<p>Total tropical primary forest loss in 2023 totaled 3.7 million hectares, the equivalent of losing almost 10 football (soccer) fields of forest per minute. <a href="https://gfw.global/3TPufvy" rel="noopener" target="_blank">View on the Global Forest Review.</a></p>


2023 Tree Cover Loss Data Analysis

overview of a river in Pantanal, Brazil

Insights

Insights | Global Forest Watch Blog

amazon

amazon | Global Forest Watch Blog

Brazil

Brazil | Global Forest Watch Blog

Congo Basin

Congo Basin | Global Forest Watch Blog

deforestation

deforestation | Global Forest Watch Blog

forest change

forest change | Global Forest Watch Blog

Indonesia

Indonesia | Global Forest Watch Blog

Mikaela Weisse

Liz Goldman

Sarah Carter

Tropical Forest Loss Drops Steeply in Brazil and Colombia, but High Rates Persist Overall

Despite progress in some countries, the rate of tropical primary forest loss in 2023 remained persistent, according to new data from UMD on GFW.

Comment

Global Forest Watch Content

https://content.globalforestwatch.org/wp-content/uploads/2018/06/gfwGlobal.png

https://secure.gravatar.com/avatar/be19f67e91ba81d93d32ae47706dcbb8?s=96&d=mm&r=g

Paige Johnson

Aerial footage of palm oil and the forest in Sentabai Village, West Kalimantan, 2017.

Data

Data | Global Forest Watch Blog

<p>New tree cover loss data from the University of Maryland’s GLAD lab (UMD) and available on Global Forest Watch (GFW) shows declines in primary forest loss in some countries but a persistent rate overall in 2023. What does the data measure, what is different this year and how does it compare to other official estimates of deforestation? Here is what you should know about the new data.</p>



<h2 class="wp-block-heading">What tree cover loss measures and how this differs from deforestation</h2>



<p>The tree cover loss data from UMD captures the annual loss of all trees taller than five meters for the calendar years between 2001 and 2023. The tree cover loss data includes the loss of trees in natural forests as well as in plantations and tree crops (though notably, the primary forest data we use to filter the loss in the tropics excludes plantations, tree crops and regrowing forests). Loss of this tree cover can be due to human or natural causes and can be permanent or temporary.</p>



<p>Deforestation differs in that it only refers to a human-caused, permanent change from forest to another land use. Some forms of tree cover loss, such as conversion of a natural forest to agricultural land, are considered deforestation, while other forms of tree cover loss, such as timber harvesting in plantation forests or natural disturbances, are not — <a href="https://research.wri.org/gfr/key-terms-definitions" target="_blank" rel="noreferrer noopener">read more about the differences here</a>. Loss in tree cover, whether deforestation or not, can be legal or illegal.</p>



<p>In some cases, such as in the <a href="https://gfw.global/4apxyiz" target="_blank" rel="noreferrer noopener">Deforestation and Restoration Targets Tracker</a> on the Global Forest Review, we use a proxy for deforestation. That proxy includes losses that are from small-scale agriculture in humid tropical primary forests and all commodity-driven deforestation and urbanization (inside and outside of humid tropical primary forests). Temporary losses, such as from fires or forestry activities, are not included. We don’t use this deforestation proxy in all of our reporting because it is based on coarse resolution data that is designed to limit the inclusion of temporary losses or losses in planted forests at the global scale only. Deforestation continues to be difficult to track with globally consistent data sets and there are inherent delays in confirming whether loss detected in satellite imagery is temporary or permanent.</p>



<h2 class="wp-block-heading">Data improvements over time result in some inconsistencies in the data timeseries</h2>



<p>Algorithm adjustments and better satellite data have <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/">improved the tree cover loss data set over time</a>.</p>



<p>This year, the loss detection algorithm used by the University of Maryland to create the tree cover loss data set was supplemented by information from the <a href="https://glad.umd.edu/dataset/DIST-ALERT" target="_blank" rel="noreferrer noopener">DIST-ALERT land disturbance data set</a>. Late season changes are sometimes missed in the tree cover loss data due to insufficient satellite data and cloud cover and are attributed to the following year. DIST-ALERT showed patches of tree cover loss that were originally missed by the tree cover loss data, and these were manually reviewed and included for 2023.</p>



<p>Employing DIST-ALERT helped especially to detect loss from late season forest fires in boreal forests and late season clearing in the tropics that would have otherwise appeared in the 2024 data. The DIST-ALERT data also helped detect some tree cover loss due to flooding and mining. The use of alerts resulted in a 3.7% increase in global tree cover loss detected. Most of the changes affected the boreal regions, while for the rest of the globe, the increase was only 1.7%. It is important to note that since most of the added areas would have otherwise been detected in 2024, we do not believe this change in the mapping approach for 2023 will have any impact on long-term trends in the data.&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</p>



<p>Additional changes to the time series include adjustments to the algorithm for the years 2011-2014 and 2015 onwards and the incorporation of Landsat 8 data starting in 2013. These changes make it easier to detect smaller scale changes, such as loss due to fires, selective logging and shifting agriculture. Variations in satellite image availability also mean there are inconsistencies with the quality and number of images available to capture data each year.</p>



<p>To address these inconsistencies, we:</p>



<ul>
<li>Focus our analysis on trends post-2015.</li>



<li>Assess the three-year moving average to interpret longer-term trends.</li>



<li>Disregard the increase in loss post-2012 in places dominated by small-scale agriculture, like Central Africa.</li>
</ul>



<h2 class="wp-block-heading">What does tree cover loss from fire capture?</h2>



<p>The tree cover loss from fire data from UMD allows us to distinguish fire-driven loss from other tree cover loss by assigning likelihood of loss due to fire to each 30-meter tree cover loss pixel, breaking the data into loss due to fire and loss due to other drivers, such as agriculture or logging. The tree cover loss due to fire data includes natural or human-ignited fires that result in direct loss of tree canopy cover. The data captures wildfires, fires used to clear land for another use and intentionally set fires that result in tree cover loss (including escaped fires started by humans for purposes related to agriculture, hunting, recreation or arson). Instances where trees are felled and later burned are not included, however, since the initial driver of loss is mechanical removal.</p>



<p>This data allows us to better understand how fire impacted tree cover loss in 2023. For example, fire played especially large roles in Bolivia and Canada — <a href="https://gfw.global/43tcLbQ" target="_blank" rel="noreferrer noopener">learn more about our findings from the 2023 data here</a>.</p>



<p>Even though fires often do not result in a permanent change in land use, they are still an important source of carbon emissions and can <a href="https://www.wri.org/insights/global-trends-forest-fires" target="_blank" rel="noreferrer noopener">lead to feedback loops</a> of increasing emissions, hotter temperatures, drier forests and more fires.</p>



<h2 class="wp-block-heading">Why we focus on the tropics</h2>



<p>In our <a href="https://gfw.global/43tcLbQ" target="_blank" rel="noreferrer noopener">analysis of the UMD data</a>, we largely focus on tree cover loss in tropical primary forests because tropical forests experience <a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&utm_medium=bitly&utm_source=GFWBlog#how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests" target="_blank" rel="noreferrer noopener">the vast majority</a> (over 96%) of the world’s deforestation, and loss in those areas has huge impacts on biodiversity and carbon storage. Even if these losses are eventually reversed, it will <a href="https://www.science.org/doi/10.1126/science.abh3629" target="_blank" rel="noreferrer noopener">take decades</a> for these habitats and carbon stocks to recover, and permanent biodiversity loss may occur.</p>



<h2 class="wp-block-heading">How does tree cover loss compare with official country data in Brazil and Indonesia?</h2>



<p><em>Brazil</em></p>



<p>PRODES — the official forest monitoring system for the Amazon from Brazil’s National Institute for Space Research (INPE) — and the UMD data have methodological and definitional differences that are important to understand when comparing these two data sources. PRODES measures clear cut deforestation and anthropogenic forest losses by fire larger than 6.25 hectares, whereas UMD captures loss larger than 0.09 hectares of all trees taller than five meters. These are both important measures that help us understand how forests are changing since deforestation, forest fires and small disturbances to forest canopy can all lead to impacts on climate, biodiversity and ecosystem services.</p>



<figure class="wp-block-image size-full"><img decoding="async" loading="lazy" width="1348" height="1042" src="https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN.png" alt="UMD and PRODES differences explained" class="wp-image-15866" srcset="https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN.png 1348w, https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN-1024x792.png 1024w, https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN-768x594.png 768w" sizes="(max-width: 1348px) 100vw, 1348px" /></figure>



<p>Both systems show a downward trend in loss for 2023 — 22% decrease for PRODES and 39% decrease for UMD (42% decrease for non-fire loss) in the Amazon biome, confirming the positive news for the Amazon.</p>



<h4 class="wp-block-heading">PRODES VS. UMD DATA IN THE BRAZILIAN AMAZON</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/17268908?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p><em>Indonesia</em></p>



<p>Overlaying the UMD/GFW primary forest loss data for 2023 with the official Indonesia land cover map from the Ministry of Environment and Forestry (MoEF) shows that 70% occurred within Indonesia’s official forest cover classes, with the remaining 30% taking place in mixed dry land agriculture, shrubland, swamp shrub and other land cover types. Approximately 144 thousand hectares of loss were found to be both within Indonesia’s official forest land cover classes and with a patch size larger than two hectares. Of these 144 thousand hectares, about 15 thousand hectares were within areas legally classified as primary forest in Indonesia.</p>



<table border="0" cellpadding="0" cellspacing="0">
	<tbody>
		<tr>
			<td style="width:350px;height:39px;">
			<p><strong>MoEF class</strong></p>
			</td>
			<td style="width:550px;height:39px;">
			<p><strong>Percent of UMD/GFW 2023 Primary Forest Loss</strong></p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Primary forest</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>9%</p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Secondary forest</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>60%</p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Plantation forest</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>1%</p>
			</td>
		</tr>
<tr>
			<td style="width:350px;height:39px;">
			<p>Non-forest</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>30%</p>
			</td>
		</tr>
	</tbody>
</table>



<p></p>



<style type="text/css"></style>



<p></p>



<p></p>

Global Forest Watch’s 2023 Tree Cover Loss Data Explained

<p>New tree cover loss data from the University of Maryland’s GLAD lab (UMD) and available on Global Forest Watch (GFW) shows declines in primary forest loss in some countries but a persistent rate overall in 2023. What does the data measure, what is different this year and how does it compare to other official estimates of deforestation? Here is what you should know about the new data.</p>



<h2 class="wp-block-heading">What tree cover loss measures and how this differs from deforestation</h2>



<p>The tree cover loss data from UMD captures the annual loss of all trees taller than five meters for the calendar years between 2001 and 2023. The tree cover loss data includes the loss of trees in natural forests as well as in plantations and tree crops (though notably, the primary forest data we use to filter the loss in the tropics excludes plantations, tree crops and regrowing forests). Loss of this tree cover can be due to human or natural causes and can be permanent or temporary.</p>



<p>Deforestation differs in that it only refers to a human-caused, permanent change from forest to another land use. Some forms of tree cover loss, such as conversion of a natural forest to agricultural land, are considered deforestation, while other forms of tree cover loss, such as timber harvesting in plantation forests or natural disturbances, are not — <a href="https://research.wri.org/gfr/key-terms-definitions" target="_blank" rel="noreferrer noopener">read more about the differences here</a>. Loss in tree cover, whether deforestation or not, can be legal or illegal.</p>



<p>In some cases, such as in the <a href="https://gfw.global/4apxyiz" target="_blank" rel="noreferrer noopener">Deforestation and Restoration Targets Tracker</a> on the Global Forest Review, we use a proxy for deforestation. That proxy includes losses that are from small-scale agriculture in humid tropical primary forests and all commodity-driven deforestation and urbanization (inside and outside of humid tropical primary forests). Temporary losses, such as from fires or forestry activities, are not included. We don’t use this deforestation proxy in all of our reporting because it is based on coarse resolution data that is designed to limit the inclusion of temporary losses or losses in planted forests at the global scale only. Deforestation continues to be difficult to track with globally consistent data sets and there are inherent delays in confirming whether loss detected in satellite imagery is temporary or permanent.</p>



<h2 class="wp-block-heading">Data improvements over time result in some inconsistencies in the data timeseries</h2>



<p>Algorithm adjustments and better satellite data have <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/">improved the tree cover loss data set over time</a>.</p>



<p>This year, the loss detection algorithm used by the University of Maryland to create the tree cover loss data set was supplemented by information from the <a href="https://glad.umd.edu/dataset/DIST-ALERT" target="_blank" rel="noreferrer noopener">DIST-ALERT land disturbance data set</a>. Late season changes are sometimes missed in the tree cover loss data due to insufficient satellite data and cloud cover and are attributed to the following year. DIST-ALERT showed patches of tree cover loss that were originally missed by the tree cover loss data, and these were manually reviewed and included for 2023.</p>



<p>Employing DIST-ALERT helped especially to detect loss from late season forest fires in boreal forests and late season clearing in the tropics that would have otherwise appeared in the 2024 data. The DIST-ALERT data also helped detect some tree cover loss due to flooding and mining. The use of alerts resulted in a 3.7% increase in global tree cover loss detected. Most of the changes affected the boreal regions, while for the rest of the globe, the increase was only 1.7%. It is important to note that since most of the added areas would have otherwise been detected in 2024, we do not believe this change in the mapping approach for 2023 will have any impact on long-term trends in the data.&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</p>



<p>Additional changes to the time series include adjustments to the algorithm for the years 2011-2014 and 2015 onwards and the incorporation of Landsat 8 data starting in 2013. These changes make it easier to detect smaller scale changes, such as loss due to fires, selective logging and shifting agriculture. Variations in satellite image availability also mean there are inconsistencies with the quality and number of images available to capture data each year.</p>



<p>To address these inconsistencies, we:</p>



<ul>
<li>Focus our analysis on trends post-2015.</li>



<li>Assess the three-year moving average to interpret longer-term trends.</li>



<li>Disregard the increase in loss post-2012 in places dominated by small-scale agriculture, like Central Africa.</li>
</ul>



<h2 class="wp-block-heading">What does tree cover loss from fire capture?</h2>



<p>The tree cover loss from fire data from UMD allows us to distinguish fire-driven loss from other tree cover loss by assigning likelihood of loss due to fire to each 30-meter tree cover loss pixel, breaking the data into loss due to fire and loss due to other drivers, such as agriculture or logging. The tree cover loss due to fire data includes natural or human-ignited fires that result in direct loss of tree canopy cover. The data captures wildfires, fires used to clear land for another use and intentionally set fires that result in tree cover loss (including escaped fires started by humans for purposes related to agriculture, hunting, recreation or arson). Instances where trees are felled and later burned are not included, however, since the initial driver of loss is mechanical removal.</p>



<p>This data allows us to better understand how fire impacted tree cover loss in 2023. For example, fire played especially large roles in Bolivia and Canada — <a href="https://gfw.global/43tcLbQ" target="_blank" rel="noreferrer noopener">learn more about our findings from the 2023 data here</a>.</p>



<p>Even though fires often do not result in a permanent change in land use, they are still an important source of carbon emissions and can <a href="https://www.wri.org/insights/global-trends-forest-fires" target="_blank" rel="noreferrer noopener">lead to feedback loops</a> of increasing emissions, hotter temperatures, drier forests and more fires.</p>



<h2 class="wp-block-heading">Why we focus on the tropics</h2>



<p>In our <a href="https://gfw.global/43tcLbQ" target="_blank" rel="noreferrer noopener">analysis of the UMD data</a>, we largely focus on tree cover loss in tropical primary forests because tropical forests experience <a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&amp;utm_medium=bitly&amp;utm_source=GFWBlog#how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests" target="_blank" rel="noreferrer noopener">the vast majority</a> (over 96%) of the world’s deforestation, and loss in those areas has huge impacts on biodiversity and carbon storage. Even if these losses are eventually reversed, it will <a href="https://www.science.org/doi/10.1126/science.abh3629" target="_blank" rel="noreferrer noopener">take decades</a> for these habitats and carbon stocks to recover, and permanent biodiversity loss may occur.</p>



<h2 class="wp-block-heading">How does tree cover loss compare with official country data in Brazil and Indonesia?</h2>



<p><em>Brazil</em></p>



<p>PRODES — the official forest monitoring system for the Amazon from Brazil’s National Institute for Space Research (INPE) — and the UMD data have methodological and definitional differences that are important to understand when comparing these two data sources. PRODES measures clear cut deforestation and anthropogenic forest losses by fire larger than 6.25 hectares, whereas UMD captures loss larger than 0.09 hectares of all trees taller than five meters. These are both important measures that help us understand how forests are changing since deforestation, forest fires and small disturbances to forest canopy can all lead to impacts on climate, biodiversity and ecosystem services.</p>



<figure class="wp-block-image size-full"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN.png" alt="UMD and PRODES differences explained" class="wp-image-15866"/></figure>



<p>Both systems show a downward trend in loss for 2023 — 22% decrease for PRODES and 39% decrease for UMD (42% decrease for non-fire loss) in the Amazon biome, confirming the positive news for the Amazon.</p>



<h4 class="wp-block-heading">PRODES VS. UMD DATA IN THE BRAZILIAN AMAZON</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/17268908?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p><em>Indonesia</em></p>



<p>Overlaying the UMD/GFW primary forest loss data for 2023 with the official Indonesia land cover map from the Ministry of Environment and Forestry (MoEF) shows that 70% occurred within Indonesia’s official forest cover classes, with the remaining 30% taking place in mixed dry land agriculture, shrubland, swamp shrub and other land cover types. Approximately 144 thousand hectares of loss were found to be both within Indonesia’s official forest land cover classes and with a patch size larger than two hectares. Of these 144 thousand hectares, about 15 thousand hectares were within areas legally classified as primary forest in Indonesia.</p>



<table border="0" cellpadding="0" cellspacing="0">
	<tbody>
		<tr>
			<td style="width:350px;height:39px;">
			<p><strong>MoEF class</strong></p>
			</td>
			<td style="width:550px;height:39px;">
			<p><strong>Percent of UMD/GFW 2023 Primary Forest Loss</strong></p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Primary forest</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>9%</p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Secondary forest</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>60%</p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Plantation forest</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>1%</p>
			</td>
		</tr>
<tr>
			<td style="width:350px;height:39px;">
			<p>Non-forest</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>30%</p>
			</td>
		</tr>
	</tbody>
</table>



<p></p>



<style type="text/css"></style>



<p></p>



<p></p>

<p>Novos dados de perda de cobertura arbórea do GLAD Lab da University of Maryland (UMD) e disponíveis na plataforma Global Forest Watch (GFW) mostram uma redução na perda de floresta primária em alguns países, mas uma taxa geral persistente em 2023. O que os dados medem, o que há de diferente este ano e como eles se comparam a outras estimativas oficiais de desmatamento? Aqui está o que você precisa saber sobre os novos dados.</p>



<h2 class="wp-block-heading">O que a perda de cobertura arbórea mede e como ela difere do desmatamento </h2>



<p>Os dados de perda de cobertura arbórea da UMD capturam a perda anual de todas as árvores com mais de cinco metros de altura entre os anos de 2001 e 2023. Os dados sobre perda de cobertura arbórea incluem a perda de árvores em florestas naturais, bem como em plantações e cultivo arbóreo (embora seja importante notar que os dados sobre florestas que usamos para filtrar a perda nos trópicos excluem plantações, cultivos de árvores e florestas em regeneração). A perda dessa cobertura pode ser devido a causas humanas ou naturais, permanente ou temporária.</p>



<p>O desmatamento difere por se referir apenas a uma mudança permanente e causada pelo homem, partindo do estado de floresta para outros usos da terra. Algumas formas de perda de cobertura arbórea, como a conversão de uma floresta natural em terras agrícolas, são consideradas desmatamento, enquanto outras formas, como a extração de madeira em florestas plantadas ou distúrbios naturais, não são. <a href="https://research.wri.org/gfr/key-terms-definitions" target="_blank" rel="noreferrer noopener">Leia mais sobre as diferenças aqui</a>. A perda de cobertura arbórea, seja desmatamento ou não, pode ser legal ou ilegal.</p>



<p>Em alguns casos, como no <a href="https://gfw.global/4apxyiz" target="_blank" rel="noreferrer noopener">Targets Tracker de Desmatamento e Restauração</a> no Global Forest Review, usamos um indicador para desmatamento. Esse indicador inclui as perdas que são de agricultura de pequena escala em florestas tropicais primárias úmidas e todo desmatamento e urbanização impulsionados por <em>commodities </em>(dentro e fora de florestas tropicais primárias úmidas). Perdas temporárias, como as decorrentes de incêndios ou atividades florestais, não foram incluídas. Não usamos esse indicador de desmatamento em todos os nossos relatórios porque é baseado em dados de baixa resolução que são concebidos para limitar a inclusão de perdas temporárias ou perdas em florestas plantadas apenas em escala global. O desmatamento continua a ser difícil de rastrear com conjuntos de dados consistentes globalmente e há atrasos inerentes para confirmar se as perdas detectadas em imagens de satélite são temporárias ou permanentes.</p>



<h2 class="wp-block-heading">Melhorias de dados ao longo do tempo resultam em algumas inconsistências na série temporal de dados</h2>



<p>Ajustes de algoritmo e melhores dados de satélite <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/">melhoraram o conjunto de dados de perda de cobertura arbórea ao longo do tempo</a>.</p>



<p>Este ano, o algoritmo de detecção de perda usado pela University of Maryland para criar o conjunto de dados de perda de cobertura arbórea foi suplementado pelas informações do <a href="https://glad.umd.edu/dataset/DIST-ALERT" target="_blank" rel="noreferrer noopener">conjunto de dados de alteração do solo do DIST-ALERT</a>. Mudanças da última estação às vezes são perdidas nos dados de perda de cobertura arbórea devido a dados de satélite insuficientes e cobertura de nuvens e são atribuídos ao ano seguinte. O DIST-ALERT mostrou fragmentos de perda de cobertura arbórea que originalmente foram omitidos pelos dados de perda de cobertura arbórea e incluídos para 2023 e esses foram revisados manualmente e incluídos em 2023.</p>



<p>Utilizar o DIST-ALERT ajudou especialmente a detectar a perda por queimadas da última estação em florestas boreais e desmatamento da última estação nos trópicos que de outro modo apareceriam nos dados de 2024. Os dados do DIST-ALERT também ajudaram a detectar parte da perda de cobertura arbórea devido à inundação e mineração. O uso de alertas resultou em um aumento de 3,7% em perda de cobertura arbórea detectada. A maioria das mudanças afetou as regiões boreais, embora para o restante do globo, o aumento foi de apenas 1,7%. É importante observar que, como a maioria das áreas adicionadas teria sido detectada em 2024 de outra forma, não acreditamos que essa mudança na abordagem de mapeamento em 2023 terá algum impacto nas tendências de longo prazo nos dados.</p>



<p>Mudanças adicionais para a série temporal inclui ajustes no algoritmo para os anos de 2011 a 2014 e a partir de 2015, bem como a incorporação de dados do Landsat 8 com início em 2013. Essas alterações facilitam a detecção de mudanças em escalas maiores, como a perda devido a incêndios, exploração madeireira seletiva e agricultura de rotação. Variações na disponibilidade de imagens por satélite também significam que há inconsistências na qualidade e no número de imagens disponíveis para capturar dados a cada ano.</p>



<p>Para resolver essas inconsistências, nós:</p>



<ul>
<li>concentramos nossa análise em tendências após 2015;</li>



<li>avaliamos a média móvel de três anos para interpretar as tendências de longo prazo; e</li>



<li>desconsideramos o aumento da perda após 2012 em locais dominados pela agricultura de pequena escala, como a África Central.</li>
</ul>



<h2 class="wp-block-heading">O que a perda de cobertura arbórea devido a incêndio capta?</h2>



<p>Os dados de perda de cobertura arbórea devido a incêndio da UMD nos permitem distinguir de outras mudanças na cobertura florestal ao atribuir a probabilidade de perda devido a incêndio para cada pixel de perda de cobertura arbórea de 30 metros, dividindo os dados em perda devido a incêndio e perda devido a outros fatores, como agricultura ou extração de madeira. Esses dados incluem incêndios naturais ou provocados pelo homem que resultam em perda direta de cobertura de copa das árvores. Os dados capturam incêndios florestais, incêndios usados para limpar a terra para outro uso e incêndios intencionais que resultam em perda de cobertura arbórea (incluindo incêndios escapados iniciados por humanos para fins relacionados à agricultura, caça, recreação ou incêndio criminoso). No entanto, os casos em que as árvores são derrubadas e posteriormente queimadas não são incluídos, uma vez que o fator inicial de perda é a remoção mecânica.</p>



<p>Esses dados nos permitem entender melhor como incêndios afetaram a perda de cobertura arbórea em 2023. Por exemplo, incêndios desempenharam papéis especialmente grandes na Bolívia e Canadá. <a href="https://gfw.global/43tcLbQ" target="_blank" rel="noreferrer noopener">Saiba mais sobre nossas descobertas a partir dos dados de 2023 aqui</a>.</p>



<p>Embora os incêndios muitas vezes não resultem em uma mudança permanente no uso da terra, eles ainda são uma importante fonte de emissões de carbono e podem <a href="https://www.wri.org/insights/global-trends-forest-fires" target="_blank" rel="noreferrer noopener">levar a ciclos de retroalimentação</a> de emissões crescentes, temperaturas mais quentes, florestas mais secas e mais incêndios.</p>



<h2 class="wp-block-heading">Por que nos concentramos nos trópicos? </h2>



<p>Na <a href="https://gfw.global/3LkdrbQ" target="_blank" rel="noreferrer noopener">análise dos dados da UMD</a>, nós nos concentramos principalmente na perda de cobertura arbórea em florestas primárias tropicais, pois as florestas tropicais sofrem <a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&amp;utm_medium=bitly&amp;utm_source=GFWBlog#how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests" target="_blank" rel="noreferrer noopener">a grande maioria</a> (mais de 96%) do desmatamento mundial, e a perda nessas áreas tem impactos enormes sobre a biodiversidade e o armazenamento de carbono. Mesmo que essas perdas acabem sendo revertidas, <a href="https://www.science.org/doi/10.1126/science.abh3629" target="_blank" rel="noreferrer noopener">levará décadas</a> para que esses habitats e estoques de carbono se recuperem, e poderá ocorrer uma perda permanente de biodiversidade.&nbsp;</p>



<h2 class="wp-block-heading">Como a perda de cobertura arbórea se compara aos dados nacionais oficiais no Brasil e na Indonésia?  </h2>



<p><em>Brasil</em>&nbsp;</p>



<p>O PRODES – o sistema oficial de monitoramento florestal para a Amazônia do Instituto Nacional de Pesquisas Espaciais (INPE) – e os dados da UMD têm diferenças metodológicas e de definição que são importantes de serem compreendidas ao comparar essas duas fontes de dados. O PRODES mede o desmatamento nítido e as perdas florestais antropogênicas por incêndio maiores que 6,25 hectares, enquanto a UMD captura perdas maiores que 0,09 hectare de todas as árvores com altura superior a cinco metros. Ambas são medidas importantes que nos ajudam a entender como as florestas estão mudando, uma vez que o desmatamento, as queimadas e pequenos distúrbios no dossel florestal podem levar a impactos no clima, na biodiversidade e nos serviços ecossistêmicos.&nbsp;</p>



<figure class="wp-block-image size-full"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/06/PT-23-05-04-gfr-technical-blog_UMDvsProdes-PT.png" alt="" class="wp-image-16134" style="aspect-ratio:3/2;object-fit:cover"/></figure>



<p>Os sistemas mostram uma tendência de queda na perda para 2023. Uma diminuição de 22% para o PRODES e 39% para a UMD (42% de redução para a perda não relacionada a incêndios) no bioma da Amazônia, confirmando as notícias positivas para a Amazônia.</p>



<h4 class="wp-block-heading">Dados do PRODES vs. UMD sobre a Amazônia brasileira</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/17601358?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p><em>Indonésia</em></p>



<p>A sobreposição dos dados de perda de floresta primária da UMD/GFW de 2023 com o mapa oficial de cobertura terrestre da Indonésia do Ministério do Meio Ambiente e Florestas (Ministry of Environment and Forestry, MoEF) mostra que 70% ocorreu em classes oficiais de cobertura florestal da Indonésia, com os restantes 30% ocorrendo em áreas de agricultura em terras secas mistas, matagais, arbustos de pântanos e outros tipos de cobertura do solo. Cerca de 144 mil hectares de perda revelaram ser tanto classes oficiais de cobertura florestal da Indonésia quanto ter um tamanho de fragmento maior do que dois hectares. Dos 144 mil hectares, cerca de 15 mil hectares estavam em áreas legalmente classificadas como floresta primária na Indonésia.</p>



<table border="0" cellpadding="0" cellspacing="0">
	<tbody>
		<tr>
			<td style="width:350px;height:39px;">
			<p><strong>Classe do MoEF</strong></p>
			</td>
			<td style="width:550px;height:39px;">
			<p><strong>Porcentagem de perda florestal primária de 2023 da UMD/GFW</strong></p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Floresta primária</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>9%</p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Floresta secundária</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>60%</p>
			</td>
		</tr>
		<tr>
			<td style="width:350px;height:39px;">
			<p>Floresta plantada</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>1%</p>
			</td>
		</tr>
<tr>
			<td style="width:350px;height:39px;">
			<p>Não florestal</p>
			</td>
			<td style="width:550px;height:39px;">
			<p>30%</p>
			</td>
		</tr>
	</tbody>
</table>



<p></p>



<style type="text/css"></style>



<p></p>



<p></p>

Explicação dos dados do Global Forest Watch sobre a perda de cobertura arbórea de 2023

New data shows persistent primary forest loss in 2023. What does the data measure and how does it compare to other official estimates of deforestation?

Global Forest Watch’s 2023 Tree Cover Loss Data Explained | GFW Blog

People

Hear stories from the forest defenders working in the field.

People | Global Forest Watch Blog

indigenous peoples

indigenous peoples | Global Forest Watch Blog

Katie Reytar

Jessica Webb

Peter Veit

<p>The Saamaka, an Afro-descendent tribal community, reside deep in the heart of Suriname’s Amazonian rainforest. They’ve maintained their plots of land and small farms for generations, relying on the bounty of fruits, nuts and medicines the surrounding tropical forest provides. &nbsp;</p>



<p>But lately, their forested territory has seen unwelcome visitors.&nbsp;</p>



<p>A logging company began bulldozing a road right through the Saamaka’s forest in early 2023. While the Suriname government sanctioned the logging concession, it came without the Saamaka’s consent, despite a 2007 ruling from the Inter-American Court of Human Rights requiring their approval for such projects. The road is just the latest threat facing the Saamaka’s land, on top of flooding from hydropower dams and water pollution from nearby mining operations. &nbsp;</p>



<p>&nbsp;“We are losing our very way of life — our food, our water, our land,” said Hugo Jabini, a community leader and spokesperson for the Saamaka. “We can no longer afford to be invisible.”</p>


<div class="wp-block-image">
<figure class="aligncenter size-full"><img decoding="async" loading="lazy" width="455" height="342" src="https://content.globalforestwatch.org/wp-content/uploads/2023/11/HugoPPT3.jpg" alt="Saamaka leader Hugo Jabini leads a kuutu, a meeting of Saamakan communities." class="wp-image-16456"/><figcaption class="wp-element-caption">Hugo Jabini leads a kuutu, a meeting of Saamakan communities.&nbsp;</figcaption></figure></div>


<p>The Saamaka’s story is a common one. Indigenous Peoples and local communities hold some of the most pristine and resource-rich lands in the world — areas highly coveted by mining and logging companies and other profiteers.&nbsp; Land grabs and other threats are especially severe in places where the government does not recognize communities’ land rights, or where anti-deforestation and other laws are weak or poorly enforced. It’s the reason many Indigenous Peoples and local communities often take land monitoring into their own hands — and some are now using digital tools to do it.&nbsp;</p>



<p>Freely available satellite imagery and data from sites like&nbsp;<a href="https://www.globalforestwatch.org/map/">Global Forest Watch</a>&nbsp;and&nbsp;<a href="https://www.landmarkmap.org/data/" target="_blank" rel="noreferrer noopener">LandMark</a>&nbsp;provide near-real-time information that tracks deforestation and land degradation. Indigenous and local communities are increasingly using tools like this to gather evidence that deforestation and degradation are happening on their lands, build their case against illegal activities and take legal action to prevent it from continuing.&nbsp;&nbsp;</p>



<p>Three examples from Suriname, Indonesia and Peru illustrate a growing trend in fighting land rights violations with data.&nbsp;&nbsp;</p>



<h2 class="wp-block-heading">In Suriname, Deforestation Data Reveals Threats to the Saamaka’s Land Rights &nbsp;</h2>



<p>In 2007, the Inter-American Court of Human Rights compelled the government of Suriname to recognize the traditional land rights of the Saamaka and other Indigenous and community groups in Suriname. &nbsp;Yet satellite data from Global Forest Watch shows that since the judgment in 2007, tree cover loss actually increased in the Saamaka’s lands, largely from logging and mining activities. Tree cover loss inside government-granted logging concessions in and surrounding the Saamaka territory totaled 44,700 hectares, a 5.5% loss in tree cover from 2001-2022.&nbsp;&nbsp;</p>



<div class="flourish-embed flourish-chart" data-src="visualisation/15824378?240776"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p>Monthly satellite images from March 2023 to October 2023 show the gradual construction of a road being carved through the heart of the Saamaka territory and leading to a government-granted logging concession. The Saamaka say the road was constructed without their knowledge or consent.&nbsp;</p>



<figure class="wp-block-image size-full"><img decoding="async" loading="lazy" width="1226" height="531" src="https://content.globalforestwatch.org/wp-content/uploads/2023/11/Suriname-logging-road.gif" alt="Gif of road construction progress through the Saamaka’s forested territory. March-October 2023" class="wp-image-16455"/><figcaption class="wp-element-caption">Road construction progresses through the Saamaka’s forested territory. Source: Planet imagery on Global Forest Watch.</figcaption></figure>



<p>Armed with this data and imagery, Jabini and seven other representatives from Suriname’s Indigenous and local communities traveled more than 2,500 miles to Washington, D.C., presenting their evidence to the Inter-American Commission on Human Rights in November 2023. Jabini hopes the commission will reopen its court case against the Suriname government to strengthen traditional land rights.&nbsp;&nbsp;</p>



<p>“We are not here to fight the government; we are just fighting for our rights,” Jabini said.</p>



<h2 class="wp-block-heading">In Indonesia, Indigenous Peoples Use Satellite Data to Track and Report Illegal Timber &nbsp;</h2>



<p>In the eastern-most edges of Indonesia lies an archipelago of 832 islands containing some of the country’s most pristine remaining rainforests. The Aru Islands are home to more than 60,000 Indigenous Aruese, who speak 14 different tribal languages. Their lives are deeply intertwined with the forests they call home, hunting deer, harvesting tree gum, and maintaining a strong spiritual connection with the forests believed to harbor the souls of their ancestors.</p>



<figure class="wp-block-image size-full"><img decoding="async" loading="lazy" width="965" height="651" src="https://content.globalforestwatch.org/wp-content/uploads/2023/11/aru-islands-2.jpg" alt="Indonesia’s Aru Islands" class="wp-image-16458" srcset="https://content.globalforestwatch.org/wp-content/uploads/2023/11/aru-islands-2.jpg 965w, https://content.globalforestwatch.org/wp-content/uploads/2023/11/aru-islands-2-768x518.jpg 768w" sizes="(max-width: 965px) 100vw, 965px" /><figcaption class="wp-element-caption">Indonesia’s Aru Islands hold some of the country’s most pristine rainforests and are home to more than 60,000 Indigenous Aruese. Photo by Forest Watch Indonesia.</figcaption></figure>



<p>As forests in more easily accessible areas of Indonesia such as Java and Borneo have been cleared, increasing pressure for valuable timber species and land for growing crops like sugar has reached the shores of the Aru Islands. Aru Indigenous Peoples and&nbsp;<a href="https://fwi.or.id/en/" rel="noreferrer noopener" target="_blank">Forest Watch Indonesia (FWI),</a>&nbsp;a civil society organization, have been working together to monitor, document and halt illegal logging using satellite data.&nbsp;&nbsp;</p>



<p>Using&nbsp;<a href="https://www.globalforestwatch.org/blog/data-and-research/early-warning-systems-for-deforestation-an-explainer/">deforestation alerts</a>&nbsp;from Global Forest Watch, FWI conducts desk analyses to identify priority locations for forest monitoring. They train Aru Forest Monitors on how to navigate to areas of forest where satellites detected recent clearings, using the&nbsp;<a href="https://forestwatcher.globalforestwatch.org/">Forest Watcher mobile app</a>. They also document evidence of deforestation with the app by taking photos and recording written or audio observations of the causes of the clearing.&nbsp;</p>



<p>Forest Monitors then continually track the extraction of timber from these deforestation sites, oftentimes following the wood to local ports, photographing identifiable information of ships containing illegally harvested timber, and transmitting this information to FWI. The organization then synthesizes the information and passes it to Indonesian authorities, who can then intercept the ships when they dock in Java before reaching international waters. In February 2019, the evidence that Indigenous Forest Monitors documented prompted the Indonesian Ministry of Environment and Forestry to seize 38 containers of illegally harvested timber. High-profile examples like this can serve as a deterrent for future illegal logging.</p>



<h2 class="wp-block-heading">Indigenous Communities in Peru Deploy Technology to Reduce Illegal Deforestation and Strengthen Land Rights</h2>



<p>A&nbsp;<a href="https://www.globalforestwatch.org/blog/people/indigenous-forest-monitors-reduce-deforestation/">two-year study</a>&nbsp;in the Northeastern Amazon region of Peru found that communities who used satellite-based forest monitoring technologies and data significantly reduced deforestation on their lands.</p>



<p>Equipped with deforestation alerts accessed from Forest Watcher and Global Forest Watch, forest monitors representing 36 local communities patrolled their ancestral lands. They documented evidence of encroachment from illicit coca plantations and illegal logging, using satellite imagery, photos, written documentation and drone footage. The monitors presented this information in community assemblies to collectively determine what action to take. Depending on the situation, they then either asked the invaders to leave by showing evidence that they had crossed over into community lands, or submitted documentation to Peruvian law enforcement, who then followed up with official field investigations.&nbsp;</p>


<div class="wp-block-image">
<figure class="aligncenter size-full"><img decoding="async" loading="lazy" width="455" height="607" src="https://content.globalforestwatch.org/wp-content/uploads/2023/11/p_paisa_20150105_113737-scaled-1.jpg" alt="" class="wp-image-16459"/><figcaption class="wp-element-caption">A community monitor in Remanso del Amazonas identifies early warning signs of deforestation. Photo by Beder Olortegui.</figcaption></figure></div>


<p>Ultimately, the communities that used forest monitoring technologies reduced deforestation in their territories by 52% in the first year and an additional 21% in the second year, as compared to a similar number of communities in a control group. The communities using monitoring technologies also reported higher satisfaction with forest governance.</p>



<p>Since the study concluded, communities with experience using monitoring technologies are now training communities in the control group to do the same. The communities who succeeded in reducing deforestation have also received direct payments from the&nbsp;<a href="https://rainforestfoundation.org/our-work/approaches/rainforest-alert/" rel="noreferrer noopener" target="_blank">Rainforest Foundation US’s Rainforest Alert program</a>, a payment system for keeping forests conserved.</p>



<p>Most recently, some communities allocated that funding towards strengthening and expanding legal recognition of their ancestral lands. Community land titling in Peru is a slow and bureaucratic process. With the funds, communities were able to georeference the limits of their lands; contract soil analysts, lawyers and Geographic Information System (GIS) specialists; and petition the government for land titles. They succeeded in obtaining titles to their territories in just 10 months – a process that normally takes years or even decades.</p>



<h2 class="wp-block-heading">Open Data Supports Communities, but More Secure Land Rights Are Essential&nbsp;</h2>



<p>Satellite data and imagery are critical tools because they provide transparent, independent evidence of suspected violations. They allow communities to rapidly monitor and document forest disturbances over large areas and long periods of time at low cost, thus reducing expensive field patrols. Moreover, much of the monitoring and documentation can be done remotely from the safety of a laptop, without risking face-to-face confrontation with encroachers. This is especially important because environmental defenders face growing threats. In 2022, at least&nbsp;<a href="https://www.globalwitness.org/en/campaigns/environmental-activists/standing-firm/" target="_blank" rel="noreferrer noopener">177 environmental defenders were killed</a>, 36% of whom were Indigenous.&nbsp;</p>



<p>Given the power of satellite data to help communities protect their land rights, data creators and managers need to ensure that communities continue to have access to open data and technology. Communities also need support in data use, including tailored&nbsp;<a href="https://www.globalforestwatch.org/grants-and-fellowships/about/">training resources</a>,&nbsp;<a href="https://www.globalforestwatch.org/grants-and-fellowships/about/" target="_blank" rel="noreferrer noopener">financial support</a>&nbsp;and facilitating the&nbsp;<a href="https://www.globalforestwatch.org/help/map/webinars/webinar-strengthening-the-role-of-indigenous-youth-in-forest-protection-perspectives-from-latin-america/">sharing of experiences</a>.</p>



<p>However, while satellite data helps communities document and take action on illegal deforestation and related infractions, governments ultimately need to recognize and enforce Indigenous Peoples’ and communities’ land rights. Stronger and more secure land rights — with higher penalties for infractions — are more likely to deter illegal activity from occurring in the first place, or at least ensure a swift and decisive end to violations.</p>



<p>Indigenous Peoples and local communities have stewarded their lands and natural resources for centuries. Though more secure land rights are ultimately needed, satellite data can help them manage threats to their lands now and conserve their homes for generations to come.</p>



<hr> 

<div class="attribute" style="”padding-bottom:" 10px;”="">
 

Originally appeared on <a href="https://www.wri.org/insights/indigenous-peoples-local-communities-use-satellite-data-deforestation" target="_blank" rel="noopener noreferrer">WRI Insights</a>. 
 

</div> 

 

<hr>



<style type="text/css"></style>

Indigenous Peoples and Local Communities Are Using Satellite Data to Fight Deforestation

<p>The Saamaka, an Afro-descendent tribal community, reside deep in the heart of Suriname’s Amazonian rainforest. They’ve maintained their plots of land and small farms for generations, relying on the bounty of fruits, nuts and medicines the surrounding tropical forest provides. &nbsp;</p>



<p>But lately, their forested territory has seen unwelcome visitors.&nbsp;</p>



<p>A logging company began bulldozing a road right through the Saamaka’s forest in early 2023. While the Suriname government sanctioned the logging concession, it came without the Saamaka’s consent, despite a 2007 ruling from the Inter-American Court of Human Rights requiring their approval for such projects. The road is just the latest threat facing the Saamaka’s land, on top of flooding from hydropower dams and water pollution from nearby mining operations. &nbsp;</p>



<p>&nbsp;“We are losing our very way of life — our food, our water, our land,” said Hugo Jabini, a community leader and spokesperson for the Saamaka. “We can no longer afford to be invisible.”</p>



<figure class="wp-block-image aligncenter size-full"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/11/HugoPPT3.jpg" alt="Saamaka leader Hugo Jabini leads a kuutu, a meeting of Saamakan communities." class="wp-image-16456"/><figcaption class="wp-element-caption">Hugo Jabini leads a kuutu, a meeting of Saamakan communities.&nbsp;</figcaption></figure>



<p>The Saamaka’s story is a common one. Indigenous Peoples and local communities hold some of the most pristine and resource-rich lands in the world — areas highly coveted by mining and logging companies and other profiteers.&nbsp; Land grabs and other threats are especially severe in places where the government does not recognize communities’ land rights, or where anti-deforestation and other laws are weak or poorly enforced. It’s the reason many Indigenous Peoples and local communities often take land monitoring into their own hands — and some are now using digital tools to do it.&nbsp;</p>



<p>Freely available satellite imagery and data from sites like&nbsp;<a href="https://www.globalforestwatch.org/map/">Global Forest Watch</a>&nbsp;and&nbsp;<a href="https://www.landmarkmap.org/data/" target="_blank" rel="noreferrer noopener">LandMark</a>&nbsp;provide near-real-time information that tracks deforestation and land degradation. Indigenous and local communities are increasingly using tools like this to gather evidence that deforestation and degradation are happening on their lands, build their case against illegal activities and take legal action to prevent it from continuing.&nbsp;&nbsp;</p>



<p>Three examples from Suriname, Indonesia and Peru illustrate a growing trend in fighting land rights violations with data.&nbsp;&nbsp;</p>



<h2 class="wp-block-heading">In Suriname, Deforestation Data Reveals Threats to the Saamaka’s Land Rights &nbsp;</h2>



<p>In 2007, the Inter-American Court of Human Rights compelled the government of Suriname to recognize the traditional land rights of the Saamaka and other Indigenous and community groups in Suriname. &nbsp;Yet satellite data from Global Forest Watch shows that since the judgment in 2007, tree cover loss actually increased in the Saamaka’s lands, largely from logging and mining activities. Tree cover loss inside government-granted logging concessions in and surrounding the Saamaka territory totaled 44,700 hectares, a 5.5% loss in tree cover from 2001-2022.&nbsp;&nbsp;</p>



<div class="flourish-embed flourish-chart" data-src="visualisation/15824378?240776"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p>Monthly satellite images from March 2023 to October 2023 show the gradual construction of a road being carved through the heart of the Saamaka territory and leading to a government-granted logging concession. The Saamaka say the road was constructed without their knowledge or consent.&nbsp;</p>



<figure class="wp-block-image size-full"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/11/Suriname-logging-road.gif" alt="Gif of road construction progress through the Saamaka’s forested territory. March-October 2023" class="wp-image-16455"/><figcaption class="wp-element-caption">Road construction progresses through the Saamaka’s forested territory. Source: Planet imagery on Global Forest Watch.</figcaption></figure>



<p>Armed with this data and imagery, Jabini and seven other representatives from Suriname’s Indigenous and local communities traveled more than 2,500 miles to Washington, D.C., presenting their evidence to the Inter-American Commission on Human Rights in November 2023. Jabini hopes the commission will reopen its court case against the Suriname government to strengthen traditional land rights.&nbsp;&nbsp;</p>



<p>“We are not here to fight the government; we are just fighting for our rights,” Jabini said.</p>



<h2 class="wp-block-heading">In Indonesia, Indigenous Peoples Use Satellite Data to Track and Report Illegal Timber &nbsp;</h2>



<p>In the eastern-most edges of Indonesia lies an archipelago of 832 islands containing some of the country’s most pristine remaining rainforests. The Aru Islands are home to more than 60,000 Indigenous Aruese, who speak 14 different tribal languages. Their lives are deeply intertwined with the forests they call home, hunting deer, harvesting tree gum, and maintaining a strong spiritual connection with the forests believed to harbor the souls of their ancestors.</p>



<figure class="wp-block-image size-full"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/11/aru-islands-2.jpg" alt="Indonesia’s Aru Islands" class="wp-image-16458"/><figcaption class="wp-element-caption">Indonesia’s Aru Islands hold some of the country’s most pristine rainforests and are home to more than 60,000 Indigenous Aruese. Photo by Forest Watch Indonesia.</figcaption></figure>



<p>As forests in more easily accessible areas of Indonesia such as Java and Borneo have been cleared, increasing pressure for valuable timber species and land for growing crops like sugar has reached the shores of the Aru Islands. Aru Indigenous Peoples and&nbsp;<a href="https://fwi.or.id/en/" rel="noreferrer noopener" target="_blank">Forest Watch Indonesia (FWI),</a>&nbsp;a civil society organization, have been working together to monitor, document and halt illegal logging using satellite data.&nbsp;&nbsp;</p>



<p>Using&nbsp;<a href="https://www.globalforestwatch.org/blog/data-and-research/early-warning-systems-for-deforestation-an-explainer/">deforestation alerts</a>&nbsp;from Global Forest Watch, FWI conducts desk analyses to identify priority locations for forest monitoring. They train Aru Forest Monitors on how to navigate to areas of forest where satellites detected recent clearings, using the&nbsp;<a href="https://forestwatcher.globalforestwatch.org/">Forest Watcher mobile app</a>. They also document evidence of deforestation with the app by taking photos and recording written or audio observations of the causes of the clearing.&nbsp;</p>



<p>Forest Monitors then continually track the extraction of timber from these deforestation sites, oftentimes following the wood to local ports, photographing identifiable information of ships containing illegally harvested timber, and transmitting this information to FWI. The organization then synthesizes the information and passes it to Indonesian authorities, who can then intercept the ships when they dock in Java before reaching international waters. In February 2019, the evidence that Indigenous Forest Monitors documented prompted the Indonesian Ministry of Environment and Forestry to seize 38 containers of illegally harvested timber. High-profile examples like this can serve as a deterrent for future illegal logging.</p>



<h2 class="wp-block-heading">Indigenous Communities in Peru Deploy Technology to Reduce Illegal Deforestation and Strengthen Land Rights</h2>



<p>A&nbsp;<a href="https://www.globalforestwatch.org/blog/people/indigenous-forest-monitors-reduce-deforestation/">two-year study</a>&nbsp;in the Northeastern Amazon region of Peru found that communities who used satellite-based forest monitoring technologies and data significantly reduced deforestation on their lands.</p>



<p>Equipped with deforestation alerts accessed from Forest Watcher and Global Forest Watch, forest monitors representing 36 local communities patrolled their ancestral lands. They documented evidence of encroachment from illicit coca plantations and illegal logging, using satellite imagery, photos, written documentation and drone footage. The monitors presented this information in community assemblies to collectively determine what action to take. Depending on the situation, they then either asked the invaders to leave by showing evidence that they had crossed over into community lands, or submitted documentation to Peruvian law enforcement, who then followed up with official field investigations.&nbsp;</p>



<figure class="wp-block-image aligncenter size-full"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/11/p_paisa_20150105_113737-scaled-1.jpg" alt="" class="wp-image-16459"/><figcaption class="wp-element-caption">A community monitor in Remanso del Amazonas identifies early warning signs of deforestation. Photo by Beder Olortegui.</figcaption></figure>



<p>Ultimately, the communities that used forest monitoring technologies reduced deforestation in their territories by 52% in the first year and an additional 21% in the second year, as compared to a similar number of communities in a control group. The communities using monitoring technologies also reported higher satisfaction with forest governance.</p>



<p>Since the study concluded, communities with experience using monitoring technologies are now training communities in the control group to do the same. The communities who succeeded in reducing deforestation have also received direct payments from the&nbsp;<a href="https://rainforestfoundation.org/our-work/approaches/rainforest-alert/" rel="noreferrer noopener" target="_blank">Rainforest Foundation US’s Rainforest Alert program</a>, a payment system for keeping forests conserved.</p>



<p>Most recently, some communities allocated that funding towards strengthening and expanding legal recognition of their ancestral lands. Community land titling in Peru is a slow and bureaucratic process. With the funds, communities were able to georeference the limits of their lands; contract soil analysts, lawyers and Geographic Information System (GIS) specialists; and petition the government for land titles. They succeeded in obtaining titles to their territories in just 10 months – a process that normally takes years or even decades.</p>



<h2 class="wp-block-heading">Open Data Supports Communities, but More Secure Land Rights Are Essential&nbsp;</h2>



<p>Satellite data and imagery are critical tools because they provide transparent, independent evidence of suspected violations. They allow communities to rapidly monitor and document forest disturbances over large areas and long periods of time at low cost, thus reducing expensive field patrols. Moreover, much of the monitoring and documentation can be done remotely from the safety of a laptop, without risking face-to-face confrontation with encroachers. This is especially important because environmental defenders face growing threats. In 2022, at least&nbsp;<a href="https://www.globalwitness.org/en/campaigns/environmental-activists/standing-firm/" target="_blank" rel="noreferrer noopener">177 environmental defenders were killed</a>, 36% of whom were Indigenous.&nbsp;</p>



<p>Given the power of satellite data to help communities protect their land rights, data creators and managers need to ensure that communities continue to have access to open data and technology. Communities also need support in data use, including tailored&nbsp;<a href="https://www.globalforestwatch.org/grants-and-fellowships/about/">training resources</a>,&nbsp;<a href="https://www.globalforestwatch.org/grants-and-fellowships/about/" target="_blank" rel="noreferrer noopener">financial support</a>&nbsp;and facilitating the&nbsp;<a href="https://www.globalforestwatch.org/help/map/webinars/webinar-strengthening-the-role-of-indigenous-youth-in-forest-protection-perspectives-from-latin-america/">sharing of experiences</a>.</p>



<p>However, while satellite data helps communities document and take action on illegal deforestation and related infractions, governments ultimately need to recognize and enforce Indigenous Peoples’ and communities’ land rights. Stronger and more secure land rights — with higher penalties for infractions — are more likely to deter illegal activity from occurring in the first place, or at least ensure a swift and decisive end to violations.</p>



<p>Indigenous Peoples and local communities have stewarded their lands and natural resources for centuries. Though more secure land rights are ultimately needed, satellite data can help them manage threats to their lands now and conserve their homes for generations to come.</p>



<hr> 

<div class="attribute" style="”padding-bottom:" 10px;”="">
 

Originally appeared on <a href="https://www.wri.org/insights/indigenous-peoples-local-communities-use-satellite-data-deforestation" target="_blank" rel="noopener noreferrer">WRI Insights</a>. 
 

</div> 

 

<hr>



<style type="text/css"></style>

Communities are increasingly using tools like GFW to gather evidence of deforestation and degradation on their lands and take legal action.  

Communities Are Using Satellite Data to Fight Deforestation

An aerial shot shows the contrast between forest and agricultural landscapes near Rio Branco, Acre, Brazil.

Tropical Primary Forest Loss Worsened in 2022, Despite International Commitments to End Deforestation

La perte de forêts primaires tropicales s’est aggravée en 2022, malgré les engagements internationaux visant à mettre fin à la déforestation

Perda de florestas tropicais primárias piora em 2022 apesar dos compromissos internacionais para acabar com o desmatamento

Empeoró La Pérdida De Bosques Tropicales Primarios en 2022, A Pesar De Los Compromisos Internacionales Contra La Deforestación

Tingkat Kehilangan Hutan Primer Tropis pada tahun 2022 Memburuk, Terlepas dari Komitmen Global untuk Mengakhiri Deforestasi

Tropical primary forest loss was 10% higher in 2022 than in 2021 despite international commitments, according to new data on GFW.

Tropical Primary Forest Loss Worsened in 2022 | GFW Blog

<p>New tree cover loss data from the University of Maryland (UMD) and available on Global Forest Watch (GFW) shows high rates of primary forest loss in 2022 despite <a href="https://gfw.global/3L1xFpo" target="_blank" rel="noreferrer noopener">political ambitions to curb loss</a>. What does the data measure, what is different this year and how does it compare to other official estimates of deforestation? Here is what you should know about the new data.</p>



<h2 class="wp-block-heading">What tree cover loss measures and how this differs from deforestation&nbsp;</h2>



<p>The tree cover loss data from UMD captures the annual loss of all trees taller than five meters for the calendar years between 2001 and 2022. Because optical satellite data collection can often be obscured by cloud cover in the humid tropics, some late-in-the year loss can be picked up in the next year following persistent cloud cover. This is a common occurrence in many tropical countries where cloud cover is prevalent.</p>



<p>The tree cover loss data includes the loss of trees in natural forests as well as in plantations and tree crops. Loss of this tree cover can be due to human or natural causes and can be permanent or temporary. Deforestation differs in that it only refers to a human-caused, permanent change from forest to another land use. Some forms of tree cover loss, such as conversion of a natural forest to agricultural land, are considered deforestation, while other forms of tree cover loss, such as timber harvesting in plantation forests or natural disturbances, are not. <a href="https://research.wri.org/gfr/key-terms-definitions" target="_blank" rel="noreferrer noopener">Read more about the differences here</a>.</p>



<p>A new tool available on the Global Forest Review, the <a href="https://gfw.global/3L1xFpo" target="_blank" rel="noreferrer noopener">Targets Tracker</a>, enables us to use tree cover loss data to measure progress toward deforestation commitments on a global level by using a coarse resolution <a href="https://research.wri.org/gfr/forest-targets-tracker#how-is-deforestation-estimated" target="_blank" rel="noreferrer noopener">proxy of deforestation</a>. To do this, the Targets Tracker identifies losses that are likely to represent deforestation in and outside of the tropics by only including tree cover loss from certain drivers. Specifically, it includes the expansion of small-scale agriculture into humid tropical primary forests, as well as all loss in commodity-driven deforestation and urbanization. It excludes temporary losses, such as from fires or forestry activities.</p>



<h2 class="wp-block-heading">Why we focus on the tropics&nbsp;</h2>



<p>In our <a href="https://gfw.global/3LkdrbQ" target="_blank" rel="noreferrer noopener">analysis of the UMD data</a>, we largely focus on tree cover loss in tropical primary forests because tropical forests experience <a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&utm_medium=bitly&utm_source=GFWBlog#how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests" target="_blank" rel="noreferrer noopener">the vast majority</a> (over 96%) of the world’s deforestation, and loss in those areas has huge impacts on biodiversity and carbon storage. Even if these losses are eventually reversed, it will <a href="https://www.science.org/doi/10.1126/science.abh3629" target="_blank" rel="noreferrer noopener">take decades</a> for these habitats and carbon stocks to recover, and permanent biodiversity loss may occur.</p>



<h2 class="wp-block-heading">How does tree cover loss compare with official country data in Brazil and Indonesia?&nbsp;&nbsp;</h2>



<h3 class="wp-block-heading">Brazil&nbsp;</h3>



<p>PRODES — the official forest monitoring system for the Amazon from Brazil’s National Institute for Space Research (INPE) — and the UMD data have methodological and definitional differences that are important to understand when comparing these two data sources. PRODES measures clear cut deforestation and anthropogenic forest losses by fire larger than 6.25 hectares, whereas UMD captures loss larger than 0.09 hectares of all trees taller than five meters. These are both important measures that help us understand how forests are changing since deforestation, forest fires and small disturbances to forest canopy can all lead to impacts on climate, biodiversity and ecosystem services.</p>



<figure class="wp-block-image size-large"><img decoding="async" loading="lazy" width="1348" height="1042" src="https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN.png" alt="UMD and PRODES differences explained" class="wp-image-15866" srcset="https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN.png 1348w, https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN-1024x792.png 1024w, https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN-768x594.png 768w" sizes="(max-width: 1348px) 100vw, 1348px" /></figure>



<p>Separating out tree cover loss from fire in the UMD data has resulted in a much closer match between these two data sets. However, the two seem to show different trends for 2022 in the Brazilian Amazon: PRODES shows an 11% decrease in deforestation while UMD shows a 21% increase in non-fire primary forest loss.</p>



<h4 class="wp-block-heading">PRODES vs. UMD data in the Brazilian Amazon&nbsp;</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/13629875?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p>The reason for this discrepancy may be the difference in reporting periods. PRODES uses the period between August and July (in this case, August 2021 to July 2022), while UMD uses the calendar year (January to December 2022). The Brazilian government’s DETER alerts — which track deforestation on a daily basis — suggest that deforestation rates increased in the second half of 2022 compared to previous years. DETER alerts detected a 2% decrease in deforestation between August 2021 and July 2022 compared to August 2020 to July 2021, but detected a 25% increase between the calendar year 2021 and 2022.</p>



<h4 class="wp-block-heading">DETER Alerts in the Brazilian Amazon&nbsp;</h4>



<div class="flourish-embed" data-src="story/1907320?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<h3 class="wp-block-heading">Indonesia</h3>



<p>This year, WRI and Indonesia’s Ministry of Environment and Forestry (MoEF) are working together on a joint analysis to better understand the UMD/GFW primary forest loss findings through the MoEF’s data and definitions. Overlaying the UMD/GFW primary forest loss data for 2022 with the MoEF land cover map shows that 67% occurs within Indonesia’s official forest cover classes, with the remaining 33% taking place in mixed dry land agriculture, shrub, swamp shrub and other land cover types. Approximately 107 thousand hectares of loss was found to be both within Indonesia’s official forest land cover classes and with a patch size larger than two hectares. Of these 107 thousand hectares, about 12 thousand hectares were within areas legally classified as primary forest in Indonesia. The upcoming joint analysis will improve our understanding of these issues.</p>



<figure class="wp-block-image size-large"><img decoding="async" loading="lazy" width="1348" height="1438" src="https://content.globalforestwatch.org/wp-content/uploads/2023/06/23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded.png" alt="Table showing UMD/GFW primary forest loss in hectares per MoEF forest classes in 2022" class="wp-image-16098" srcset="https://content.globalforestwatch.org/wp-content/uploads/2023/06/23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded.png 1348w, https://content.globalforestwatch.org/wp-content/uploads/2023/06/23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded-960x1024.png 960w, https://content.globalforestwatch.org/wp-content/uploads/2023/06/23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded-768x819.png 768w" sizes="(max-width: 1348px) 100vw, 1348px" /></figure>



<h2 class="wp-block-heading">Data improvements over time result in inconsistencies in the data before 2015&nbsp;&nbsp;</h2>



<p>Algorithm adjustments and better satellite data have <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/">improved the tree cover loss data set over time</a>. Changes to the tree cover loss detection algorithm for the years 2011-2014 and 2015 onwards and the incorporation of Landsat 8 data starting in 2013 make it easier to detect smaller scale changes, such as loss due to fires, selective logging and shifting agriculture.&nbsp;Variations in satellite image availability also mean there are inconsistencies with the quality and number of images available to capture data each year.</p>



<p>To address these issues, we:&nbsp;&nbsp;</p>



<ul><li>Focus our analysis on trends post-2015.&nbsp;&nbsp;</li><li>Assess the three-year moving average to interpret longer-term trends.&nbsp;</li><li>Disregard the increase in loss post 2012 in places dominated by small-scale agriculture, like Central Africa. This increase is likely due to data improvements that better detect the loss later in the time series.&nbsp;&nbsp;</li></ul>



<h2 class="wp-block-heading">How will the data change over time?&nbsp;</h2>



<p>UMD is working on improvements to the tree cover extent and change data, which will result in changes to the data in the coming years. The new data will track the annual extent of tree cover and offer annual loss and gain information using a consistent algorithm across the time series. It will also track multiple gain and loss events in a single location, allowing us to better track dynamic areas of forest change such as forestry activities and wildfires. Incorporating annual gain and repeated loss events will provide a more complete picture of global forest change.</p>



<style type="text/css"></style>

Global Forest Watch’s 2022 Tree Cover Loss Data Explained

<p>New tree cover loss data from the University of Maryland (UMD) and available on Global Forest Watch (GFW) shows high rates of primary forest loss in 2022 despite <a href="https://gfw.global/3L1xFpo" target="_blank" rel="noreferrer noopener">political ambitions to curb loss</a>. What does the data measure, what is different this year and how does it compare to other official estimates of deforestation? Here is what you should know about the new data.</p>



<h2>What tree cover loss measures and how this differs from deforestation&nbsp;</h2>



<p>The tree cover loss data from UMD captures the annual loss of all trees taller than five meters for the calendar years between 2001 and 2022. Because optical satellite data collection can often be obscured by cloud cover in the humid tropics, some late-in-the year loss can be picked up in the next year following persistent cloud cover. This is a common occurrence in many tropical countries where cloud cover is prevalent.</p>



<p>The tree cover loss data includes the loss of trees in natural forests as well as in plantations and tree crops. Loss of this tree cover can be due to human or natural causes and can be permanent or temporary. Deforestation differs in that it only refers to a human-caused, permanent change from forest to another land use. Some forms of tree cover loss, such as conversion of a natural forest to agricultural land, are considered deforestation, while other forms of tree cover loss, such as timber harvesting in plantation forests or natural disturbances, are not. <a href="https://research.wri.org/gfr/key-terms-definitions" target="_blank" rel="noreferrer noopener">Read more about the differences here</a>.</p>



<p>A new tool available on the Global Forest Review, the <a href="https://gfw.global/3L1xFpo" target="_blank" rel="noreferrer noopener">Targets Tracker</a>, enables us to use tree cover loss data to measure progress toward deforestation commitments on a global level by using a coarse resolution <a href="https://research.wri.org/gfr/forest-targets-tracker#how-is-deforestation-estimated" target="_blank" rel="noreferrer noopener">proxy of deforestation</a>. To do this, the Targets Tracker identifies losses that are likely to represent deforestation in and outside of the tropics by only including tree cover loss from certain drivers. Specifically, it includes the expansion of small-scale agriculture into humid tropical primary forests, as well as all loss in commodity-driven deforestation and urbanization. It excludes temporary losses, such as from fires or forestry activities.</p>



<h2>Why we focus on the tropics&nbsp;</h2>



<p>In our <a href="https://gfw.global/3LkdrbQ" target="_blank" rel="noreferrer noopener">analysis of the UMD data</a>, we largely focus on tree cover loss in tropical primary forests because tropical forests experience <a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&amp;utm_medium=bitly&amp;utm_source=GFWBlog#how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests" target="_blank" rel="noreferrer noopener">the vast majority</a> (over 96%) of the world's deforestation, and loss in those areas has huge impacts on biodiversity and carbon storage. Even if these losses are eventually reversed, it will <a href="https://www.science.org/doi/10.1126/science.abh3629" target="_blank" rel="noreferrer noopener">take decades</a> for these habitats and carbon stocks to recover, and permanent biodiversity loss may occur.</p>



<h2>How does tree cover loss compare with official country data in Brazil and Indonesia?&nbsp;&nbsp;</h2>



<h3>Brazil&nbsp;</h3>



<p>PRODES — the official forest monitoring system for the Amazon from Brazil’s National Institute for Space Research (INPE) — and the UMD data have methodological and definitional differences that are important to understand when comparing these two data sources. PRODES measures clear cut deforestation and anthropogenic forest losses by fire larger than 6.25 hectares, whereas UMD captures loss larger than 0.09 hectares of all trees taller than five meters. These are both important measures that help us understand how forests are changing since deforestation, forest fires and small disturbances to forest canopy can all lead to impacts on climate, biodiversity and ecosystem services.</p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/05/23-05-04-gfr-technical-blog_UMDvsProdes-EN.png" alt="UMD and PRODES differences explained" class="wp-image-15866"/></figure>



<p>Separating out tree cover loss from fire in the UMD data has resulted in a much closer match between these two data sets. However, the two seem to show different trends for 2022 in the Brazilian Amazon: PRODES shows an 11% decrease in deforestation while UMD shows a 21% increase in non-fire primary forest loss.</p>



<h4>PRODES vs. UMD data in the Brazilian Amazon&nbsp;</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/13629875?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p>The reason for this discrepancy may be the difference in reporting periods. PRODES uses the period between August and July (in this case, August 2021 to July 2022), while UMD uses the calendar year (January to December 2022). The Brazilian government’s DETER alerts — which track deforestation on a daily basis — suggest that deforestation rates increased in the second half of 2022 compared to previous years. DETER alerts detected a 2% decrease in deforestation between August 2021 and July 2022 compared to August 2020 to July 2021, but detected a 25% increase between the calendar year 2021 and 2022.</p>



<h4>DETER Alerts in the Brazilian Amazon&nbsp;</h4>



<div class="flourish-embed" data-src="story/1907320?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<h3>Indonesia</h3>



<p>This year, WRI and Indonesia’s Ministry of Environment and Forestry (MoEF) are working together on a joint analysis to better understand the UMD/GFW primary forest loss findings through the MoEF’s data and definitions. Overlaying the UMD/GFW primary forest loss data for 2022 with the MoEF land cover map shows that 67% occurs within Indonesia’s official forest cover classes, with the remaining 33% taking place in mixed dry land agriculture, shrub, swamp shrub and other land cover types. Approximately 107 thousand hectares of loss was found to be both within Indonesia’s official forest land cover classes and with a patch size larger than two hectares. Of these 107 thousand hectares, about 12 thousand hectares were within areas legally classified as primary forest in Indonesia. The upcoming joint analysis will improve our understanding of these issues.</p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/06/23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded.png" alt="Table showing UMD/GFW primary forest loss in hectares per MoEF forest classes in 2022" class="wp-image-16098"/></figure>



<h2>Data improvements over time result in inconsistencies in the data before 2015&nbsp;&nbsp;</h2>



<p>Algorithm adjustments and better satellite data have <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/">improved the tree cover loss data set over time</a>. Changes to the tree cover loss detection algorithm for the years 2011-2014 and 2015 onwards and the incorporation of Landsat 8 data starting in 2013 make it easier to detect smaller scale changes, such as loss due to fires, selective logging and shifting agriculture.&nbsp;Variations in satellite image availability also mean there are inconsistencies with the quality and number of images available to capture data each year.</p>



<p>To address these issues, we:&nbsp;&nbsp;</p>



<ul><li>Focus our analysis on trends post-2015.&nbsp;&nbsp;</li><li>Assess the three-year moving average to interpret longer-term trends.&nbsp;</li><li>Disregard the increase in loss post 2012 in places dominated by small-scale agriculture, like Central Africa. This increase is likely due to data improvements that better detect the loss later in the time series.&nbsp;&nbsp;</li></ul>



<h2>How will the data change over time?&nbsp;</h2>



<p>UMD is working on improvements to the tree cover extent and change data, which will result in changes to the data in the coming years. The new data will track the annual extent of tree cover and offer annual loss and gain information using a consistent algorithm across the time series. It will also track multiple gain and loss events in a single location, allowing us to better track dynamic areas of forest change such as forestry activities and wildfires. Incorporating annual gain and repeated loss events will provide a more complete picture of global forest change.</p>



<style type="text/css"></style>

<p>De nouvelles données sur la perte de la couverture arborée, fournies par l’Université du Maryland (UMD) et disponibles sur le site Global Forest Watch (GFW), révèlent des taux élevés de perte de forêts primaires en 2022, en dépit des&nbsp;<a href="https://gfw.global/3L1xFpo">ambitions politiques visant à freiner cette perte</a>. Qu’est-ce qui est mesuré par ces données, quelles sont les différences cette années et comment se comparent-elles aux autres estimations officielles de la déforestation&nbsp;? Voici ce qu'il faut savoir sur ces nouvelles données.</p>



<h2>Qu’est-ce que la perte du couverture arborée et en quoi cela diffère-t-elle de la déforestation ?</h2>



<p>Les données recueillies par l'UMD concernant la perte du couvert arboré comprennent les pertes annuelles de tous les arbres d'une hauteur supérieure à cinq mètres entre 2001 et 2022. Dans les régions tropicales humides, où la couverture nuageuse est fréquente, la collecte de données par satellite optique peut être entravée, ce qui peut entraîner une récupération partielle des pertes en fin d'année lorsqu'il y a une couverture nuageuse persistante. Ce phénomène est courant dans de nombreux pays tropicaux où les nuages sont prédominants.</p>



<p>Ces données sur la perte du couvert arboré englobent les pertes d'arbres dans les forêts naturelles, les plantations et les cultures arboricoles. La perte de ce couvert arboré peut être attribuée à des facteurs humains ou naturels et peut être permanente ou temporaire. La déforestation se distingue de la perte de couvert arboré dans le sens où elle se réfère uniquement à un changement permanent de la forêt vers une autre utilisation des terres, causé par l'homme. Certaines formes de perte de couvert arboré, telles que la conversion d'une forêt naturelle en terres agricoles, sont considérées comme de la déforestation, tandis que d'autres formes de perte de couvert arboré, comme l'exploitation du bois dans les plantations forestières ou les perturbations naturelles, ne sont pas considérées comme de la déforestation. Pour en savoir plus sur les différences, vous pouvez cliquer <a href="https://research.wri.org/gfr/key-terms-definitions">ici</a>.</p>



<p>Le <a href="https://research.wri.org/gfr/forest-targets-tracker?utm_campaign=treecoverloss2022&amp;utm_medium=bitly&amp;utm_source=GFWBlog">Targets Tracker</a>, un nouvel outil disponible sur la Revue mondiale des forêts, permet de mesurer les progrès réalisés dans la lutte contre la déforestation à l'échelle mondiale en utilisant des données sur la perte du couvert arboré. Cet outil utilise une estimation approximative de la <a href="https://research.wri.org/gfr/forest-targets-tracker#how-is-deforestation-estimated">déforestation à résolution grossière</a> en identifiant les pertes qui pourraient représenter la déforestation, tant dans les régions tropicales que hors de celles-ci. Il se concentre uniquement sur les pertes du couvert arboré liées à certains facteurs, tels que l'expansion de l'agriculture à petite échelle dans les forêts tropicales humides primaires, ainsi que les pertes causées par la déforestation et l'urbanisation liées aux produits de base. Les pertes temporaires, telles que celles causées par les incendies ou les activités forestières, sont exclues de cette évaluation.</p>



<h2>Pourquoi nous concentrons-nous sur les tropiques ?</h2>



<p>Dans notre&nbsp;<a href="https://gfw.global/3LkdrbQ">analyse des données de l’UMD</a>, nous nous concentrons principalement sur la&nbsp;perte du couvert arboré dans les forêts primaires tropicales. Cette focalisation est justifiée par le fait que les forêts tropicales représentent la&nbsp;<a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&amp;utm_medium=bitly&amp;utm_source=GFWBlog#how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests">grande majorité</a>&nbsp;(plus de 96 %) des zones touchées par la déforestation&nbsp;dans le monde. La perte de ces forêts a un impact considérable sur la biodiversité et le stockage du carbone. Même si ces pertes sont finalement inversées, il faudra&nbsp;<a href="https://www.science.org/doi/10.1126/science.abh3629">des décennies</a>&nbsp;pour que ces habitats et ces stocks de carbone se reconstituent. De plus, une perte permanente de biodiversité est susceptible de se produire.</p>



<h2>Comment la perte du couvert arboré se compare-t-elle aux données officielles du Brésil et de l’Indonésie ?</h2>



<h3>Brésil</h3>



<p>Il est essentiel de comprendre les différences méthodologiques et définitionnelles entre PRODES, le système officiel de surveillance des forêts amazoniennes de l'Institut national brésilien de recherche spatiale (INPE), et les données de l'UMD lors de la comparaison de ces deux sources d'information. PRODES mesure la déforestation par coupes rases et les pertes forestières anthropiques résultant d'incendies sur des surfaces de plus de 6,25 hectares. En revanche, l'UMD enregistre les pertes du couvert forestier de plus de 0,09 hectare, incluant tous les arbres de plus de cinq mètres de haut. Ces deux mesures sont importantes pour comprendre l'évolution des forêts, car la déforestation, les incendies de forêt et les perturbations mineures du couvert forestier peuvent tous avoir un impact sur le climat, la biodiversité et les services écosystémiques.</p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/07/FR-23-05-04-gfr-technical-blog_UMDvsProdes-FR.png" alt="Explication des différences UMD et PRODES" class="wp-image-16148"/></figure>



<p>Lorsqu'on isole la perte du couvert arboré causée par les incendies dans les données de l'UMD, on observe une correspondance beaucoup plus étroite entre ces deux ensembles de données. Cependant, pour l'année 2022 en Amazonie brésilienne, les deux sources semblent présenter des tendances différentes : PRODES indique une diminution de 11 % de la déforestation, tandis que l'UMD révèle une augmentation de 21 % de la perte de forêt primaire non liée aux incendies.</p>



<h4>Données PRODES vs UMD dans l’Amazonie brésilienne&nbsp;</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/14261358?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p>La différence entre les données PRODES et UMD dans la région amazonienne du Brésil peut s'expliquer par l'écart entre les périodes de déclaration utilisées par ces sources. PRODES utilise la période d'août à juillet (par exemple, août 2021 à juillet 2022), tandis que l'UMD se base sur l'année civile (janvier à décembre 2022). Les alertes DETER du gouvernement brésilien, qui surveillent la déforestation quotidiennement, suggèrent que les taux de déforestation ont augmenté au cours du second semestre 2022 par rapport aux années précédentes. Les alertes DETER ont montré une diminution de 2 % de la déforestation entre août 2021 et juillet 2022 par rapport à la période d'août 2020 à juillet 2021, mais ont enregistré une augmentation de 25 % entre l'année civile 2021 et 2022.</p>



<h4>Alertes DETER en Amazonie brésilienne</h4>



<div class="flourish-embed" data-src="story/1962332?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<h3>Indonésie</h3>



<p>Cette année, le WRI (World Resources Institute) et le ministère indonésien de l'environnement et des forêts (MoEF) travaillent ensemble sur une analyse commune dans le but de mieux comprendre les résultats de la perte de forêt primaire selon les données et les définitions de l'UMD/GFW (Université du Maryland/Global Forest Watch). En superposant les données de perte de forêt primaire de l'UMD/GFW pour l'année 2022 avec la carte de couverture terrestre du ministère indonésien de l'environnement et des forêts, il a été constaté que 67 % des pertes se produisent dans les classes officielles de couverture forestière de l'Indonésie, tandis que les 33 % restants se produisent dans des zones de terres agricoles mixtes sèches, d'arbustes, d'arbustes marécageux et d'autres types de couverture terrestre. De plus, il a été relevé qu'environ 107 000 hectares de perte se trouvent à la fois dans les classes officielles de couverture forestière de l'Indonésie et ont une taille de parcelle supérieure à deux hectares. Parmi ces 107 000 hectares, environ 12 000 hectares se situent dans des zones légalement désignées comme des forêts primaires en Indonésie. L'analyse conjointe en cours permettra d'améliorer notre compréhension de ces questions.</p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/07/FR-23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded-FR.png" alt="Perte de forêt primaire de l'UMD/GFW en hectares par classe de forêt du MoEF en 2022" class="wp-image-16149"/></figure>



<h2>L’amélioration des données au fil du temps a entraîné des incohérences dans les données antérieures à 2015&nbsp;&nbsp;</h2>



<p>Les ajustements des algorithmes et de meilleures données satellitaires ont <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/">amélioré les données de perte de la couverture arborée au cours du temps</a>. Les changements dans l’algorithme de détection de la perte de la couverture arborée pour les années 2011 à 2014 ainsi que 2015 et au-delà, de même que l’intégration des données de Landsat 8 à partir de 2013, facilitent la détection des changements à plus faible échelle, comme les pertes dues aux incendies, la coupe sélective et des changements agricoles.&nbsp;Les variations de disponibilité des images satellites signifient également qu’il y a des incohérences dans la qualité et le nombre d’images disponibles pour la capture de données chaque année.</p>



<p>Pour corriger ce problème, nous&nbsp;:</p>



<ul><li>Concentrons nos analyses sur les tendances après 2015</li><li>Évaluer la moyenne mobile sur trois ans pour interpréter les tendances à long terme.</li><li>Ne pas tenir compte de l’augmentation des pertes après 2012 dans les régions dominées par l’agriculture à petite échelle, comme l’Afrique centrale. Cette augmentation est probablement due à l’amélioration des données qui permet de mieux détecter la perte plus tard dans la série chronologique.</li></ul>



<h2>Comment les données évolueront-elles dans le temps ?</h2>



<p>L’UMD travaille à l’amélioration des données sur l’étendue et l’évolution du couvert arboré, ce qui entraînera des modifications des données dans les années à venir. Les nouvelles données permettront de suivre l’étendue annuelle du couvert arboré et fourniront des informations sur les pertes et les gains annuels à l’aide d’un algorithme cohérent pour l’ensemble de la série chronologique. Elles permettront également de suivre plusieurs événements de gain et de perte en un seul endroit, ce qui nous permettra de mieux suivre les zones dynamiques de changement forestier, telles que les activités forestières et les incendies de forêt. La prise en compte des gains annuels et des pertes répétées permettra d’obtenir une image plus complète de l’évolution des forêts à l’échelle mondiale.</p>



<style type="text/css"></style>

Explication des données du Global Forest Watch sur la perte du couvert arboré en 2022

<p>Novos dados sobre a perda de cobertura arbórea da Universidade de Maryland (University of Maryland, UMD) disponíveis no Global Forest Watch (GFW) mostram taxas altas de perda de florestas primárias em 2022, apesar das&nbsp;<a href="https://gfw.global/3L1xFpo">ambições políticas de conter a perda</a>. O que os dados medem, o que há de diferente este ano e como eles se comparam a outras estimativas oficiais de desmatamento? Aqui está o que você precisa saber sobre os novos dados.</p>



<h2>O que a perda de cobertura arbórea mede e como ela difere do desmatamento</h2>



<p>Os dados de perda de cobertura arbórea da UMD capturam a perda anual de todas as árvores com mais de cinco metros de altura entre os anos de 2001 e 2022. Como a coleta de dados por satélite óptico pode, muitas vezes, ser atrapalhada por nuvens nos trópicos úmidos, algumas perdas no final do ano podem ser detectadas no próximo ano após uma cobertura de nuvens persistente. Essa é uma ocorrência comum em muitos países tropicais onde a cobertura de nuvens é predominante.</p>



<p>Os dados sobre perda de cobertura arbórea incluem a perda de árvores em florestas naturais, bem como em plantações e cultivo arbóreo. A perda dessa cobertura pode ser devido a causas humanas ou naturais, permanente ou temporária. O desmatamento difere por se referir apenas a uma mudança permanente e causada pelo homem, partindo do estado de floresta para outros usos da terra. Algumas formas de perda de cobertura arbórea, como a conversão de uma floresta natural em terras agrícolas, são consideradas desmatamento, enquanto outras formas, como a extração de madeira em florestas plantadas ou distúrbios naturais, não são. <a href="https://research.wri.org/gfr/key-terms-definitions">Leia mais sobre as diferenças aqui</a>.</p>



<p>Uma nova ferramenta disponível na Global Forest Review, o&nbsp;<a href="https://gfw.global/3L1xFpo">Targets Tracker</a>&nbsp;(Rastreador de metas), nos permite usar dados sobre perda de cobertura arbórea para medir o progresso em relação aos compromissos de desmatamento em nível global, usando uma&nbsp;<a href="https://research.wri.org/gfr/forest-targets-tracker#how-is-deforestation-estimated">aproximação de desmatamento</a>&nbsp;de resolução grosseira. Para isso, o Targets Tracker identifica as perdas que&nbsp;provavelmente&nbsp;representam o desmatamento dentro e fora dos trópicos, incluindo apenas a perda de cobertura arbórea de determinados fatores. Especificamente, ele inclui a expansão da agricultura de pequena escala em florestas primárias tropicais úmidas, bem como todas as perdas no desmatamento e na urbanização impulsionados por commodities. Também exclui perdas temporárias, como as decorrentes de incêndios ou atividades florestais.</p>



<h2>Por que nos concentramos nos trópicos?</h2>



<p>Na&nbsp;<a href="https://gfw.global/3LkdrbQ">análise dos dados da UMD</a>, nós nos concentramos principalmente na perda de cobertura arbórea em florestas primárias tropicais, pois as florestas tropicais sofrem&nbsp;<a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&amp;utm_medium=bitly&amp;utm_source=GFWBlog#how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests">a grande maioria</a>&nbsp;(mais de 96%) do desmatamento mundial, e a perda nessas áreas tem impactos enormes sobre a biodiversidade e o armazenamento de carbono. Mesmo que essas perdas acabem sendo revertidas,&nbsp;<a href="https://www.science.org/doi/10.1126/science.abh3629">levará décadas</a>&nbsp;para que esses habitats e estoques de carbono se recuperem, e poderá ocorrer uma perda permanente de biodiversidade.</p>



<h2>Como a perda de cobertura arbórea se compara aos dados nacionais oficiais no Brasil e na Indonésia?</h2>



<h3>Brasil&nbsp;</h3>



<p>O PRODES – o sistema oficial de monitoramento florestal para a Amazônia do Instituto Nacional de Pesquisas Espaciais (INPE) – e os dados da UMD têm diferenças metodológicas e de definição que são importantes de serem compreendidas ao comparar essas duas fontes de dados. O PRODES mede o desmatamento nítido e as perdas florestais antropogênicas por incêndio maiores que 6,25 hectares, enquanto a UMD captura perdas maiores que 0,09 hectare de todas as árvores com altura superior a cinco metros. Ambas são medidas importantes que nos ajudam a entender como as florestas estão mudando, uma vez que o desmatamento, as queimadas e pequenos distúrbios no dossel florestal podem levar a impactos no clima, na biodiversidade e nos serviços ecossistêmicos.</p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/06/PT-23-05-04-gfr-technical-blog_UMDvsProdes-PT.png" alt="Explicação das diferenças da UMD e do PRODES" class="wp-image-16134"/></figure>



<p>Separar a perda de cobertura arbórea de incêndios nos dados da UMD resultou em uma correspondência muito mais próxima entre esses dois conjuntos de dados. No entanto, eles parecem mostrar tendências diferentes na Amazônia brasileira no ano de 2022: o PRODES mostra uma redução de 11% no desmatamento, enquanto a UMD mostra um aumento de 21% na perda de floresta primária sem relação com incêndios.</p>



<h4>Dados do PRODES vs. UMD sobre a Amazônia brasileira&nbsp;</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/14224655?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p>O motivo para essa discrepância pode ser a diferença nos períodos dos relatórios. O PRODES se utiliza do período entre agosto e julho (nesse caso, agosto de 2021 a julho de 2022), enquanto a UMD usa o ano-calendário (janeiro a dezembro de 2022). Os alertas do DETER do governo brasileiro, que rastreiam diariamente o desmatamento, sugerem que as taxas de desmatamento aumentaram no segundo semestre de 2022 em comparação com os anos anteriores. Os alertas do DETER detectaram uma redução de 2% no desmatamento entre agosto de 2021 e julho de 2022 em comparação com agosto de 2020 a julho de 2021, mas detectaram um aumento de 25% entre o ano-calendário de 2021 e 2022.</p>



<h4>Alertas do DETER na Amazônia brasileira</h4>



<div class="flourish-embed" data-src="story/1959112?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<h3>Indonésia</h3>



<p>Neste ano, o WRI e o Ministério do Meio Ambiente e Florestas da Indonésia (MoEF) estão trabalhando juntos para analisar e compreender os dados da Universidade de Maryland e do GFW de perda de florestas primárias pela lente das definições e classificações do MoEF. Sobrepondo os dados de perda de florestas primárias de 2022 da UMD/GFW com os mapas de uso e cobertura da terra, podemos identificar que 67% da perda ocorre em classes de cobertura de florestas definidas pela Indonésia, enquanto 33% restantes se dão em áreas de outros usos, como agricultura mista, arbustos, áreas alagáveis e outros tipos de cobertura de vegetação. Aproximadamente 107 mil hectares de perda foram identificados dentro das classes oficiais do governo. Desses 107 mil hectares, cerca de 12 mil hectares estão áreas legalmente classificadas como florestas primárias na Indonésia. A análise conjunta irá melhorar nossa compreensão sobre essas questões. &nbsp;</p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/06/PT-23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded-PT.png" alt="Perda de florestas primárias UMD/GFW em hectares segundo as classes do MoEF em 2022" class="wp-image-16135"/></figure>



<h2>Melhorias de dados ao longo do tempo resultam em inconsistências nos dados antes de 2015&nbsp;</h2>



<p>Ajustes de algoritmo e melhores dados de satélite <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/">melhoraram o conjunto de dados de perda de cobertura arbórea ao longo do tempo</a>. Mudanças no algoritmo de detecção de perda de cobertura arbórea para os anos de 2011 a 2014 e 2015 em diante e a incorporação de dados do Landsat 8 a partir de 2013 facilitam a detecção de mudanças de menor escala, como perdas devido a incêndios, corte seletivo e agricultura itinerante. As variações na disponibilidade de imagens de satélite também significam que há inconsistências com a qualidade e o número de imagens disponíveis para capturar dados a cada ano.</p>



<p>Para abordar esses problemas, nós:</p>



<ul><li>concentramos nossa análise em tendências após 2015;</li><li>avaliamos a média móvel de três anos para interpretar as tendências de longo prazo; e</li><li>desconsideramos o aumento da perda após 2012 em locais dominados pela agricultura de pequena escala, como a África Central. Esse aumento provavelmente se deve a melhorias nos dados que detectam melhor a perda posterior na série temporal.</li></ul>



<h2>Como os dados mudarão com o passar do tempo?</h2>



<p>A UMD está trabalhando para aprimorar os dados de extensão e alteração de cobertura arbórea, o que resultará em mudanças nos dados para os próximos anos. Os novos dados rastrearão a extensão anual de&nbsp;cobertura arbórea e oferecerão informações anuais de perdas e ganhos usando um algoritmo consistente em toda a série temporal. Eles também rastrearão vários eventos de ganho e perda em um mesmo local, o que nos permitirá rastrear melhor as áreas dinâmicas de mudança florestal, como atividades de reflorestamento e incêndios florestais. A incorporação de eventos de ganho e perda anuais e repetidos fornecerá um cenário mais completo das mudanças florestais globais.</p>



<style type="text/css"></style>

Explicação dos dados do Global Forest Watch sobre a perda de cobertura arbórea de 2022

<p>Nuevos datos sobre la pérdida de cobertura arbórea facilitados por la Universidad de Maryland (University of Maryland, UMD) y disponibles en Global Forest Watch (GFW) muestran tasas altas de pérdida de bosques primarios en 2022, a pesar de las <a href="https://gfw.global/3L1xFpo" target="_blank" rel="noreferrer noopener">ambiciones políticas para frenar la pérdida</a>. ¿Qué miden los datos, qué es diferente este año y cómo se compara con otras estimaciones oficiales de deforestación? Esto es lo que usted debe saber sobre los nuevos datos.&nbsp;</p>



<h2>Qué mide la pérdida de cobertura arbórea y cuál es su diferencia con la deforestación</h2>



<p>Los datos sobre la pérdida de cobertura arbórea de la UMD registran la pérdida anual de todos los árboles de más de cinco metros de altura en los años naturales comprendidos entre 2001 y 2022. Debido a que la recolección de datos ópticos por satélite a menudo puede estar oscurecida por la cobertura de nubes en los trópicos húmedos, se puede registrar alguna pérdida tardía en el año siguiente después de la cobertura de nubes persistente. Este fenómeno es común en muchos países tropicales en los que predomina la cobertura de nubes.&nbsp;&nbsp;</p>



<p>Los datos sobre la pérdida de cobertura arbórea incluyen la pérdida de árboles en bosques naturales, así como en plantaciones y cultivos de árboles. La pérdida de esta cobertura arbórea puede ser de origen humano o natural y puede ser permanente o temporal. La deforestación consiste en un cambio permanente y provocado por el hombre de un bosque a otro uso del suelo. Algunas formas de pérdida de cobertura arbórea, como la conversión de un bosque natural en tierras agrícolas, se consideran deforestación, mientras que otras formas de pérdida de cobertura arbórea, como la tala de madera en plantaciones forestales o las alteraciones naturales, no. <a href="https://research.wri.org/gfr/key-terms-definitions" target="_blank" rel="noreferrer noopener">Lea más sobre las diferencias aquí</a>.&nbsp;&nbsp;</p>



<p>Una nueva herramienta disponible en Global Forest Review, el <a href="https://gfw.global/3L1xFpo" target="_blank" rel="noreferrer noopener">Rastreador de Objetivos</a>, nos permite utilizar los datos de pérdida de cobertura arbórea para medir el progreso hacia el cumplimiento de los compromisos relacionados con la deforestación a nivel mundial gracias a un <a href="https://research.wri.org/gfr/forest-targets-tracker#how-is-deforestation-estimated" target="_blank" rel="noreferrer noopener">proxy de deforestación</a> de resolución gruesa. Para ello, el Rastreador de Objetivos identifica las pérdidas que podrían representar deforestación dentro y fuera de los trópicos incluyendo únicamente la pérdida de cobertura arbórea causada por determinados factores. Más específicamente, incluye la expansión de la agricultura a pequeña escala en los bosques primarios tropicales húmedos, así como toda pérdida en la deforestación provocada por productos básicos de consumo y urbanización. Quedan excluidas las pérdidas temporales, como las provocadas por incendios o actividades forestales.&nbsp;&nbsp;</p>



<h2>¿Por qué nos centramos en los trópicos?&nbsp;</h2>



<p>En nuestro <a href="https://gfw.global/3LkdrbQ" target="_blank" rel="noreferrer noopener">análisis de</a> los datos de la Universidad de Maryland (UMD), nos centramos principalmente en la pérdida de cobertura arbórea en los bosques tropicales primarios, ya que en ellos se registra <a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&amp;utm_medium=bitly&amp;utm_source=GFWBlog#how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests" target="_blank" rel="noreferrer noopener">la mayor parte</a> (más del 96%) de la deforestación mundial, y la pérdida en esas áreas tiene enormes repercusiones sobre la biodiversidad y el almacenamiento de carbono. Aunque estas pérdidas puedan revertirse eventualmente, los hábitats y las reservas de carbono <a href="https://www.science.org/doi/10.1126/science.abh3629" target="_blank" rel="noreferrer noopener">tardarán décadas</a> en recuperarse y podría producirse una pérdida permanente de biodiversidad.&nbsp;</p>



<h2>¿Cómo se compara la pérdida de cobertura arbórea con los datos oficiales de Brasil e Indonesia?</h2>



<h3>Brasil&nbsp;</h3>



<p>PRODES (el sistema oficial de monitoreo forestal para la Amazonia del Instituto Nacional De Investigación Espacial de Brasil (INPE)) y los datos de la Universidad de Maryland (UMD) tienen diferencias metodológicas y de definición que son importantes de entender cuando se comparan estas dos fuentes de datos. PRODES mide la deforestación total de más de 6.25 hectáreas y pérdidas de bosques antropogénicos por incendios, mientras que la Universidad de Maryland (UMD) registra la pérdida de más de 0.09 hectáreas de todos los árboles de más de cinco metros de altura. Ambas son medidas importantes que nos ayudan a entender cómo cambian los bosques, ya que la deforestación, los incendios forestales y las pequeñas alteraciones del dosel forestal pueden tener impactos en el clima, la biodiversidad y los servicios de los ecosistemas.&nbsp;</p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/07/SP-23-05-04-gfr-technical-blog_UMDvsProdes-ES.png" alt="" class="wp-image-16173"/></figure>



<p>Al separar la pérdida de cobertura arbórea de los incendios en los datos de la UMD se obtuvo una correspondencia mucho más estrecha entre estos dos conjuntos de datos. Sin embargo, ambos parecen mostrar tendencias distintas para 2022 en la Amazonia brasileña: PRODES muestra un descenso del 11 % en la deforestación, mientras que UMD muestra un aumento del 21 % en la pérdida de bosques primarios no afectados por incendios.&nbsp;</p>



<h4>Datos de PRODES vs. Universidad de Maryland (UMD) en la Amazonia brasileña&nbsp;</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/14295074?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p>El motivo de esta discrepancia puede ser la diferencia en los periodos reportados. PRODES utiliza el periodo que comprende entre agosto y julio (en este caso, de agosto de 2021 a julio de 2022), mientras que la UMD utiliza el año calendario (de enero a diciembre de 2022). Las alertas DETER del gobierno brasileño, que realizan un seguimiento diario de la deforestación, sugieren que las tasas de deforestación aumentaron en el segundo semestre de 2022 en comparación con años anteriores. Las alertas DETER detectaron una disminución del 2 % de la deforestación entre agosto de 2021 y julio de 2022 en comparación con agosto de 2020 y julio de 2021, pero detectaron un aumento del 25 % entre el año calendario 2021 y 2022.&nbsp;</p>



<h4>Alertas DETER en la Amazonia brasileña&nbsp;</h4>



<div class="flourish-embed" data-src="story/1965302?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<h3>Indonesia&nbsp;&nbsp;</h3>



<p>Este año, WRI y el Ministerio de Medio Ambiente y Silvicultura (MoEF) de Indonesia están trabajando juntos en un análisis conjunto para comprender mejor los hallazgos de la pérdida de bosque primario según UMD/GFW a través de los datos y definiciones del MoEF. Al superponer los datos de pérdida de bosque primario de UMD/GFW para 2022 con el mapa de cobertura de tierras del MoEF, se observa que el 67% ocurre dentro de las clases oficiales de cobertura forestal de Indonesia, mientras que el 33% restante tiene lugar en áreas de agricultura mixta de tierras secas, arbustos, pantanos y otros tipos de cobertura del suelo. Aproximadamente, se encontró una pérdida de 107 mil hectáreas que se ubicaba tanto dentro de las clases oficiales de cobertura forestal de Indonesia con un área mayor a dos hectáreas. De estas 107 mil hectáreas, alrededor de 12 mil hectáreas se encontraban dentro de áreas legalmente clasificadas como bosque primario en Indonesia. El próximo análisis conjunto mejorará nuestra comprensión de estos problemas.&nbsp;</p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/07/SP-23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded-ES.png" alt="" class="wp-image-16174"/></figure>



<h2>Las mejoras en los datos a lo largo del tiempo generan incoherencias en los datos anteriores a 2015 </h2>



<p>Los ajustes del algoritmo y la mejora de los datos satelitales han <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/" target="_blank" rel="noreferrer noopener">perfeccionado el conjunto de datos de pérdida de cobertura arbórea a lo largo del tiempo</a>. Los cambios en el algoritmo de detección de la pérdida de cobertura arbórea para los años 2011-2014 y 2015 en adelante y la incorporación de los datos de Landsat 8 a partir de 2013 facilitan la detección de cambios a menor escala, como la pérdida debida a los incendios, la explotación selectiva y la agricultura migratoria. Las variaciones en la disponibilidad de imágenes satelitales también significa que hay inconsistencias con la calidad y el número de imágenes disponibles para registrar datos cada año.&nbsp;&nbsp;</p>



<p>Para abordar estos problemas, realizamos lo siguiente:&nbsp;&nbsp;</p>



<ul><li>Nos centramos en nuestros análisis de tendencias después del 2015.&nbsp;&nbsp;</li><li>Se evalúaa el promedio móvil de tres años para interpretar las tendencias a más largo plazo.&nbsp;</li><li>Se ignora el aumento de las pérdidas después de 2012 en lugares donde predomina la agricultura a pequeña escala, como África Central. Es probable que este aumento se deba a la mejora de los datos, que detectan mejor las pérdidas más adelante en la serie de tiempo.&nbsp;&nbsp;</li></ul>



<h2>¿Cómo evolucionarán los datos con el tiempo?&nbsp;</h2>



<p>La UMD está trabajando para mejorar los datos de cambios y extensión de la cobertura arbórea, lo que resultará en cambios en los datos en los próximos años. Los nuevos datos permitirán hacer un seguimiento de la extensión anual de la cobertura arbórea y ofrecerán información sobre las pérdidas y ganancias anuales utilizando un algoritmo consistente para toda la serie de tiempo. También hará un seguimiento de varios episodios de ganancia y pérdida en una única ubicación, lo que nos permitirá detectar mejor las áreas dinámicas de cambio en los bosques, como actividades forestales e incendios forestales. La incorporación de episodios de ganancia anual y pérdida recurrente proporcionará una imagen más completa del cambio en los bosques a escala mundial.&nbsp;&nbsp;</p>



<style type="text/css"></style>

Explicación de los datos de Global Forest Watch sobre la pérdida de cobertura arbórea en 2022

<p>Data kehilangan tutupan pohon yang baru dari University of Maryland (UMD) dan tersedia di Global Forest Watch (GFW) menunjukkan tingginya laju kehilangan hutan primer pada tahun 2022 kendati adanya <a href="https://gfw.global/3L1xFpo" target="_blank" rel="noreferrer noopener">ambisi politik untuk mengatasi kehilangan hutan</a>. Apa yang diukur oleh data tersebut, apa yang berbeda tahun ini, dan bagaimana perbandingannya dengan perkiraan resmi deforestasi lainnya? Inilah yang harus Anda ketahui tentang data baru.&nbsp;</p>



<h2>Apa yang diukur oleh kehilangan tutupan pohon dan apa perbedaannya dengan deforestasi&nbsp;</h2>



<p>Data kehilangan tutupan pohon dari UMD mencatat kehilangan semua pohon yang memiliki tinggi lebih dari 5 meter setiap tahunnya selama tahun kalender antara 2001 sampai 2022. Karena pengumpulan data satelit optik kerap terkendala tutupan awan di daerah tropis yang lembap, data kehilangan tutupan pohon yang terjadi pada akhir tahun dapat terdeteksi di tahun berikutnya setelah tutupan awan pekat berkurang. Hal ini umum terjadi di banyak negara tropis yang lazim mengalami tutupan awan.&nbsp;&nbsp;</p>



<p>Data kehilangan tutupan pohon ini tidak hanya mencakup kehilangan pohon di hutan alami, melainkan juga di perkebunan dan tanaman pohon. Kehilangan tutupan pohon ini dapat diakibatkan oleh penyebab manusia atau alam serta dapat bersifat permanen atau sementara. Deforestasi berbeda karena deforestasi hanya mengacu pada perubahan dari hutan menjadi penggunaan lahan lainnya yang bersifat permanen dan disebabkan oleh manusia. Beberapa bentuk kehilangan tutupan pohon, seperti konversi hutan alami menjadi lahan pertanian, dianggap sebagai deforestasi, sedangkan bentuk kehilangan tutupan pohon lainnya, seperti pemanenan kayu di hutan perkebunan atau gangguan alami, tidak termasuk deforestasi. <a href="https://research.wri.org/gfr/key-terms-definitions" target="_blank" rel="noreferrer noopener">Baca selengkapnya mengenai perbedaannya di sini</a>.&nbsp;&nbsp;</p>



<p>Tersedia alat bantu baru di Global Forest Review, <a href="https://gfw.global/3L1xFpo" target="_blank" rel="noreferrer noopener">Pelacak Target</a>, yang memungkinkan kami menggunakan data kehilangan tutupan pohon untuk mengukur kemajuan komitmen deforestasi pada tingkat global dengan menggunakan resolusi kasar <a href="https://research.wri.org/gfr/forest-targets-tracker%22%20/l%20%22how-is-deforestation-estimated" target="_blank" rel="noreferrer noopener">proksi deforestasi</a>. Untuk melakukan hal ini, Pelacak Target mengidentifikasi kehilangan yang kemungkinan mewakili deforestasi di dalam dan di luar daerah tropis hanya dengan mencakup kehilangan tutupan pohon oleh penyebab tertentu. Secara khusus, alat bantu ini mencakup perluasan pertanian berskala kecil menuju hutan primer tropis lembap, serta kehilangan dalam urbanisasi dan deforestasi yang didorong oleh komoditas. Alat bantu ini mengecualikan kehilangan sementara, seperti karena kebakaran atau aktivitas kehutanan.&nbsp;&nbsp;</p>



<h2>Mengapa kami berfokus pada daerah tropis&nbsp;</h2>



<p>Dalam <a href="https://gfw.global/3LkdrbQ" target="_blank" rel="noreferrer noopener">analisis data UMD</a>, kami kebanyakan berfokus pada kehilangan tutupan pohon di hutan primer tropis, sebab hutan tropis menjadi korban dari <a href="https://research.wri.org/gfr/forest-extent-indicators/forest-loss?utm_campaign=treecoverloss2021&amp;utm_medium=bitly&amp;utm_source=GFWBlog%22%20\l%20%22how-much-tree-cover-is-lost-in-tropical-versus-temperate-and-boreal-forests" target="_blank" rel="noreferrer noopener">sebagian besar</a> (lebih dari 96%) deforestasi di dunia, dan kehilangan di kawasan tersebut berdampak besar pada keanekaragaman hayati dan penyimpanan karbon. Meskipun kehilangan tersebut pada akhirnya akan pulih kembali, <a href="https://www.science.org/doi/10.1126/science.abh3629" target="_blank" rel="noreferrer noopener">dibutuhkan waktu puluhan tahun</a> agar habitat dan stok karbon ini dapat pulih, dan kehilangan keanekaragaman hayati permanen dapat terjadi.&nbsp;</p>



<h2>Bagaimana perbandingan kehilangan tutupan pohon dengan data negara resmi di Brasil dan Indonesia?&nbsp;&nbsp;</h2>



<h3>Brasil&nbsp;</h3>



<p>PRODES — sistem pemantauan hutan resmi untuk Amazon dari Institut Nasional untuk Penelitian Angkasa Brasil (National Institute for Space Research, INPE) — dan data UMD memiliki perbedaan metodologis dan definisi yang penting untuk dipahami ketika membandingkan kedua sumber data ini. PRODES mengukur deforestasi tebang habis dan kehilangan hutan antropogenik akibat kebakaran yang lebih besar dari 6,25 hektare, sedangkan UMD mencatat kehilangan yang lebih besar dari 0,09 hektare dari semua pohon yang tingginya lebih dari lima meter. Keduanya merupakan langkah penting yang membantu kita memahami bagaimana hutan berubah karena deforestasi, kebakaran hutan, dan gangguan kecil pada kanopi hutan, semuanya dapat berdampak pada iklim, keanekaragaman hayati, dan jasa ekosistem. </p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/08/23-05-04-gfr-technical-blog_UMDvsProdes-IND.png" alt="" class="wp-image-16218"/></figure>



<p>Memisahkan kehilangan tutupan pohon karena kebakaran dalam data UMD memberikan perbandingan yang lebih mirip antara kedua kumpulan data ini. Namun demikian, keduanya tampak menunjukkan tren yang berbeda untuk tahun 2022 di Amazon Brasil: PRODES menunjukkan penurunan deforestasi sebesar 11%, sedangkan UMD menunjukkan kenaikan kehilangan hutan primer nonkebakaran sebesar 21%.&nbsp;</p>



<h4>PRODES vs. Data UMD di Amazon Brasil&nbsp;</h4>



<div class="flourish-embed flourish-chart" data-src="visualisation/14222900?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<p>Alasan dari ketidaksesuaian ini mungkin dikarenakan oleh perbedaan dalam periode pelaporan. PRODES menggunakan periode antara Agustus sampai Juli (dalam hal ini, Agustus 2021 sampai Juli 2022), sedangkan UMD menggunakan tahun kalender (Januari sampai Desember 2022). Peringatan DETER dari pemerintah Brasil — yang melacak deforestasi setiap harinya — menandakan tingkat deforestasi naik pada paruh kedua tahun 2022 dibandingkan dengan tahun sebelumnya. Peringatan DETER mendeteksi penurunan deforestasi sebesar 2% antara Agustus 2021 sampai Juli 2022 dibandingkan dengan periode Agustus 2020 sampai Juli 2021, namun mendeteksi peningkatan sebesar 25% antara tahun kalender 2021 dan 2022.&nbsp;</p>



<h4>Peringatan DETER di Amazon Brasil</h4>



<div class="flourish-embed" data-src="story/1958956?163031"><script src="https://public.flourish.studio/resources/embed.js"></script></div>



<h3>Indonesia&nbsp;&nbsp;</h3>



<p>Tahun ini, WRI dan Kementerian Lingkungan Hidup dan Kehutanan (KLHK) bekerja sama dalam analisis gabungan untuk memahami lebih lanjut temuan kehilangan hutan primer UMD/GFW menggunakan data dan definisi dari KLHK. Membandingkan data kehilangan hutan primer UMD/GFW 2022 dengan peta tutupan lahan KLHK memperlihatkan bahwa 67% kehilangan terjadi dalam beberapa kelas tutupan hutan resmi Indonesia, sementara 33% lainnya mencakup lahan pertanian kering campuran, belukar, rawa, dan tipe tutupan lahan lainnya.&nbsp;</p>



<p>Sekitar 107 ribu hektare hutan primer yang hilang ditemukan pada data kelas tutupan hutan resmi Indonesia dan dengan potongan yang lebih besar dari 2 hektare. Dari 107 ribu hektare tersebut, sekitar 12 ribu hektare mencakup Kawasan yang secara legal diklasifikasikan sebagai hutan primer di Indonesia. Analisis gabungan yang akan diterbitkan segera akan memperdalam pemahaman kita akan hal ini. </p>



<figure class="wp-block-image size-large"><img src="https://content.globalforestwatch.org/wp-content/uploads/2023/08/23-05-04-gfr-technical-blog_Indonesia-error-bars-table-unrounded-IND.png" alt="" class="wp-image-16219"/></figure>



<h2>Perbaikan data dari waktu ke waktu menghasilkan inkonsistensi dalam data sebelum 2015&nbsp;&nbsp;</h2>



<p>Penyesuaian algoritme dan data satelit yang lebih baik telah <a href="https://www.globalforestwatch.org/blog/data-and-research/tree-cover-loss-satellite-data-trend-analysis/" target="_blank" rel="noreferrer noopener">meningkatkan kumpulan data kehilangan tutupan pohon dari waktu ke waktu</a>. Perubahan pada algoritme pendeteksian kehilangan tutupan pohon untuk tahun 2011-2014 dan 2015 dan seterusnya dan penggabungan data Landsat 8 mulai tahun 2013 memudahkan untuk mendeteksi perubahan skala yang lebih kecil, seperti kehilangan karena kebakaran, tebang pilih, dan pertanian berpindah. Variasi ketersediaan citra satelit juga berarti adanya inkonsistensi dengan kualitas dan jumlah citra yang tersedia untuk diambil data setiap tahunnya.&nbsp;&nbsp;</p>



<p>Untuk mengatasi masalah ini, kami:&nbsp;&nbsp;</p>



<ul><li>Memfokuskan analisis kami pada tren pasca-2015.&nbsp;&nbsp;</li><li>Menilai rata-rata pergerakan selama tiga tahun untuk menginterpretasikan tren dalam jangka waktu yang lebih lama.&nbsp;</li><li>Mengabaikan peningkatan kehilangan pasca-2012 di tempat yang didominasi oleh pertanian skala kecil, seperti Afrika Tengah. Peningkatan ini kemungkinan diakibatkan oleh perbaikan data yang mendeteksi kehilangan tersebut secara lebih baik di kemudian waktu.&nbsp;&nbsp;</li></ul>



<h2>Bagaimana data akan berubah dari waktu ke waktu?&nbsp;</h2>



<p>UMD tengah mengupayakan perbaikan terhadap sebaran tutupan pohon dan data perubahan, yang akan menghasilkan perubahan terhadap data di tahun-tahun mendatang. Data yang baru akan melacak sebaran tutupan pohon tahunan serta menawarkan informasi kehilangan dan perolehan tahunan menggunakan algoritme yang konsisten di seluruh rangkaian waktu. Data ini juga akan melacak beberapa peristiwa perolehan dan kehilangan di satu lokasi, yang memungkinkan kami melacak dengan lebih baik luas dinamis perubahan hutan, seperti kegiatan kehutanan dan kebakaran hutan. Menggabungkan peristiwa perolehan dan kehilangan berulang tahunan akan memberikan gambaran perubahan hutan global yang lebih lengkap.&nbsp;&nbsp;</p>



<style type="text/css"></style>

Penjelasan Data Kehilangan Tutupan Pohon 2022 dari Global Forest Watch

New data shows high rates of primary forest loss in 2022. What does the data measure and how does it compare to other official estimates of deforestation?

Global Forest Watch’s 2022 Tree Cover Loss Data Explained | GFW Blog

Forest Loss Remained Stubbornly High in 2021

La pérdida de bosques se mantuvo obstinadamente elevada en 2021

Perda florestal permanece resistentemente alta em 2021

Kehilangan Hutan Tetap Tinggi di Tahun 2021

La perte des forêts est restée chroniquement élevée en 2021

Forest loss, including tropical deforestation, remained stubbornly high in 2021 according to new data on the Global Forest Watch.

What Happened to Forests in 2021? | Global Forest Watch Blog