The Oil Palm

Global Forest Watch’s Rubbery Plantation Numbers

Global Forest Watch recently issued a study and a blogpost on forest loss and plantations that was nothing less than surprising and at times bizarre. 

The study itself and its objectives seemed reasonably clear. The authors note that there is a lack of data on the location and extent of plantation areas within locales that are considered forest area, particularly because maps and imaging technology are generally unable to distinguish between plantations and natural forests. 

The study’s technical objectives are also reasonably clear:

(1) to create a better understanding of where and to what extent plantations constitute landscapes across seven tropical countries and (2) to estimate how much tree cover change identified … falls within planted areas.

Why is this important to Global Forest Watch?

Most assessments of plantation area rely on government statistics on plantation area. WRI and its Global Forest Watch tool use satellite modelling that contructs data on forest cover, not forest area. So, this kind of distinction is important for WRI to be able to have more accurate data. 

But what the organisation actually ends up publishing as a blogpost is something different altogether. 

There are four points about the data that could be considered contentious. 

First, is the macro level data. GFW states “plantations cover almost a third (30.2 percent) of all land area in Malaysia … the majority are palm oil.”

There’s something of a misnomer in this. It’s not entirely inaccurate, but not accurate either. When we’re talking about a food crop such as palm oil we’re actually talking about agriculture and horticulture, i.e. farms. GFW is basically saying that all farms are plantations, and that there’s forest cover loss taking within these areas. And these are included in the final tally. 

This explains to some extent why the results don’t square that well with government figures on plantation area. 

Second is the definitions used. So, the data limitations stipulate that a medium-sized mosiac area comprising, say oil palm, fruit trees and forest, will be considered ‘oil palm’ if more than half that area is oil palm – and this goes up to areas of 50ha. So, if there’s a 100ha area with 50ha of oil palm in it, it will count as 100ha of oil palm. 

Why is this a problem? Because if there are cases where farmers have set aside forest areas as riparian buffers, or to even join fragmented forest areas, they are not counted as forest per se. This is certainly the case in mosaic forest areas where pulpwood plantations exist alongside natural forest and palm oil. 

Third is the general accuracy of the data so far. WRI says they plan to improve it, but there are a few howlers that indicate these preliminary results shouldn’t have been released at all. 

In an area in Bali, for example, an area listed as ‘oil palm’, contains a seaside resort, large urbanised areas, and zero oil palms. The palms are, rather, coconut palms. This is a perfect illustration of how inaccurate data can lead to conclusions that are simply not factual. This has to raise questions about the overall reliability of GFW’s data.

But there are two other broader points that must be considered. 

First is the utility of having ‘forest cover’ statistics for plantation area in the first place. ‘Forest cover’ is different from ‘forest area’ as a metric, in that forest area takes into account disturbances such as harvesting, fires or pests. 

So, for something like plantation acacia in Sumatra, which has a species rotation of seven years, forest cover changes aren’t a particularly valuable measurement. Forest area is much more relevant.

And it would seem that GFW, like a number of other campaign groups is attempting to row against the use of ‘forest area’ as a credible metric and instead use ‘forest cover’. The most recent FAO assessment of the world’s forests indicated that Malaysia had in fact reversed deforestation, with gains in forest area over the past 15 years. 

But forest cover is GFW’s headline metric. It permits the small losses on a year-to-year basis to be presented as something bigger, ignoring longer-term trends around forest re-growth.

But consider the following: according to GFW’s headline metric of forest cover, US forest cover loss from 2001 to 2014 was 29 million ha; Indonesia’s loss in the same period was 18 million ha.  But which country is actually losing forest area?

But most importantly, GFW’s work has unintentionally caught plantations as a form of tree cover – and by extension forest cover. This is something  that no campaign group would consider as credible, just for political reasons, despite plantations fitting within credible definitions of forest cover. Second is the overall negative perception of plantations. And there are two aspects to this. 

Lumping Indonesian pulp plantations together with oil palm plantations makes about as much sense as lumping North American conifer plantations in with soybean. The purpose of the plantations are so different that they can’t and generally aren’t considered remotely similar by anyone in agricultural or silvicultural fields. So why put them together? It does not make sense.

And, the blog states that ‘natural forests offer benefits’ that plantations can’t. This is certainly true. But in the case of palm oil, a cash crop, this is like saying ‘natural forests offer benefits that farming can’t’.  Again, a parallel would be attempting to compare the benefits of a tropical forest with a coffee plantation. By the same token, plantations can offer many benefits that natural forests can’t.

So, the question is, what’s GFW’s agenda here? 

Are they attempting to muddy the generally positive data that has emerged from countries such as Malaysia, which appears to have reversed deforestation trends over the past decades? Other activists have already done this

Already GFW has placed a new metric on its website: ‘tree cover loss outside of plantations’, reported on a year-to-year basis. This might make good headlines, but it won’t provide other reasons for forest cover loss: fire, disease, sustainable logging, vegetation management. 

Are they objecting to plantations across the board and therefore objecting to farming? That is a parlous position to be in: GFW will know that millions of small farmers benefit from planting oil palm: being opposed to this planting means being opposed to the poverty alleviation it brings.

These GFW mis-steps highlight the need for accuracy, and for facts to prevail over hyperbole. While we’re sure GFW’s technical work will yield some interesting results over the next few years on landscapes, the way the current study is being promoted in the media at the moment needs some serious rethinking.