1

THE FUTURE TREND OF A MULTI-FUNCTIONAL AWS NETWORK IN AN URBAN SETTING LIKE HONG KONG

Edwin W.L. Ginn(1) and K.H. Tam(2)

(1) Hong Kong Observatory, eginn@hko.gov.hk, 134A Nathan Road, Kowloon, Hong Kong, China

(2) Hong Kong Observatory, khtam@hko.gov.hk, 134A Nathan Road, Kowloon, Hong Kong, China

Abstract Meteorological services face increasing demands from the local community and the global travelling public for more comprehensive weather information. This points towards the need for a multi-functional automatic weather station (AWS) network capable of delivering a great variety of weather and related information than before. With recent advances in instrumentation and increasing cost-effectiveness in telecommunication, this need can now be met. The use of renewable energy not only improves the reliability of operations, but also contributes to conserving the environment and projecting a positive image for the meteorological services. It is also possible to deploy a central station capable of real-time data quality control for early detection of faults and to facilitate immediate remedial actions. This would result in enhanced data quality and availability. To extend the weather monitoring coverage in a cost-effective manner, cooperation with non-meteorological organizations such as schools could also be explored. This paper presents the Hong Kong Observatory's strategy and experience in the development of a multi-functional AWS network to meet the diverse needs of the public and specialised users for weather information in an urban setting. Key words: AWS, multi-functional, urban, weather 1. Introduction In recent years, meteorological services are faced with increasing demands for more comprehensive weather information from various sectors of the local community and overseas visitors. In a high density urban setting like Hong Kong, there are additional challenges and constraints. The need for weather and related information from an Automatic Weather Station (AWS) network often exceed those provided by traditional AWSs. Furthermore, to cope with the fast pace of life in a city and the variety of communication devices that the public is equipped with, there are requirements for reliable high-resolution spatial weather information captured and delivered in a timely manner via a number of fixed and mobile computing devices. These demands not only pose a unique challenge but also provide an opportunity for meteorological services to reassert the value of their services as well as to raise their image.

2

As background information, Hong Kong is situated at the south-eastern tip of China. It has a total area of 1104 square kilometres. It covers Hong Kong Island, Kowloon Peninsula and the New Territories — the more rural section of Hong Kong, which also includes 262 outlying islands. Hong Kong's population was about 6.88 million in mid-2004. The population density was 6380 people per square kilometre. Hong Kong's climate is sub-tropical, tending towards the temperate for nearly half the year. Temperatures can drop below 10 degrees Celsius in winter and often exceed 32 degrees Celsius in summer. About 90 per cent of the annual rainfall of 2214 mm occurs between April and September. The AWS network of the Hong Kong Observatory started operation in the mid-1980s and currently consists of about 80 stations throughout the territory. 2. Demands of an AWS network in an urban setting The development of a multi-functional AWS network is mainly driven by the requirements from users. These users include: weather forecasters; other government departments in such fields as civil engineering, drainage services, recreation and leisure and the police; special clients such as transport and bridge operators, container terminals; organizers of theme parks and sports organizations; and last but not the least, members of the public and tourists. Understanding the needs and demands of these multifarious users is crucial to the successful design and implementation of an AWS network. 2.1 The requirements Apart from the conventional temperature, relative humidity, pressure and wind readings, requirements from the public have emerged, ranging from visibility, weather photos, weather stress, environmental radiation levels and ultra-violet index. Different users have different requirements but the following are some common characteristics and qualities they would like to have:

a. reliable and of high quality; b. timely dissemination – latest data available to users in minutes; c. available in different channels – including radio, TV, PCs, mobile phones and computing

devices; d. user friendly for easy digestion; e. location specific – high spatial resolution; f. time specific – high temporal resolution; and g. environmentally friendly.

3. Meeting user requirements The following strategy and solutions have been adopted by the Hong Kong Observatory:

a. explore the use of new sensors and innovative use of data processing; b. ensure reliable operation of AWSs with minimum down-time;

3

c. improve data quality and availability; d. improve data transmission and reliability; e. develop user friendly and packaged products for easy digestion by users; f. improve product dissemination in a variety of format and channels; g. use of green energy; and h. expand the AWS network coverage within the confines of available resources.

3.1 New sensors and data processing techniques In recent years, urban dwellers are increasingly aware of the effects of weather on their well being, such as ultra-violet radiation and heat stress they would experience when engaging in outdoor activities. Reduced visibility due to meteorological conditions as well as anthropogenic activities such as construction and industrial development is also of growing concern. Various new sensors have been incorporated into the AWS network to satisfy these needs. Examples include:

a. sensors for ultra-violet index (UV Index); b. weather photos for the public and forecasters; c. visibility meters; d. wet bulb globe temperature (WBGT) for heat stress monitoring; and e. sensors for ambient radiation level, in connection with the operation of nuclear power

plants some 50 km away. New data and image processing techniques, such as edge detection, are also being explored to allow automatic determination of visibility using real-time weather photos. 3.2 Reliable operation with minimum down-time To ensure reliable operations with minimum down-time, the Observatory’s approach is to install redundant sensors, improve power supply, improve lightning protection to equipment and to provide on-site data logging to ensure that all data including those for climatological purposes can eventually be recovered in the event of failures in telecommunication lines. To improve reliability of AWSs, the Observatory have obtained positive results in using sensors and equipment that operate on DC as far as possible and installing battery banks to isolate power surges from city mains electricity during thunderstorms. Adding surge filters also help to protect sensors, electronics and communication equipment from lightning damages, thus ensuring reliable operation. All the above measures, especially the installation of redundant sensors, entail additional cost and a balance has to be struck between cost effectiveness and reliability of operation.

4

3.3 Improved data quality and availability with real-time data quality assurance system AWS data is typically received by a central data acquisition system which passes on the data to downstream systems for processing and archival. A system can be developed to provide one-stop data quality assurance (QA) by carrying out quality control of the data received by the central data acquisition system in real time. Such a system has been developed and implemented at the HKO in 2005. The data flow and processing are summarised in Figure 1. The system is highly automatic. It serves to enhance data quality and at the same time reduce manual labour in quality control. Operation of the AWSs can also be monitored via a webpage which displays the status of the AWS network in real time (Figure 2). Through various real-time and non-real-time automatic data quality control processes, the system carries out quality assurance for each data received from the AWS by assigning a quality assurance flag to the data, filtering out erroneous data from the AWS data stream, and alerting maintenance staff to immediate remedial action via automatic emails (Figure 3). The advantage of this automatic alerting feature is that it enables early detection and diagnosis of faults, thus enhancing data quality and availability. Apart from monitoring the operation of the AWS, the QA flags also serve as an indication of quality, facilitating reference by users in future case studies and climatological research. Details of the QA algorithms are given in WMO 1996, Shaver 2000, Tam et al. 2004 and 2005. The system is also capable of imposing threshold on data and issuing automatic warnings to users or prompting forecasters to issue weather warnings. 3.4 Improved data transmission and reliability – redundant communication path and

higher bandwidth Besides traditional land communication lines and radio links, public mobile data communication is now readily available at reasonable cost and with good spatial coverage, thus providing flexibility in the deployment of AWSs. For key stations, a redundant communication path can be accomplished through the combined use of land lines, radio links and public mobile data communication. This not only improves data transmission and reliability, but the higher bandwidth now available from public mobile data services such as GPRS and 3G would also allow the transmission of more AWS data at higher temporal resolution and make possible the dissemination of graphics products of large file size such as weather photos. 3.5 User-friendly and packaged products to meet the demands of users To cater for the specific needs of various users, the Observatory has developed new, user friendly products of weather information which are packaged in an easily digestible way. One such example is the provision of real-time weather photos on the Observatory website which enables the public and tourists to assess the weather conditions at their travel locations in a direct and intuitive manner. On the website, animation of weather photos for the past few hours as well

5

as some notable weather events of the past are also shown, thus promoting public education on weather (see Figure 4). Traditional rainfall information is now provided in an analysed form to the public through the use of coloured isohyets so that the public can easily appreciate the situation, especially during rainstorms. Integrated wind/gust display panels (see Figure 5) have also been developed to facilitate weather forecasters to better appreciate the wind situation in Hong Kong and to issue gust warnings to users such as traffic controllers and container terminal operators in a timely fashion. 3.6 Improved product dissemination in a diversity of format With the increasing popularity of the Internet and the enhanced multi-media capability of mobile phones and computing devices, the Observatory has taken the opportunity to revolutionize its product dissemination means to take advantage of the latest technology. Weather information are now delivered in text, images, video clips and voice files on the HKO Internet website, and on mobile phones and PDAs through specially designed webpages, thus catering for the public as well as the under-privileged. 3.7 Use of renewable energy and improved power supply AWSs are often sited in remote areas with no city power supply. These AWSs should therefore be designed to operate with instruments which work on DC and with low power consumption. Normally, solar panels can be used to power the AWS. However, insufficient sunshine due to prolonged periods of overcast or rainy weather will occasionally result in inadequate power for the station. A hybrid of solar panels and wind-powered generators can be deployed. Experience indicates that the windy conditions that normally accompany overcast weather complement the sunshine shortage quite well. This way, the hybrid is capable of providing a continuous supply of renewable energy. Moreover, as AWSs with renewable energy do not run on city power, they are immune to power interruptions from city mains and not susceptible to damages caused by lightning surges through the power line, thus enhancing data availability and reducing maintenance effort. Down time due to failure of power supply has since been much reduced. The use of renewable energy also contributes to conserving the environment and projecting a positive image for the meteorological service. Figure 6 shows a well received press event organized by the Observatory in 2005 announcing the operation of an AWS which operates entirely on renewable energy. 3.8 Expanding the AWS coverage Schools and organizations in Hong Kong are increasing aware of the importance of weather and climate to their well-being. Through the use of robust and inexpensive sensor packages, the number of weather instruments and thus weather observations are rising steadily. It may thus be worthwhile to enlist the co-operation of schools and the local community to extend the

6

local weather coverage. While these instruments and location may not be fully compliant with WMO’s rigorous requirements for accuracy and exposure, they nevertheless still provide useful information to the community given the proper advice, training and assistance in their optimal siting, operation and maintenance. Plan is in hand to explore such a co-operation. Figure 7 illustrates the Observatory’s efforts in liaising with schools in expanding the AWS coverage in Hong Kong. 4. Discussions Over the past few years, the AWS network in Hong Kong has gone through extensive development and improvements as discussed above. The introduction of new weather services giving the public more weather information, in quantity, in type and in detail, has been received very positively. Feedback from the public and specialised users has enabled the HKO to make further improvements to its services. At the Observatory, efforts are on-going to improve the reliablility of AWSs within staff and funding constraints. Plans are also underway to add more functions to AWSs through the installation of new sensors such as WBGT for heat stress monitoring, UV monitors for protection against UV and to deploy more network cameras at strategic locations to provide real-time weather photos for the public and weather monitoring by forecasters. These new sensors offer the opportunity for more weather services and products. For instance, images from network cameras can be usefully deployed in monitoring atmospheric visibility and sea states, which have implications on the city’s aesthetic quality and marine safety respectively. Plan is also in hand to cooperate with local schools and organisations to set up community meteorological network so as to enhance the weather coverage in Hong Kong. References 1. Shafer, M.A., C.A. Fiebrich, D.S. Arndt, S.E. Fredrickson, and T.W. Hughes, 2000: Quality assurance procedures in the Oklahoma Mesonet, J. Atmos. Oceanic Technol., 17, 474-494. 2. Tam, K.H., W.M. Tse and W.S. Ip, 2004: Integrated Meteorological Data Quality Assurance System (in Chinese only), 18th Guangdong-Hong Kong-Macau Seminar on Hazardous Weather, Hong Kong, 16 -18 February 2004. 3. Tam, K.H., B.Y. Lee and K.W. Chan, 2005: New automatic weather station system in Hong Kong featuring one-stop quality assurance, internet technology and renewable energy, WMO Technical Conference on Meteorological and Environmental Instruments and Methods of Observation (TECO-2005), Bucharest, Romania, 4 -7 May 2005. 4. World Meteorological Organization, 1996: Guide to Meteorological Instruments and Methods of Observation, WMO-No.8, Geneva, pp II.1-10 to pp II 1-12.

Figure 1: Data flow and processing by the integrated AWS data quality assurance system of the Hong Kong

Observatory.

Figure 2: A typical display of real-time data quality monitoring of the Hong Kong Observatory AWS network.

7

(a)

(b)

Figure 3: A spike occurring in the wind record (a) was captured by the real-time data quality assurance

system and an automatic email (b) was generated by the system to alert maintenance staff for

follow-up actions.

8

Figure 4: The “weather photo” webpage on the Hong Kong Observatory website.

Figure 5: The maximum gust display panel for weather forecasters to monitor high winds in Hong Kong and to

issue gust warnings to special users.

9

Figure 6: A press event organized by the Hong Kong Observatory in 2005. The use of renewable energy on this

island AWS reflects an effort towards conserving the environment and projects a positive image for

the department.

Figure 7: Establishing cooperation with schools to expand the AWS network coverage: (a) a school AWS;

(b) a typical school; (c) prototype school AWS webpage; and (d) discussions with teachers on future cooperation.

10