Monitoring towers

There are a number of different survey techniques that are used for monitoring tower structures.The selection generally depends on accuracy, cost and time constraints of the monitoring scheme. Three different methods for monitoring towers that are independent of GNSS observations have been explained below.

 

1. Historical method

Johnston (1964) observed a steel weather tower at Mascot Airport, Sydney which is around 50 metres high as seen in Figure 2. The method involved establishing two control stations about 100 metres away from the structure and observing to the centre of a lightening conductor at top of the tower.

 

Figure 2 - Weather tower at Mascot (Johnston, 1964)

 

The readings were taken by plumbing down from the top of the tower to a specially calibrated ruler about 30 metres from the observer as shown in Figure 3. By taking two set of face left and face right observations along with weather & temperature readings every half hour it was possible to detect for movement of this tower structure.

 

Figure 3 - Calibrated ruler (Johnston, 1964)


The benefits of this monitoring technique for high structures are its simplicity and the elimination of systematic errors.By using the calibrated ruler for taking readings, it is not necessarily to measure angles saving considerable time in the field.

 

Also, when measuring between top of the tower and a reference point, the dislevelment of the instrument and non-horizontality of the trunion axis can affect horizontal angle readings. By using the ruler technique, only collimation error is present that is eliminated by taking the face left and face right readings.(Johnston, 1964)

 

However, there are limitations when using this method that relate to measurement accuracy as well as the height of the tower structure that can be efficiently monitored. For tall buildings we need to consider other monitoring techniques that have been explained in the sections below

 

2. Modern method using ATR

Modern robotic total stations can measure automatically, continuously and have features such as automatic target recognition (ATR) to locate and track a target. In the ATR mode, the prism needs to be roughly sighted to take a distance measurement. Once a lock is established, the instrument turns to the prism takes a more precise angle and distance measurement.

 

This is especially useful for monitoring tower structures where limitations of the human eye could lead to a distorted line of sight to prism under poor visibility e.g. when sun is low on the horizon.


Leica (2009) quotes an angular accuracy (both Hz and V) of 1” second and a base positioning accuracy of ±1mm when using the ATR on the TPS1200+ total stations.(See fig. 4)

 

Figure 4 - Leica TPS1200+ with ATR (Leica, 2009)

 

Robotic total stations with ATR perform better than GPS in a stop and go situation i.e. taking measurements of a moving object only when it is stationary. (Radovanovic & Teskey, 2001)


The limitations with ATR are its low data sampling rate (compared to GPS) that makes it difficult to detect certain high frequency motions of the object. Also, ATR measurements suffer from reduced measuring ranges under certain weather conditions e.g. bright reflections, heat shimmer,etc. (Harvey, 2008)


Therefore, the ATR method is better suited for measuring slow dynamic deformation of large tower structures that move at slower frequencies and are not too far away from observer.

 

3. Resection method

The monitoring of tower structures can be done by taking resection measurements from a point over multiple epochs. This method is used by the Land & Property Management Authority (LPMA) for the establishment and monitoring of CORSnet NSW (http://corsnet.lpma.nsw.gov.au/), a network of permanent satellite tracking stations in New South Wales.

 

A CORSnet station is essentially a GNSS antenna that is mounted on a stable platform so it moves with the Earth’s crust and can be used in predicting crustal motion. For the required stability, a concrete pillar measuring about 1.5m in height and 40cm across is selected as the mounting platform. The pillar holds a pillar plate and is surrounded by four reference pins set in concrete as seen in Figure 5. (Janssen, 2009)

 

Figure 5 - Pillar plate with reference pins (Janssen, 2010)

 

Janssen (2009) states that every CORSnet station has at least three reference marks (RM) placed in the bedrock about 25 metres away in a triangular pattern around the monument. Additionally, a pillar benchmark (BM) is established near the monument for measuring height differences between pillar plate and reference marks as shown in Figure 6.

 

Figure 6 - Arrangement of the Pillar and RMs (Janssen, 2010)

 

The resection method involves the total station being set up on the pillar and 5 sets of observations are taken to the three reference marks. The LPMA uses this monitoring technique regularly by precise connection to the three reference marks in order to ensure the monument’s integrity.

 

The stated accuracy of the horizontal survey is of the order of 0.5” for directions and ±1mm+1 ppm for distances. The levelling measurement meets the L2A specifications (~0.3mm permitted misclose). (Janssen, 2010).


Therefore repeated measurements by the resection method give a high order of accuracy that is ideal for high precision monitoring applications. However, resection requires the occupation of the target point i.e. the tower structure, which may not be possible in certain situations and other monitoring methods have to be considered.