IPWEA NSW Division Annual Conference 2005
Streetlight Assessment Program - GIS application to model streetlight compliance
Glen Moody,
GIS / Assets Administrator, Kogarah Municipal Council, NSW Australia
Paper Summary
The Streetlight Assessment Project is a geographical based computer program used to assist in the management of street lighting infrastructure. The project was developed in response to a broader review of street lighting conducted by the Southern Sydney Regional Organisation of Councils. The need for this project arose as a consequence of significant deficiencies in the local electricity distributor’s management of street lighting assets that were identified by councils in 2003.
The GIS application allows councils to readily evaluate the capabilities of their street lighting on residential roads, understand the potential improvements that proposed lighting changes should bring and identify areas with inadequate light. While the application does not undertake detailed lighting calculations, it does provide a useful indication of likely areas of compliance with AS1158 Lighting for Roads and Public Spaces.
Page 1
IPWEA NSW Division Annual Conference 2005
Introduction
The Streetlight Assessment Program is a geographical based computer program used to assist in the management of street lighting infrastructure. The project was developed in response to a broader review of street lighting conducted by the Southern Sydney Regional Organisation of Councils (SSROC). The need for this project arose as a consequence of significant deficiencies in the local electricity distributor’s management of street lighting assets that were identified by councils in 2003.
Historically, local councils decided when streets needed to be lit. Many of these lights were first installed decades ago before an accepted national standard existed. In recent decades, EnergyAustralia and its predecessors have made a number of lighting technology selection and installation decisions that amounted to a de facto lighting standard. This de facto standard (as illustrated in EnergyAustralia Network Standard NS0118) was not always well aligned with accepted practice under AS1158. The practices of the electricity distributor were not necessarily efficient in the consumption of energy or the distribution of lighting levels as required by the current Australian Standard.
The GIS application provides the ability to model lighting levels of particular light types and mounting heights across an entire street lighting network. This was achieved by combining spatial information from council's GIS databases, EnergyAustralia's infrastructure data and lighting manufacturers’ photometric data on luminaire performance. The project is believed to be the first of its kind, and allows variable input data to model multiple scenarios and ensure portability across council lighting networks.
The principle functions of the application are to assess current levels of compliance against standards, identify and prioritise the worst areas of non-compliance, identify excessive lighting levels and determine the potential benefits of switching to new luminaires.
Overall, it will enhance the ability of council to assess the operational efficiency of lighting networks and consequently provide financial, social and environmental benefits to councils and the community.
Background
It is estimated that 2.5 percent of the total Australian electricity production is expended on night-time outdoor lighting. About thirty percent of this lighting ends up in the night sky as light pollution. (Dudley, 2001).
Roadway lighting is one of the single largest energy expenditure items for councils. It is therefore essential that this large infrastructure asset is managed efficiently to meet social and environment needs whilst maintaining a safe roadway.
In 2002 a group of 17 councils in southern Sydney, including Kogarah Council, initiated a review of the street lighting network as presently managed by EnergyAustralia. The review identified certain deficiencies as well as opportunities for substantial improvement. The scope of this review was expanded and in 2003 consultants Next Energy were contracted by SSROC and the ‘Street Lighting Improvement Program’ (SLI Program) was established. This program now includes some 210,000 lights across 29 councils representing approximately 45% of all street lights in NSW.
Next Energy put forward an invitation to all member councils to research a methodology for modelling street lighting on a large scale. Kogarah Council accepted this offer and the potential for a GIS solution was realised.
Kogarah Council worked closely with Next Energy to ensure the project would meet the required specification. Next Energy provided all background information relating to street lighting including Australian Standards and lighting manufacturers’ product information including isolux diagrams. A number of joint field studies were conducted within the Kogarah municipality to validate assumptions and ensure the accuracy of the project’s outcomes.
Objectives
The project addresses four principle improvement objectives.
- Provide the ability to model variable street lighting options across an entire network.
- Allow councils to identify existing levels of non-compliance of street lighting against Australian Standard 1158.3.1 for residential roads.
- Provide the opportunity to improve the operational efficiency of lighting technology in existing networks.
- Facilitate the introduction of future lighting technology design concepts.
The flow on benefits have the potential to deliver; more appropriate lighting solutions, reduced light pollution, improved safety and reduced greenhouse gas emissions by ensuring best of breed solutions are implemented.
Methodology
The Australian Standard for roadway lighting (AS1158) specifies that the minimum maintained level of light for residential category streets (AS1158.3.1) is 0.07 lux as measured in the horizontal plane.
Initially it was thought that mapping the spread of light from each luminaire could be achieved through utilising the buffer function common to all GIS software, however the light from a streetlight does not spread equally in all directions but instead has a specific shape and orientation. Each luminaire type has it’s own theoretical spread of light which is represented as an isolux diagram. Isolux diagrams for various streetlights were obtained from the respective manufacturers and the 0.07lux and 0.035lux lines were drawn to scale as templates within the GIS (see Figure 1).
In order to map these objects relative to each luminaire a custom application was written. This was done in MapBasic, a BASIC-like programming environment for MapInfo Professional which enables the extension of geographic functionality, as well as the ability to automate repetitive operations.
Light Orientation
Two data sets were supplied from Energy Australia, the first being streetlight point locations with attribute information, the second was a drawing file (.DXF) showing the orientation of the light. A site survey in Kogarah found light location was reasonably accurate and was usually within 3m of actual location. Overlaying of aerial photography was also used to confirm the accuracy of pole location. The light orientation dataset however was found to have some significant errors.
To correct these inconsistencies, new orientation data was required. This was achieved using a separate automated process that was later incorporated into the application. Using light pole point location data it was assumed that the orientation of the luminaire was, in most cases, perpendicular to the road centreline. The nearest road to each luminaire was identified and a line drawn perpendicular to this from each light pole.
The angle of this line could then be obtained by the program and used to map the template isolux shapes with the correct orientation.
Compliance
The minimal level of light intensity defined by the Australian Standard for Roadway lighting for residential roads is 0.07 lux. It is important to note that this level of light can be achieved through accumulation of overlapping light spreads. Hence the program was used to generate two geographical objects around each luminaire. These represented the area of light around the luminaire above 0.035lux and an inner object above 0.07lux shown in figure 1 below.
Figure 1. - spread of light showing 0.07lux (compliant) and 0.035lux
It was then possible to develop the program to include areas where the accumulated effect of two lights contributing 0.035 lux resulted in compliance with the standard.
From here the GIS could be used to map the spread of light from each luminaire and intersect this with the road reserve to determine the theoretical level of light reaching the surface of the road reserve throughout the entire road network.
Assumptions and Limitations
As with all GIS applications, the value of the information it produces is related to the quality of the data provided. In this project some assumptions were made, most of which were considered to have been conservative in nature. The most significant are as follows:
a) Actual Height Data Unavailable - The spread of light or isolux are theoretical spreads of lights based on luminaires positioned six metres above a flat road surface. The height of the luminaire affects the spread of light; a higher luminaire will disperse light further. To be conservative, a worst-case assumption was made that the average height of most luminaires was six metres which was the historic mounting height in Sydney.
b) Luminaire Orientation Data Inaccurate - As previously stated, orientation data for luminaires proved to be highly inaccurate. This data was discarded and luminaires were aligned with the road reserve. While this approach appears to be more accurate, councils may need to do further validation of this, particularly with respect to corners and a variety of roadway anomalies.
c) Luminaire Location Information Accuracy Needs Further Assessment – To date, only a small sample of luminaire location information has been assessed for accuracy. Councils will need to do further validation of the accuracy of this data and consider the implications of data variability.
d) Real World Variability - The spread of light assumes controlled environmental variables, and does not consider variations in slope, reflectance, or obstacles such as parked cars and trees. Taking these variables into consideration would be practically impossible to do so in a GIS. However, the Street Lighting Improvement Program noted that the Australian Standard AS1158, being a modelled standard, does not account for these variables either. In practice, where such factors produce localised difficulties, councils would have to rely on visual assessment. More often than not, such investigations will be in response to resident complaints.
Results
Map layers of areas 0.07lux or greater were created for the entire Kogarah streetlight network. The 0.07lux and greater layer was cut against the road reserve layer to produce areas of non-compliance. Each area of non-compliance retained the data of the road reserve layer allowing easy identification and reporting.
Some areas of non-compliance were identified as too small to be practical and were removed from the data (<1m2). It was also found that the non-compliance areas between two lights were often represented by several smaller areas therefore mis-representing the number of non-compliance areas. These areas were combined to represent a single non-compliance area. Although this was not essential it allowed better reporting on the location and number of additional street lights required to meet the current standard.
The above process was repeated for each of the three different light types; the current twin 20 Watt fluorescent, and the two newer technology luminaires. A non-compliance report was produced for each of these light types.
Figures 2 and 3 show how areas of non-compliance can be easily mapped across a local government area. Non compliant areas can be sorted by size to allow a prioritised investigation schedule. Simple non-compliance issues can be investigated in- house using the GIS, reducing the need for site inspections. This is greatly improved with a good aerial orthophoto dataset.
Figure 2. Map output showing potential non-compliance areas.
Figure 3. Close-up view showing mapped light spreads around each luminaire.
Tabular non-compliance reports were also produced as shown in table 1.
Road Name / Count of Non compliant areas / Total Area m (non compliant)JELLY STREET / 2 / 303
ROSE AVE / 2 / 298
BILSON STREET / 2 / 296
DALE AVENUE / 1 / 288
DINWULLA CRESCENT / 2 / 284
ROSE CRESCENT / 2 / 281
CONDELL CRESCENT / 1 / 279
PRINCE EDWARD STREET / 3 / 279
VISTARAMA STREET / 1 / 278
DOVER LANE / 2 / 274
Table 1. - Sample of non-compliance report, showing count of non-compliance areas per street as well as cumulative area.
Transfer
A presentation was made to all SLI Program participants in July 2004. A workshop was then conducted to provide training to participants and hand over the application. Councils were also encouraged to bring their own street lighting data to resolve any discrepancies in data structures.
At this stage the program can only be used with MapInfo software. This was due to the availability of this software at Kogarah Council, however the data once modelled can be easily exported to all major GIS software.
This project was initially designed for use by member councils of the SLI Program, However the application of the project would be beneficial to any council Australia-wide that has a responsibility for the management of street lighting. It was also recognised that the program had the potential for broader commercial application and could be expanded and adapted to meet the requirements of other parties.
Conclusion
Councils are now faced with increasing responsibility for specifying street lighting technology. While the Streetlight GIS application does not undertake detailed lighting calculations, it does provide a useful indication of likely areas of non-compliance and as such delivers a feasible mechanism for councils to undertake broad-scale assessments.
There are several limitations of the application. However these can mostly be resolved by improved accuracy of street lighting inventory data. Attempts to improve this data across the entire network are currently underway.
This project clearly shows that GIS can play a significant role in managing street lights. The application has provided direct benefit to councils by allowing them to readily estimate the capabilities of their existing street lighting networks on residential roads, understand the potential improvements that proposed lighting changes should bring with regard to compliance and identify areas with inadequate light.
Demonstrations and discussions with councils' regional group, SSROC, EnergyAustralia, the primary street lighting manufacturers and an independent lighting consultant have confirmed that the project is considered to be a first. Next Energy and Kogarah Council have been commended by each of the stakeholders involved.