Critical threshold in Amazon deforestation

A large-scale new study identifies a critical threshold for significant species loss in the Amazon due to deforestation, but also suggests a strategy for action

The deforestation of the Amazon is reaching a critical point for maintaining biodiversity. A year-long study of 3 million hectares in Brazil’s southwestern state of Rondôni revealed that each loss of 10 percent of forest cover meant the extinction of one or two species. But when 43 percent of forest cover is lost, this becomes a threshold where the risk to biodiversity increases rapidly – with each extra 10 percent of disappearing forest cover, the species lost accelerates to between two and eight. 

The research team, lead by Dr Jose Manuel Ochoa-Quintero from Cambridge University’s Department of Zoology, estimate that about a third of the Amazon region is heading to, or is already at, this threshold point, and that unless urgent action is taken these areas will lose between 31-44% of species by 2030. 

The problem of deforestation is exacerbated by the fact that the species in the patches or fragments of forest that remain, have reduced connectivity with the habitats that maintained them. ‘We’re checking to see whether the reason why some of the species disappear is because interaction is being lost in the process of forest fragmentation,’ explains Dr Ochoa-Quintero. ‘You lose the connection between the subpopulation and the different remaining native vegetation. And also within those remaining areas, you are losing interaction so it may entail losing species.’

Underlying dynamics of species loss

The complex interactions where species feed off other species, or species which depend on other species to produce seeds or for pollination, begins to unravel. ‘All these species interactions are very important,’ says Dr Ochoa-Quintero. It’s a challenge to track this change he says, because the issue isn’t simply identifying species loss, but also identifying ‘the mechanisms that generate those responses.’ 

There were two sources of data. Dr Ochoa-Quintero spent a year with an assistant driving around this three million hectares area which they divided into 1,223 squares of 10,000km, selecting 31 squares representative of the spectrum of forest cover across the region. Dr Ochoa-Quintero’s assistant used to hunt in the area and knew the species well. They used traditional observation techniques, logged evidence such as footprints and faeces, and also played recordings of songs to try and elicit responses from small birds who are difficult to see, but are territorial. ‘It’s hard work, the forest is full of mosquitoes,’ reflects Dr Ochoa-Quintero.

Getting access in the field

Support from the University of Sao Paolo, local institutions and people was crucial in giving them access to the forests; four of the areas they studied were Protected Areas – PAs – while 27 of the areas were private lands. These were people, ‘who know local farmers very well,’ explains Dr Ochoa-Quintero. ‘They allowed us to enter because most of the places were private properties, you need to get local support and permission to enter, ask questions and afterwards [they] let us go into the forest to collect the data, to spend some time on the property. You need to spend some time with them, to get their confidence in order to gain access.’  

Collecting this second set of data involved between 5-6 interviews a day. The interviewees, says Dr Ochoa-Quintero, are themselves a slice of Brazilian history. At a time of high unemployment in the 1970s the Brazilian government made an offer of land. The government, explains Dr Ochoa-Quintero ‘told the people “you have a place to work and the land is free.” These people arrived and colonised the area. At the beginning the production was basically cacao plantations, some coffee plantations, nowadays it’s mainly cattle ranches. These were the people at the basis of our interviews, most were living in plots from between 100-150 hectares acres up to 500 hectares.’

The combined evidence of the two separate datasets (the species tracking and the interviews) suggested that something significant was happening. ‘That is where we found a threshold,’ says Dr Ochoa-Quintero. ‘Dr Isabel Rosa at Imperial College was doing a probabilistic model of deforestation for the entire Amazon basin so we decided to combine information she had on the deforestation model with the information we had for the species loss. That is why we tried to interpolate the data we have for the coming 20 years to see to see what is going to happen in the future, given the current forest loss. By combining these sets of information we arrived at these amazing results.’ While there are limits on the amount of land that can be made a Protected Area, the findings are significant as they indicate a strategy for action, suggesting that the focus should be on reforestation efforts (and preventing deforestation) in areas that are on the threshold of species loss.