Brazilian algorithm projects the future of the Amazon rainforest – 05/23/2023 – Environment

A group of researchers from Unicamp (State University of Campinas) developed an algorithm capable of projecting the future of the Amazonian vegetation, presenting scenarios with changes in the forest caused by climate change.

One of the results shows that a drier climate in the region, with a 50% reduction in precipitation, could increase diversity, but with lower carbon storage rates.

This is because there would be an increase in the storage of carbon dioxide (CO_two) in the roots of vegetation to the detriment of absorption through leaves and trunks, which have a greater capacity for accumulation. Taking into account different situations, the scientists calculated that the absorption could be between 57.48% and 57.75% lower compared to regular weather conditions.

The first exclusively Brazilian algorithm of its kind, it was nicknamed Caetê, which in the Tupi-Guarani language means “virgin forest”. The name comes from the acronym CARbon and Ecosystem functional-Trait Evaluation model (in free translation: model for evaluating functional characteristics of carbon and ecosystem). Its first results are described in an article published in the scientific journal Ecological Modeling.

Caetê simulates phenomena of nature using mathematical equations fed by data from environmental conditions, such as rainfall, solar incidence and CO levels_two. With this information, the algorithm answers what the rate of photosynthesis may be under certain conditions or in which part the plant will store more carbon (roots, leaves or trunks). Through this information, it is possible to arrive at the amount of carbon that the forest can store and at which point the native vegetation no longer recovers.

“The main result of the research was to show that the inclusion of diversity in vegetation models improves the ability to forecast climate change, increasing credibility. And the second point, an unexpected result, shows that, after applying a 50% reduction in precipitation, there was an increase in the diversity of plant strategies, but with less removal of carbon from the atmosphere. This may have a different result on the mitigation of climate change. In this case, increasing diversity may not necessarily indicate a positive balance”, says Bianca Fazio Rius, the first author of the article and a doctoral student at Unicamp’s Institute of Biology (IB).

Rius receives support from Fapesp, which also funded the study through a grant to João Paulo Darela Filho and from AmazonFACE, a research program that studies, through field experiments, how CO2 increases_two weather affects the Amazon rainforest, its biodiversity and ecosystem services (find out here).

The researcher is part of the Earth System Science Laboratory team, coordinated by Professor David Montenegro Lapola, who supervised the study.

“With Caetê, at the same time that we seek to improve the representation of the enormous biological diversity of the largest tropical forest in the world, we also create a stimulus for the collection of data in the field, which is still necessary for this type of model”, he explains. Lapola to FAPESP Agency.

The professor was one of the Brazilians who, along with 34 other scientists from national and international institutions, signed an article featured on the cover of Science magazine, earlier this year, showing that about 38% of the current area of ​​the Amazon is suffering from some type of degradation caused by four factors—fire, selective logging (mostly illegal), edge effects (which are changes in forested regions next to deforested zones), and extreme droughts. The result is that carbon emissions derived from the gradual loss of vegetation are equivalent to or even greater than those recorded by deforestation (read more here).

Advantages and disadvantages

Vegetation models have been widely used to explore the fate of the Amazonian forest’s carbon budget under projected future climate conditions. Previous studies have already shown that in the last 40 years the Amazon has become 1°C warmer and there has been a reduction of up to 36% in the level of rainfall in some areas. As a reflection of deforestation, plant degradation and global warming, the forest has also lost its ability to absorb CO_two.

In addition, a report released on May 17 by the World Meteorological Organization warns that the global temperature should reach record levels in the next five years due to the gases that cause the greenhouse effect and the El Niño phenomenon, with a forecast of reduction in the rainfall regime for the Amazon.

However, current algorithms are based on a small set of so-called plant functional types (PFT), with underrepresentation of diversity. With this, the combination of characteristics found in the model ecosystems is simplified given the complexity of the largest tropical forest in the world, generating limited scenarios or scenarios that overestimate the impacts of environmental changes.

Among the types that currently exist are dynamic models of global vegetation (DGVMs) — software that makes simulations and projections of the vegetational dynamics of a region, including Jena Diversity (JeDi). On the other hand, among the advantages of simulations is the fact that they do not depend on logistics and large investments, necessary for carrying out a large-scale field experiment.

Inflection point

Rius explains that the study does not focus on species. “We use the idea that each individual, even individuals within a species, can be considered a type of strategy for dealing with the environment. The computationally created strategies do not necessarily belong to a certain species”, he says.

The researcher explains that the strategies represent a set of characteristics of plants or any living being that will tell how it responds to or affects the environment. That is, a plant that adapts the depth of the root to be able to access water depending on the height of the water table. This can determine the survival and reproduction of these individuals, being linked to ecosystem services, such as the ability to absorb carbon or generate moisture for the rainfall cycle.

“What we saw with the climate getting drier was a change in the occurrence of types of life strategies in the Amazon. We observed an increase in the occurrence of strategies similar to those of the Cerrado. works have already brought”, completes Rius.

The scientists point out that the research with Caetê brought more evidence that the inclusion of variability and diversity may have implications for modeling the so-called “turning point” of the Amazon, when the natural vegetation is no longer able to recover. One of the first articles to address the topic was signed by researchers Thomas Lovejoy (1941-2021), a biologist who coined the term “biological diversity”, and Carlos Nobre, co-president of the Scientific Panel for the Amazon, highlighting the importance of the forest’s hydrological cycle tropical not only for Brazil but also for South America and other regions.

Through evapotranspiration, the forest guarantees moisture throughout the year, which contributes, for example, to rainfall in parts of the Prata river basin, especially in southern Paraguay, Brazil, Uruguay and central-eastern Argentina.

Historic

Caetê began to be created in 2015 based on the CPTEC-PVM2 ecosystem model, which had Lapola as one of those involved in the initial development.

“Most vegetation models represent the Amazon with two or three types of strategies. The proposal was to include greater diversity. We will continue to develop it, because a model is never ready”, says Rius.

Along these lines, Unicamp IB doctoral student Bárbara Cardeli joined the group and is working on the model to include a module aimed at quantifying ecosystem services. “The idea is that, in a simple way, it is possible to see through some processes, for example, how vegetation strategies allocate carbon, which ecosystem services are assured or not. We want to include numerical data, values, how is the offering these services”, says Cardeli.

In the researchers’ view, Caetê may become a tool that provides data for decision-making and the construction of public policies aimed at the carbon market. During the United Nations Conference on Climate Change (COP26), Brazil assumed the commitment to reduce the country’s carbon emissions by 50% by 2030 compared to 2005 data and neutralize them by 2050.

The article “Higher functional diversity improves modeling of Amazon forest carbon storage” can be read here.