Those who monitor global #forest cover, or have any other interest in global #forests should mask global non-forest areas such as #GrassyBiomes by using resources such as the IUCN Global Ecosystem Typology (https://www.iucn.org/resources/conservation-tool/iucn-global-ecosystem-typology) or the RESOLVE Ecoregions 2017© biomes map (https://ecoregions.appspot.com/). They are much better than the global #treecover / #treeheight maps that so many use.

#SavannasMatter
#deforestation
#climatechange
#ClimateCrisis

The worst ever map of global #forest cover is in Fig 1 of this paper:

https://doi.org/10.1016/j.scitotenv.2023.162167

#SavannaScience #SavannasMatter #GrassyBiomes

The FAO’s definition of #forest is based on tree height > 5 m and tree cover > 10%. They introduced the definition in 1980, based on UNESCO’s 'International classification and mapping of vegetation' published in 1973. The FAO’s definition has persisted for > 4 decades despite being outdated and leading to #TropicalGrassyBiomes (savannas) being falsely classified as ‘forest’, 'degraded forest' or 'deforested'.

#SavannasMatter

The problem with large, freely available, datasets is that they are too easy to use blindly. It’s all very well comparing different #RemoteSensing forest cover datasets, but if the definition of forest is wrong, then such studies are wasted resources. The same goes for comparing remote sensing data with ground data, if the ground data are from misclassified sites. Forest inventory data used by #FAO are poor reference data against which to compare satellite products. #SavannasMatter

There is interesting circularity in the use of global forest/non-forest (FNF) and global forest change (GFC) products. FNF maps are used to mask non-forest areas in studies of global forest change, and GFC maps are used to update FNF maps, yet both types of map are almost always based only on vegetation structure (height or tree cover). The situation is analogous to using the wrong screwdriver on a screw that will never tighten, even with the correct screwdriver.

#SavannasMatter #SavannaScience

#TropicalGrassyBiomes like the #Cerrado of South America are under threat of degradation by #Forestization.

Stop large-scale #TreePlanting in #GrassyBiomes. #Savannas are NOT "degraded forests".

Stop mapping global "forests" in terms of vegetation height and woody cover alone. That leads to misclassification of savannas as #forests.

#SavannasMatter

https://news.mongabay.com/2023/04/to-restore-brazils-cerrado-planting-trees-is-a-bad-option-experts-say/

Why is there still an obsessive interest in mapping global forests on the sole basis of remotely sensed vegetation height and woody cover?

For example:

https://www.sciencedirect.com/science/article/pii/S0034425720305381#f0005

https://www.tandfonline.com/doi/full/10.1080/13658816.2023.2183959

Such outdated approaches are prone to misclassifying #GrassyBiomes as "forest" or "degraded forest".

It's now known that plant functional traits and other data not easily acquired by satellites must be included.

For example:

https://spj.science.org/doi/10.34133/2022/9835284

https://www.nature.com/articles/s41586-022-05318-4

#SavannasMatter

I'll be talking about #WoodyPlants and #FunctionalTraits in the Kruger Park version of the SAVANNA #EcosystemModel at this year's #SavannaScience Network Meeting.

https://www.sanparks.org/scientific-services/events/savanna-science-network-meeting-2023

#SavannasMatter

Almost everyone who is anyone in #SavannaScience attends the Savanna Science Network Meeting in Skukuza, Kruger National Park.

I'll be posting updates from this year's meeting, which is the 20th edition.

https://www.sanparks.org/scientific-services/events/savanna-science-network-meeting-2023

#SavannasMatter

@TobiasKuemmerle
Interesting analysis, but depends heavily on how deforestation is defined and the ability to differentiate between forest, degraded forest and #TropicalGrassyBiomes (i.e. savannas). They acknowledge the inability to differentiate between forest and agro-forests, but not between degraded forest and savanna. Use of Hansen et al's (2013) Global Forest Change data is problematic because forest is defined as >50 % crown cover of trees taller than 5 m height.

#SavannasMatter

Savannas are perpetually misclassified as forests.
https://besjournals.onlinelibrary.wiley.com/doi/epdf/10.1111/2041-210X.14040

Savannas like longleaf pine savanna in Florida are not "forested ecosystems". They are #TropicalGrassyBiomes whose plants are adapted fire.

#SavannasMatter

@marijnvdpas

News headlines like this create a negative perception of savannas in the eyes of the general public.

I can't read the whole NYT article, but I guess it's based on the recent paper in PNAS 119(52), e2203200119, which I have yet to read thoroughly, but my first impression was mixed. On the downside, portraying #TropicalGrassyBiomes as a stage in linear succession from grassland to forest is outdated, misleading and ignores current thinking in #SavannaScience.

#SavannasMatter

@wclark

Into "savanna states" or into "savanna-like states"? One assumes that forest and savanna are alternate stable states composed of species with very different functional traits. The other does not.

Savannas and other #GrassyBiomes are not "degraded forests".

#SavannasMatter

@pmeyfroidt

Important work.

But the analysis depends on how restoration priority areas are determined, which depends on how ecosystems are defined. Strassburg et al. (2020) is seriously flawed in that regard. Their Extended Data Figs 1 and 5 do not distinguish between "forest" and #GrassyBiomes such as savannas, especially in Africa and South America. This leads to intact savanna areas being misclassified as "degraded forest" and mislabelled as "restoration priority areas".

#SavannasMatter

@dominic

Important work.

Caveat: The analysis depends on how restoration priority areas are determined, which depends on how ecosystems are defined. Strassburg et al. (2020) is seriously flawed in that regard. Their Extended Data Figs 1 and 5 do not distinguish between "forest" and #GrassyBiomes such as savannas, especially in Africa and South America. This leads to intact savanna areas being misclassified as "degraded forest" and mislabelled as "restoration priority areas".

#SavannasMatter

@hakim

Excellent review of satellite remote sensing for mapping #GrassyBiomes at regional and global scales.

This is crucial for #SavannaScience and the prevention of #Forestization of #OpenNaturalEcosystems such as #TropicalGrassyBiomes.

The EU and FAO need to read this, and other recent related work, and then realise they need to update their antiquated definition of "forest" based only on veg structure and ignoring #FunctionalTraits.

#SavannasMatter

It's time we stopped referring to fire and herbivores as 'disturbances' in #SavannaScience.

They are important *consumers* in #TropicalGrassyBiomes, and savanna plants are adapted to them.

#SavannasMatter

@resilienceSci

Interesting concept, but neither approach used to define "forest" can distinguish undegraded savanna from forest, so the results are misleading, especially with regards to Africa.

#SavannasMatter

@TobiasKuemmerle

Why is there stubborn refusal by the FAO to revise the definition of "forest" it has clung to since 1973 and acknowledge the importance of ecologically functional attributes in the definition?

Why does FAO not follow IUCN?

https://global-ecosystems.org/explore/biomes/T4

#SavannasMatter

@TobiasKuemmerle

"They set an innovative concept for the definition of ‘forest degradation’ meaning the structural changes to forest cover, taking the form of the conversion of naturally regenerating forests and primary forests into plantation forests and other wooded land and the conversion of primary forests into planted forests."

https://www.consilium.europa.eu/en/press/press-releases/2022/12/06/council-and-parliament-strike-provisional-deal-to-cut-down-deforestation-worldwide/

What is "other wooded land"? Not all "other wooded land" is degraded forest.

#SavannasMatter