice. You just need to call-out the engineer promptly when you see something has "gone wrong". But how do you process the data based on the engineer's technical report? You need to understand what went wrong to either ignore, delete or correct the data.

There are numerous ways that instruments and calibrations can produce poor quality data. Some are so rare that you are unlikely to ever experience the fault in your network. You may struggle to diagnose the new problem and make the right decision but your final data and statistics must look good. Breakdowns can produce high peaks that exceed the air quality limits and dramatically affect your reports. How do you know what is good or bad?

Why bother recording the instrument diagnostics?  The instrument diagnostics are often hard to interpret and even the engineers do not always understand the significance. Above is a typical calibration spreadsheet for a ML 9841B NOx instrument. The operator has made a comment and the Chassis temperature was hot on arrival. So what? The carbon scrubber on the NOx exhaust needs to be changed to prevent ozone leaking into the enclosure and, more importantly, damaging the pump valves. The service engineer will probably want to replace the scrubber now rather than wait for the pump to fail. The instrument chassis temperature was hot on arrival then cooled to normal while the enclosure door was open. Perhaps the ventilation grills need cleaning and the cooling fans need to be checked. NOx instruments can go crazy during hot summer afternoons and reducing the life some expensive components. Fix the ventilation now before too late.

I have decades of experience examining data from 450 operational sites and 1,000 closed sites. I can readily spot anomalies, diagnose the problem and make the appropriate correction without unnecessary data loss. Here are some typical corrections that make a huge difference to your data and annual statistics. Ramped (proportional) zero and sensitivity scalings. Rescaling ozone sensitivities after a photometer test. Correcting NO2 converter inefficiencies. Correcting for BAM and TEOM low flow rates. Correcting TEOM k0 anomalies. Rescaling poisoned calibration cylinders.

Applying the final polish to your data is not difficult or expensive if you know what can go wrong. I can help you produce high quality data and statistics in several ways. Processing the data you have collected. Calculating your statistics. Training you to spot unusual faults. Reviewing your processed data and suggesting a course of action. Showing you how to create a Ratification Production Line to meet deadlines. Creating a Quality Circle that pools your knowledge and drives forward improvements.

The cost is small in comparison to the operation of your monitoring site. You have spent an enormous amount of money on instruments, cylinders, cabins, site visits, calibrations, service contracts, spare parts and data collection. All this is wasted if you skimp on a few hundred pounds for the final expert data ratification and report dodgy data.

I'm always happy to discuss air quality data and provide free advice. Contact me Geoff.Broughton@aqdm.co.uk