Photochemistry and Photobiology, 2014, 90: 903–910
Autonomous Portable Solar Ultraviolet Spectroradiometer (APSUS) – a New CCD Spectrometer System for Localized, Real-Time Solar Ultraviolet (280–400 nm) Radiation Measurement Rebecca Hooke*, Andy Pearson and John O’Hagan Public Health England, Didcot, Oxfordshire, UK Received 23 August 2013, accepted 22 January 2014, DOI: 10.1111/php.12251
ABSTRACT
There exist a number of fixed-site ground-level solar UV monitoring networks around the world (8–11) including one in the UK which has been operated by Public Health England (and its predecessor organizations) since 1988 (12). These networks offer valuable data for long-term monitoring of UV levels and are an excellent resource for climate change investigations. However, this type of solar monitoring has some limitations with regard to benefitting public health in the immediate short term. Firstly, fixed-site networks generally cannot provide direct UV radiation measurements in particular locations where many people are outside being exposed to solar UV radiation. Location-specific UV level information is important since there is significant variation in UV levels over local areas (13). However, there are often great distances (>200 km) between measurement sites in fixed-site networks and these sites are not always situated in locations where many people are likely to be outside and exposed to solar UV radiation, such as city centers, parks, beaches etc., but instead are often in more rural areas. Information on solar UV levels for these particular locations cannot be provided by these networks apart from by inference or modeling. Secondly, most fixed-site networks use broadband solar UV detectors and cannot therefore provide detailed spectral UV information. Broadband detectors are relatively inexpensive and they are smaller and simpler than spectral UV detectors. However, broadband detectors are only designed to apply one weighting function to one spectral shape and since the solar UV spectral shape changes over the course of the day (13) broadband detectors have a significant degree of inaccuracy due to spectral mismatch (8). Spectral data, on the other hand, are far more versatile since they can be weighted according to any biological weighting function and therefore allow various biological effects of solar UV spectral irradiance for the same data to be compared. Although less common than broadband solar UV monitoring, spectral solar UV monitoring is carried out (14–16); spectral solar UV data has been gathered by Public Health England (PHE, formerly the Health Protection Agency and the National Radiological Protection Board) since 1995. Spectral solar UV monitoring is normally carried out using a laboratory-based spectroradiometer. These types of spectroradiometer provide reliable UV measurements. However, they take a significant amount of time to scan one wavelength at a time across the UV spectrum (circa 2 min for the PHE spectroradiometer). Since weather conditions can easily change over this timescale, the resulting UV spectrum gathered from one of these instruments may show distortion. CCD (charge-coupled device) spectrometers measure all
Terrestrial solar ultraviolet (UV) radiation has significant implications for human health and increasing levels are a key concern regarding the impact of climate change. Monitoring solar UV radiation at the earth’s surface is therefore of increasing importance. A new prototype portable CCD (charge-coupled device) spectrometer-based system has been developed that monitors UV radiation (280–400 nm) levels at the earth’s surface. It has the ability to deliver this information to the public in real time. Since the instrument can operate autonomously, it is called the Autonomous Portable Solar Ultraviolet Spectroradiometer (APSUS). This instrument incorporates an Ocean Optics QE65000 spectrometer which is contained within a robust environmental housing. The APSUS system can gather reliable solar UV spectral data from approximately April to October inclusive (depending on ambient temperature) in the UK. In this study the new APSUS unit and APSUS system are presented. Example solar UV spectra and diurnal UV Index values as measured by the APSUS system in London and Weymouth in the UK in summer 2012 are shown.
INTRODUCTION Solar ultraviolet (UV) radiation reaching the earth’s surface has many well-documented health effects. These include detrimental effects such as sunburn, cataract formation and skin cancer (1,2) as well as the important beneficial effect of the production of vitamin D which is known to be an essential requirement for good bone health (3). Monitoring solar UV radiation is of great value for public health purposes, both in the immediate short term and in the long term. UV radiation is invisible and cannot be felt (4) and UV radiation levels cannot always be discerned intuitively from weather conditions (5); monitoring of UV levels in the immediate short term and the provision of this information to the public in real time would enable people to make informed decisions as to whether to protect themselves from solar UV. Long-term UV monitoring is important for climate change research and the monitoring of long-term trends in UV levels (6,7). *Corresponding author email: [email protected] (Rebecca Hooke) © 2014 Crown copyright. Photochemistry and Photobiology © 2014 The American Society of Photobiology. This article is published with the permission of the Controller of HMSO and the Queen’s Printer for Scotland and Public Health England.
903
904
Rebecca Hooke et al.
wavelengths at once, thus avoiding distorted spectra. They are also less expensive and smaller than the laboratory-based spectroradiometers. However, for a long time it has been considered that this type of spectrometer is not suitable for solar UV monitoring (17). In this study, the development of a new prototype CCD spectrometer-based instrument and communications system designed to benefit people’s health in the short term by providing real-time UV level information to the public is presented. The new instrument is called the Autonomous Portable Solar Ultraviolet Spectroradiometer (APSUS) unit and its associated system is called the APSUS system. The APSUS unit utilizes a CCD spectrometer to measure terrestrial UV radiation (280–400 nm) spectrally. It is portable; it can be taken to particular locations where people are outside being exposed to solar radiation in order to monitor UV radiation levels there directly. The APSUS system can feed the UV data gathered by the APSUS unit back to the public in a simplified form in real time via both a standard web page and a Smartphone-optimized web page. Spectral UV data gathered by the APSUS unit are processed and simplified UV data are stored in a PHE database. To provide an example of the data that can be gathered by the APSUS system in pertinent locations, example spectral data and UV Index data from the first official APSUS unit deployment are shown. There are currently two APSUS units and these units were officially deployed for the first time over the course of the summer of 2012 in London and Weymouth, respectively, in the UK. These locations were chosen because they were the sites of numerous high profile sporting events and large numbers of people were therefore likely to be outside and exposed to solar UV radiation in these locations.
METHODS Development of the optical system. A comparison was made of all the potentially suitable CCD spectrometers that could be found on the market (in early 2010) and which could measure over the UV wavelength range of 200–400 nm. The characteristics of the different spectrometers were compared. Since the instrument was to be utilized for public health purposes, one of the primary considerations for the choice of spectrometer was its ability to measure the low irradiance of the shortest terrestrial UV wavelengths, since these wavelengths have the greatest biological effect (18). The QE65000 spectrometer from Ocean Optics (Dunedin, Florida, USA) was selected to be the spectrometer used in this new instrument. The QE65000 performed well in comparison to the other spectrometers under all the criteria except for stray light and the size and weight of the spectrometer. The size and weight were considered to be of secondary importance, pertaining only to the portability of the final instrument. Stray light, however, was a very important issue. An alternative spectrometer was strongly considered due to its improved stray light rejection, however, its performance under the other criteria was not as favorable as the QE65000. The QE65000 had a dynamic range that was an order of magnitude better, and the largest dynamic range of all surveyed spectrometers. The QE65000 also had the largest range of possible integration times (8 ms to 15 min) and furthermore the QE65000 was the only spectrometer we found available with thermoelectric cooling of the CCD array. This cooling enabled the QE65000 to have much greater signal-tonoise ratio and therefore provided it with improved sensitivity which would be a great advantage for measuring the shortest terrestrial UV wavelengths. The QE65000 spectrometer was considered on balance to be the best CCD array spectrometer available for our instrument. The QE65000 was configured specifically for the purpose of measuring solar UV radiation. It was fitted with a #H5 grating which provided the best efficiency for short wavelength UV detection. A slit of 50 lm was chosen, considered to be wide enough to gain sufficient throughput without losing required resolution. The QE65000 was fitted with a
OF1-U325C bandpass filter which approaches being a solar blind filter— having a sharp cut-off at 400 nm and thus filtering out almost all visible light; since the intensity of visible light is much greater than that of UV in the solar spectrum, preventing this radiation from entering the spectrometer would significantly reduce the amount of potential stray light inside the spectrometer. An integrating sphere was chosen as the input optics for the spectrometer since it was found to have better overall radiation throughput than a diffuser and this was especially the case at the shortest wavelengths of terrestrial UV radiation (between 290 and 330 nm). The selected integrating sphere was 1.5 inches in diameter. The integrating sphere was connected to the spectrometer by a short optical fiber. Bespoke housing, optimized for angular response, was made for the integrating sphere (see “APSUS unit angular response” section) and the entrance to the integrating sphere was covered by a fused silica quartz dome. Completion of the APSUS unit. A number of other components were chosen to be included in the instrument. Devices were selected to gather temperature, humidity, tilt and GPS (Global Positioning System) information. A GSM (Global System for Mobile Communications) antenna was also included in the instrument to allow it to communicate remotely. An embedded computer was incorporated to bring all these components together and to control the instrument, allowing it to function autonomously. Bespoke control software was written for the instrument and was hosted on the embedded computer. An outline of the functions of this control system is as follows. The control system contains a module that controls the data acquisition of the spectrometer. This module provides a number of options for the method of acquiring spectral measurements, 1 Allowing spectra to be obtained with fixed integration times or with automatically optimized integration times. 2 Allowing one measurement to be taken or two measurements to be taken and the two spectra spliced together at a particular wavelength. Parameters defining the way in which measurements should be taken and their frequency are provided in a config file. The control system carries out all instrument communications. It reads the spectral UV data and writes it to a raw textfile alongside the temperature, humidity, tilt and GPS information. It transmits these generated raw textfiles over the GSM network to a processing computer based at PHE in Chilton, UK. All the instrument components were housed inside a bespoke environmental housing. This housing was machined out of an aluminum alloy. It comprised a top section that was painted white and a base plate that was anodized black. The housing, with all the other selected components, was called the APSUS head unit. A mounting plate was fitted to the base of the APSUS head unit to enable it to be mounted on a tripod. The final mounted device was called the APSUS unit (see Figs. 1 and 2). Two APSUS units have been constructed. Overview of the APSUS system. A system of communications allowing data gathered by the APSUS units to be stored in a PHE database and disseminated to the public, called the APSUS system, was developed. Following is an outline of how the APSUS system works. The APSUS system comprises the APSUS units, a processing computer, a PHE database, a server and a number of software programs. It allows spectral solar UV data to be collected, saved into a PHE database, processed into a simplified form and disseminated to the public in real time by means of a standard web page and also a Smartphone-optimized web page. Spectral solar UV data are gathered by APSUS units and are transmitted over the GSM network to a processing computer at PHE, Chilton. The processing computer saves the spectral UV data into a PHE database and processes the spectral data to calculate the UV Index (4,6). A file containing UV Index data for the current day is then updated and uploaded to a server. A standard and Smartphone-optimized web page are then updated to display the new UV Index data. The web page allows daily UV Index data for different sites at which APSUS unit measurements have been taken around the UK to be observed—both current data and historic data. The Smartphone-optimized web page allows UV Index data from the APSUS units for the current day to be observed. The UV Index data are presented on a UV Index Dial—showing the current UV Index for current data and the maximum UV Index over the course of the day for historic data (on the web page only). When current UV Index data are displayed, one sentence of public health advice relating to the current UV Index is given below the UV Index Dial. The UV Index data are also presented in the form of a graph in order to show the variation in the
Irradiance (mWm-2nm-1)
Irradiance (mWm-2nm-1)
Photochemistry and Photobiology, 2014, 90
905
1000 800 600 400 200 0 280
300
320
300
320
340 Wavelength (nm)
360
380
400
360
380
400
1000 100 10 1 0.1 280
340
0.01 0.001 0.0001 Jobin Yvon
Wavelength (nm) APSUS 1 APSUS 2
Figure 3. APSUS units’ spectra in comparison to the spectroradiometer spectrum at 15:00 on the day they were calibrated. The top graph has a linear scale and the bottom graph has a log scale.
Figure 2. APSUS head unit.
UV Index over the course of the day for both current and historic data (web page only). APSUS unit irradiance calibration. The APSUS units have been calibrated, traceably to PTB (Physikalisch-Technische Bundesanstalt), using an ISA-Jobin Yvon-Spex D3180 Series scanning spectroradiometer. The APSUS units gathered spectral UV data from the sky at the same time that this spectroradiometer scanned the sky and a sensitivity function was calculated for each APSUS unit that would set the APSUS unit spectrum to equal the spectroradiometer spectrum. This method of calibration was used since it meant that the spectral shape of the calibration source would be the same as the measurement source and this was found to be effective at taking account of stray light within the APSUS unit. The calibration was performed near the summer solstice, at 12 noon on 22 July 2012, under stable weather conditions (variation in UV irradiance over 5 min as measured by PHE broadband detectors was
Autonomous Portable Solar Ultraviolet Spectroradiometer (APSUS) – a New CCD Spectrometer System for Localized, Real-Time Solar Ultraviolet (280–400 nm) Radiation Measurement Rebecca Hooke*, Andy Pearson and John O’Hagan Public Health England, Didcot, Oxfordshire, UK Received 23 August 2013, accepted 22 January 2014, DOI: 10.1111/php.12251
ABSTRACT
There exist a number of fixed-site ground-level solar UV monitoring networks around the world (8–11) including one in the UK which has been operated by Public Health England (and its predecessor organizations) since 1988 (12). These networks offer valuable data for long-term monitoring of UV levels and are an excellent resource for climate change investigations. However, this type of solar monitoring has some limitations with regard to benefitting public health in the immediate short term. Firstly, fixed-site networks generally cannot provide direct UV radiation measurements in particular locations where many people are outside being exposed to solar UV radiation. Location-specific UV level information is important since there is significant variation in UV levels over local areas (13). However, there are often great distances (>200 km) between measurement sites in fixed-site networks and these sites are not always situated in locations where many people are likely to be outside and exposed to solar UV radiation, such as city centers, parks, beaches etc., but instead are often in more rural areas. Information on solar UV levels for these particular locations cannot be provided by these networks apart from by inference or modeling. Secondly, most fixed-site networks use broadband solar UV detectors and cannot therefore provide detailed spectral UV information. Broadband detectors are relatively inexpensive and they are smaller and simpler than spectral UV detectors. However, broadband detectors are only designed to apply one weighting function to one spectral shape and since the solar UV spectral shape changes over the course of the day (13) broadband detectors have a significant degree of inaccuracy due to spectral mismatch (8). Spectral data, on the other hand, are far more versatile since they can be weighted according to any biological weighting function and therefore allow various biological effects of solar UV spectral irradiance for the same data to be compared. Although less common than broadband solar UV monitoring, spectral solar UV monitoring is carried out (14–16); spectral solar UV data has been gathered by Public Health England (PHE, formerly the Health Protection Agency and the National Radiological Protection Board) since 1995. Spectral solar UV monitoring is normally carried out using a laboratory-based spectroradiometer. These types of spectroradiometer provide reliable UV measurements. However, they take a significant amount of time to scan one wavelength at a time across the UV spectrum (circa 2 min for the PHE spectroradiometer). Since weather conditions can easily change over this timescale, the resulting UV spectrum gathered from one of these instruments may show distortion. CCD (charge-coupled device) spectrometers measure all
Terrestrial solar ultraviolet (UV) radiation has significant implications for human health and increasing levels are a key concern regarding the impact of climate change. Monitoring solar UV radiation at the earth’s surface is therefore of increasing importance. A new prototype portable CCD (charge-coupled device) spectrometer-based system has been developed that monitors UV radiation (280–400 nm) levels at the earth’s surface. It has the ability to deliver this information to the public in real time. Since the instrument can operate autonomously, it is called the Autonomous Portable Solar Ultraviolet Spectroradiometer (APSUS). This instrument incorporates an Ocean Optics QE65000 spectrometer which is contained within a robust environmental housing. The APSUS system can gather reliable solar UV spectral data from approximately April to October inclusive (depending on ambient temperature) in the UK. In this study the new APSUS unit and APSUS system are presented. Example solar UV spectra and diurnal UV Index values as measured by the APSUS system in London and Weymouth in the UK in summer 2012 are shown.
INTRODUCTION Solar ultraviolet (UV) radiation reaching the earth’s surface has many well-documented health effects. These include detrimental effects such as sunburn, cataract formation and skin cancer (1,2) as well as the important beneficial effect of the production of vitamin D which is known to be an essential requirement for good bone health (3). Monitoring solar UV radiation is of great value for public health purposes, both in the immediate short term and in the long term. UV radiation is invisible and cannot be felt (4) and UV radiation levels cannot always be discerned intuitively from weather conditions (5); monitoring of UV levels in the immediate short term and the provision of this information to the public in real time would enable people to make informed decisions as to whether to protect themselves from solar UV. Long-term UV monitoring is important for climate change research and the monitoring of long-term trends in UV levels (6,7). *Corresponding author email: [email protected] (Rebecca Hooke) © 2014 Crown copyright. Photochemistry and Photobiology © 2014 The American Society of Photobiology. This article is published with the permission of the Controller of HMSO and the Queen’s Printer for Scotland and Public Health England.
903
904
Rebecca Hooke et al.
wavelengths at once, thus avoiding distorted spectra. They are also less expensive and smaller than the laboratory-based spectroradiometers. However, for a long time it has been considered that this type of spectrometer is not suitable for solar UV monitoring (17). In this study, the development of a new prototype CCD spectrometer-based instrument and communications system designed to benefit people’s health in the short term by providing real-time UV level information to the public is presented. The new instrument is called the Autonomous Portable Solar Ultraviolet Spectroradiometer (APSUS) unit and its associated system is called the APSUS system. The APSUS unit utilizes a CCD spectrometer to measure terrestrial UV radiation (280–400 nm) spectrally. It is portable; it can be taken to particular locations where people are outside being exposed to solar radiation in order to monitor UV radiation levels there directly. The APSUS system can feed the UV data gathered by the APSUS unit back to the public in a simplified form in real time via both a standard web page and a Smartphone-optimized web page. Spectral UV data gathered by the APSUS unit are processed and simplified UV data are stored in a PHE database. To provide an example of the data that can be gathered by the APSUS system in pertinent locations, example spectral data and UV Index data from the first official APSUS unit deployment are shown. There are currently two APSUS units and these units were officially deployed for the first time over the course of the summer of 2012 in London and Weymouth, respectively, in the UK. These locations were chosen because they were the sites of numerous high profile sporting events and large numbers of people were therefore likely to be outside and exposed to solar UV radiation in these locations.
METHODS Development of the optical system. A comparison was made of all the potentially suitable CCD spectrometers that could be found on the market (in early 2010) and which could measure over the UV wavelength range of 200–400 nm. The characteristics of the different spectrometers were compared. Since the instrument was to be utilized for public health purposes, one of the primary considerations for the choice of spectrometer was its ability to measure the low irradiance of the shortest terrestrial UV wavelengths, since these wavelengths have the greatest biological effect (18). The QE65000 spectrometer from Ocean Optics (Dunedin, Florida, USA) was selected to be the spectrometer used in this new instrument. The QE65000 performed well in comparison to the other spectrometers under all the criteria except for stray light and the size and weight of the spectrometer. The size and weight were considered to be of secondary importance, pertaining only to the portability of the final instrument. Stray light, however, was a very important issue. An alternative spectrometer was strongly considered due to its improved stray light rejection, however, its performance under the other criteria was not as favorable as the QE65000. The QE65000 had a dynamic range that was an order of magnitude better, and the largest dynamic range of all surveyed spectrometers. The QE65000 also had the largest range of possible integration times (8 ms to 15 min) and furthermore the QE65000 was the only spectrometer we found available with thermoelectric cooling of the CCD array. This cooling enabled the QE65000 to have much greater signal-tonoise ratio and therefore provided it with improved sensitivity which would be a great advantage for measuring the shortest terrestrial UV wavelengths. The QE65000 spectrometer was considered on balance to be the best CCD array spectrometer available for our instrument. The QE65000 was configured specifically for the purpose of measuring solar UV radiation. It was fitted with a #H5 grating which provided the best efficiency for short wavelength UV detection. A slit of 50 lm was chosen, considered to be wide enough to gain sufficient throughput without losing required resolution. The QE65000 was fitted with a
OF1-U325C bandpass filter which approaches being a solar blind filter— having a sharp cut-off at 400 nm and thus filtering out almost all visible light; since the intensity of visible light is much greater than that of UV in the solar spectrum, preventing this radiation from entering the spectrometer would significantly reduce the amount of potential stray light inside the spectrometer. An integrating sphere was chosen as the input optics for the spectrometer since it was found to have better overall radiation throughput than a diffuser and this was especially the case at the shortest wavelengths of terrestrial UV radiation (between 290 and 330 nm). The selected integrating sphere was 1.5 inches in diameter. The integrating sphere was connected to the spectrometer by a short optical fiber. Bespoke housing, optimized for angular response, was made for the integrating sphere (see “APSUS unit angular response” section) and the entrance to the integrating sphere was covered by a fused silica quartz dome. Completion of the APSUS unit. A number of other components were chosen to be included in the instrument. Devices were selected to gather temperature, humidity, tilt and GPS (Global Positioning System) information. A GSM (Global System for Mobile Communications) antenna was also included in the instrument to allow it to communicate remotely. An embedded computer was incorporated to bring all these components together and to control the instrument, allowing it to function autonomously. Bespoke control software was written for the instrument and was hosted on the embedded computer. An outline of the functions of this control system is as follows. The control system contains a module that controls the data acquisition of the spectrometer. This module provides a number of options for the method of acquiring spectral measurements, 1 Allowing spectra to be obtained with fixed integration times or with automatically optimized integration times. 2 Allowing one measurement to be taken or two measurements to be taken and the two spectra spliced together at a particular wavelength. Parameters defining the way in which measurements should be taken and their frequency are provided in a config file. The control system carries out all instrument communications. It reads the spectral UV data and writes it to a raw textfile alongside the temperature, humidity, tilt and GPS information. It transmits these generated raw textfiles over the GSM network to a processing computer based at PHE in Chilton, UK. All the instrument components were housed inside a bespoke environmental housing. This housing was machined out of an aluminum alloy. It comprised a top section that was painted white and a base plate that was anodized black. The housing, with all the other selected components, was called the APSUS head unit. A mounting plate was fitted to the base of the APSUS head unit to enable it to be mounted on a tripod. The final mounted device was called the APSUS unit (see Figs. 1 and 2). Two APSUS units have been constructed. Overview of the APSUS system. A system of communications allowing data gathered by the APSUS units to be stored in a PHE database and disseminated to the public, called the APSUS system, was developed. Following is an outline of how the APSUS system works. The APSUS system comprises the APSUS units, a processing computer, a PHE database, a server and a number of software programs. It allows spectral solar UV data to be collected, saved into a PHE database, processed into a simplified form and disseminated to the public in real time by means of a standard web page and also a Smartphone-optimized web page. Spectral solar UV data are gathered by APSUS units and are transmitted over the GSM network to a processing computer at PHE, Chilton. The processing computer saves the spectral UV data into a PHE database and processes the spectral data to calculate the UV Index (4,6). A file containing UV Index data for the current day is then updated and uploaded to a server. A standard and Smartphone-optimized web page are then updated to display the new UV Index data. The web page allows daily UV Index data for different sites at which APSUS unit measurements have been taken around the UK to be observed—both current data and historic data. The Smartphone-optimized web page allows UV Index data from the APSUS units for the current day to be observed. The UV Index data are presented on a UV Index Dial—showing the current UV Index for current data and the maximum UV Index over the course of the day for historic data (on the web page only). When current UV Index data are displayed, one sentence of public health advice relating to the current UV Index is given below the UV Index Dial. The UV Index data are also presented in the form of a graph in order to show the variation in the
Irradiance (mWm-2nm-1)
Irradiance (mWm-2nm-1)
Photochemistry and Photobiology, 2014, 90
905
1000 800 600 400 200 0 280
300
320
300
320
340 Wavelength (nm)
360
380
400
360
380
400
1000 100 10 1 0.1 280
340
0.01 0.001 0.0001 Jobin Yvon
Wavelength (nm) APSUS 1 APSUS 2
Figure 3. APSUS units’ spectra in comparison to the spectroradiometer spectrum at 15:00 on the day they were calibrated. The top graph has a linear scale and the bottom graph has a log scale.
Figure 2. APSUS head unit.
UV Index over the course of the day for both current and historic data (web page only). APSUS unit irradiance calibration. The APSUS units have been calibrated, traceably to PTB (Physikalisch-Technische Bundesanstalt), using an ISA-Jobin Yvon-Spex D3180 Series scanning spectroradiometer. The APSUS units gathered spectral UV data from the sky at the same time that this spectroradiometer scanned the sky and a sensitivity function was calculated for each APSUS unit that would set the APSUS unit spectrum to equal the spectroradiometer spectrum. This method of calibration was used since it meant that the spectral shape of the calibration source would be the same as the measurement source and this was found to be effective at taking account of stray light within the APSUS unit. The calibration was performed near the summer solstice, at 12 noon on 22 July 2012, under stable weather conditions (variation in UV irradiance over 5 min as measured by PHE broadband detectors was
