PG&E Report on
Bird Strike Monitor Performance Testing

Publication Number: 600-00-027
Publication Date: May 2000

The executive summary, abstract and table of contents for this report are available below. This publication is available as an Adobe Acrobat Portable Document Format Files. In order to download, read and print PDF files, you will need a copy of the free Acrobat Reader software installed in and configured for your computer. The software can be downloaded from Adobe Systems Incorporated's website.

---

Executive Summary

Migratory and resident birds can be injured or killed from collisions with overhead conductors or ground wires (Brown et al. 1984, APLIC 1994). For species with limited numbers, this mortality can be significant to the population as well as the individual (Brown et al. 1987, Crivelli et al. 1988). Bird strikes with powerlines can also result in temporary power outages.

Several factors lead utilities, resource agencies, and others to try and quantify bird collisions with wires. First, appropriate remedial action can significantly reduce collisions (Morkill and Anderson 1991, Brown and Drewien 1995). Common methods used to reduce collision mortality include various types of line marking devices. But as these are expensive to install, and may cause visual impacts, they should generally not be used unless needed. Second, resource agencies and environmental groups are increasingly concerned about bird loss from collisions. Studies of collision rates may be needed for permitting documents, or to address concerns raised by agencies or the public. Study methods must be sufficiently reliable that all parties will accept the results.

Developed in the late 1970s to provide worst case estimates of potential bird mortality for use in environmental assessment documents, the current methods to identify responsible lines and provide an accurate count of bird collisions are both labor intensive and unreliable. The primary method is to search under wires for dead birds, which may be difficult to detect, removed by scavengers, or decomposed. Dead bird searches may be impossible in locations over water or wetlands, yet powerlines in these habitats are known to be particularly hazardous to birds (Faanes, 1989). And unless reinforced by direct observations of strikes, correlations with specific line outages, or the presence of characteristic injuries, finding dead birds under lines provides only circumstantial evidence of collisions. Nevertheless, estimates of collision mortality are commonly based on dead bird searches. In addition, inaccuracies are multiplied by corrections for biases presumed to be present in the methods (APLIC 1994).

To provide an alternate, potentially more accurate method, the Pacific Gas & Electric Company developed the Bird Strike Monitor as a cost effective, reliable technique of detecting bird collisions with wires. Affordable, small, lightweight, and easy to install, the monitor is designed for use by transmission and distribution line owners and regulatory agencies to identify and mitigate powerlines responsible for collision-caused bird mortalities. They should function on energized or unenergized conductors of commonly used voltages. To be acceptable to regulatory agencies, the system must be shown to obtain reliable, rigorous data that can be correlated with data collected by conventional dead bird searches.

Objectives

The overall goal of this project was to advance the development and testing of the Bird Strike Monitor to the point where it could be released for fabrication or construction and confidently used in studies of bird strikes. The objectives were to:

• Redesign and improve the Bird Strike Monitor
• Conduct performance tests that simulated live bird strikes with wires to determine the correlation between actual strikes and those recorded by the Bird Strike Monitor.

Outcomes

This project produced two major outcomes:

• The bird strike monitor as designed and tested did not produce consistently reliable data.
• We were unable to correlate the number of bird strikes recorded by the monitor with dead birds or their remains. The primary reason for performance test problems was the drastically changed transmission environment.

Conclusions

The project demonstrated that the design of the bird strike monitor is no longer functional in areas with significant radio traffic in the 900 MHz range. Nevertheless, the system must be able to perform in hostile environments to be credible.

• To produce a bird strike monitor that will reliably record bird strikes, the current system will have to redesigned and incorporate a new radio and new software.
• Existing design can only be used in areas where radio site surveys have shown that interference would not create a problem.

Recommendations

To encourage further development of the system, the following recommendations are offered:

• Obtain recommendations from specialists in wireless data communication special applications on radio frequency bandwidth and technology for future radio selection.
• Depending on the radio chosen, develop software to take advantage of the technology's abilities to resist interference.
• Since the radio must be changed, the monitor's circuit board must be redesigned, and firmware modified. New waterproof housing is required, and can be designed to accommodate the new components.
• Upgrade the data collection method to make it Internet and pager-capable. The system would call the ground station as frequently as desired, and the software would conduct an automatic download of the ground station. If data were available for download, the software would email or page people on a warning list.
• Additional improvements can be made to the functionality of the system:

• Upgrade ground station hardware. By adding an up-to-date single board computer and flash-memory storage system, the existing ground station hardware would be replaced with more capable hardware and software. The ground station could automatically call the server to announce that a bird strike had occurred. It could transmit its data seconds after the sensor had activated. This feature could be quite helpful if it led to the discovery of a downed bird before the carcass was lost to predators.
• It would be useful to have a method of changing the monitor's trigger point short of burning in a ROM chip.
• Clean up minor bugs in ground station software and the modem interface program.

• Once the necessary changes have been made to the system's design, it is critical that prototypes be thoroughly tested in the laboratory and the field.

Benefits to California

The Bird Strike Monitor would benefits California electric ratepayers by providing an improved method of detecting bird collisions with overhead wires, allowing utilities to more easily identify spans in which alleviating measures are needed. It would also be very useful in evaluating the effectiveness of marking and other measures used to reduce bird collisions with wires. Since bird collisions can cause outages (mainly momentary outages), reducing bird strikes contributes to improved electric reliability as well as enhanced environmental quality (especially for the affected birds).

Although the Bird Strike Monitor Performance Testing Project fell short of its original goals, it identified a conflict between the technology and the recent proliferation of wireless devices. This conflict must be resolved to ensure reliable data collection. The funds provided by the California Energy Commission have allowed us to document the system sufficiently to allow development to continue.

Abstract

The goal of this project was to complete development and testing of the Bird Strike Monitor, a device that detects and records bird strikes with powerlines. The monitor would provide a more accurate and less costly method of measuring the number of bird collisions with powerlines.

Pacific Gas and Electric Company's Research and Development Department developed the Bird Strike Monitor, which was field tested near Klamath Falls in 1995 and 1996. The field testing showed that the Bird Strike Monitor could reliably record bird strikes, but it did not yield an unequivocal correlation between dead birds and monitor strikes, mostly due to a paucity of bird strikes. It was felt that additional testing would show that the monitor reliably recorded all bird strikes.

We proposed to test the monitor in a controlled situation. A bird cannon was built to propel dead birds (or other objects) at a test line. We also proposed to make several improvements to the system to correct deficiencies discovered during the Klamath Falls testing.

However, the performance testing did not produce reliable data. A large percentage of strike records suffered from corrupted data from loss of bytes in transmission. The waveforms obtained from the corrupted data were similar to each other, but different from almost anything obtained during laboratory and field tests in 1995. After much troubleshooting, we traced the problem to radio interference in the 900 MHz band used by the monitor's transceivers. This was compounded by the lack of retransmission protocols.

The solution is to use another radio. Changing the radio is not trivial - it requires that several components of the monitor be redesigned, and both firmware and software modified. The redesign was beyond the scope of this project. We have documented the hardware and software of the current system, and will provide it to anyone interested in furthering the Bird Strike Monitor's development. Although the project failed in its original objective, it gave us a clear idea of the modifications needed to make the monitor work in today's changed wireless environment.

Table of Contents

Preface

Executive Summary

Abstract

1.0 Introduction

2.0 Methods

3.0 Outcomes

4.0 Discussion

5.0 Conclusions and Recommendations

6.0 References

Appendices

Appendix I         Monitor Hardware Documentation

Appendix II        Monitor Software Documentation

Appendix III       Ground Station Hardware Documentation

Appendix IV       Ground Station Software Documentation

List of Figures

Figure 1. Bird Strike Monitor

Figure 2. Bird Strike Monitor Components

Figure 3. Monitor Ground Station

Figure 4. Installing Test Line

Figure 5. Bird Cannon Ready to Fire

Figure 6. Typical Waveform Data Produced by the Bird Strike Monitor

Figure 7. Waveform Produced by Corrupted Data

List of Tables

Table 1. Characteristics of two types of spread spectrum radios