Monday 20th September 2010
Recent checks by EEG of VicForests corporate reports have found examples of VicForests blatantly misrepresenting its environmental performance in a recent corporate report. VicForests has boasted in its 2008 Corporate Plan Highlights report that there had been “measurable improvements in the environmental performance of forest management operations on State forest.”
VicForests has boasted in its 2008 Corporate Plan Highlights report that there had been “measurable improvements in the environmental performance of forest management operations on State forest.” In support of this claim, VicForests presented a graph based on EPA Audits of Compliance to try and prove that things had really looked up in Victorian State forests since the Department of Sustainability and Environment had been dumped as manager of logging operation.
Figure 1 graphs the improvement in VicForests’ environmental compliance scores in EPA audits undertaken since 2004/05. It also shows the downward trend in the number of recommendations that the EPA has made for improved performance. Compliance has improved and the number of recommendations has decreased since VicForests took over responsibility for commercial activities on State forest.
VicForests plotted two series of data to try and prove its point
- Counts of the number of recommendations made in each report
- Scores from the EPA Audits
In the graph, VicForests tries to create the impression that the EPA was becoming less concerned about environmental breaches since VicForests had taken over from the Department of Sustainability and Environment in 2004. VicForests has used a number of tricks that could have come straight out of “How to Lie with Statistics” and is a textbook case of statistical misrepresentation.
Trick 1: Selective omission of data
This one is a favourite of climate sceptics and other data benders. If you have a series of results over a number of years and the results at either end get in the way of the point you are trying to make just chop of the offending bits of the series.
Trick 2: Count anything and pretend it’s relevant to the issue
To prove its point, VicForests decided to use the number of recommendations made by the EPA as a measure of noncompliance with the Code. The trouble with this approach is that problems discovered in later Audits will not end up as a recommendation if the problem has already being raised as a recommendation in an earlier report. A more realistic comparison would have been to count the number of new recommendations AND all the open recommendations. (You can see these results in the table below). However, there is a bigger problem with the VicForests approach – many problems identified in these Audits did not end up creating a recommendation. For example, in the 2007 report, over one third of the measurements made to check soils were incorrect. In East Gippsland, Vicforests had failed to complete regeneration surveys for more than 80% of the coupes and the auditors found 4 major environmental breaches but neither of these matters ended up as a recommendation.
Trick 3: Hide any other incriminating data
This one is VicForests’ best trick; in 2005, VicForests was in charge of logging when the following 4 major environmental breaches were detected by an EPA Special Audit:
- Logging of a part of the Errinundra National Park
- Illegal logging outside approved boundaries at two other East Gippsland coupes
- A large part of an SPZ for Superb Parrots was destroyed by illegal logging in the Barmah state Forest
This appalling bungling resulted in the EPA making another 20 recommendation in 2005, but none of these appear in the Vicforests graph.
Trick 4: Pretend the audits are conducted in a political vacuum.
The final EPA audit was conducted during 2007 and released the following year. The review was carried out under very difficult conditions. The Department of Sustainability and Environment had decided to review the EPA Audit program. The review was conducted by Sinclair Knight Merz, and their report makes very clear the widespread hostility to the EPA coming from industry groups and VicForests. VicForests was apparently peeved at having its fantastic work scrutinised by outsiders whose work they considered substandard. Comments had been made during stakeholder forums that some participants did not even believe that auditors had any jurisdiction to make recommendations. DSE ignored the recommendation of the review to keep the audit program as an independent review and shut it down in 2008. Given the hostility shown to the EPA, it is hardly surprising they bothered making many recommendations.
Even though VicForests use of a count of recommendations is pretty wonky, the extent of misrepresentation becomes clearer when the missing years are added to the data.
Here you can see that even if VicForests method is followed, their claim that things got much better after they took over in 2004/05 is nonsense – after 2003/04, there is a big increase the following year. The figure for 2005/06 does not include the 20 recommendations from the Special Audit.
In 2003/04, the EPA started reporting the number of environmental breaches. These could be things like logging in an SPZ, felling trees into a buffer etc. Using Trick 3, VicForests has conveniently forgotten to plot these figures, because the pattern it is trying to present to the public of steady improvement is absent.
In addition to these tricks, VicForests had one further trick up its sleeves –
Trick 5: Pretend that data for different years is comparable
This trick mostly applies to the blue graph showing a tiny increase in the EPAs overall result for each audit. In fact these scores are not at all easy to compare because they are based on a small sample of all the coupes. The selection of these coupes was very biased – in the three years covered in the graph, the Central Gippsland FMA was included in every audit, but the Dandenong FMA was not checked once!
Also, the way the audit score was calculated effectively hid any signs of serious ecological damage done to the forest. For example if in 2007, the auditors had found all the habitat trees burnt on the East Gippsland coupes then the overall result would have dropped by about 1%. The audit process gave more weight to checking that all the logger’s litter had been cleared from the logging coupe than ecologically important things like habitat tree protection. The audit process often found serious environmental breaches detected (for example, one of the 109 Environmental Impacts in the table above) but the overall audit score could still end up being quite high. VicForests publicity is very misleading in suggesting that a small increase in an audit score has anything to do with the lack of environmental damage caused by logging.
After having the book thrown at them by the EPA in 2005, you’d need a hide as thick as a rhinoceros to keep pumping out corporate spin without missing a beat but Vicforests seems filled with these types who just can’t see the forest for the wood.