Understanding forest loss patterns in Amazonia, the Earth’s largest rainforest region, is critical for effective forest conservation and management. Following the most detailed analysis to date, spanning the entire Amazon and extending over a 14-year period (2001–2014), we reveal significant shifts in deforestation dynamics of Amazonian forests. Firstly, hotspots of Amazonian forest loss are moving away from the southern Brazilian Amazon to Peru and Bolivia. Secondly, while the number of new large forest clearings (>50 ha) has declined significantly over time (46%), the number of new small clearings (<1 ha) increased by 34% between 2001–2007 and 2008–2014. Thirdly, we find that small-scale low-density forest loss expanded markedly in geographical extent during 2008–2014. This shift presents an important and alarming new challenge for forest conservation, despite reductions in overall deforestation rates.

Protected areas (PAs) have been established to conserve tropical forests, but their effectiveness at reducing deforestation is uncertain. To explore this issue, we combined high resolution data of global forest loss over the period 2000–2012 with data on PAs. For each PA we quantified forest loss within the PA, in buffer zones 1, 5, 10 and 15 km outside the PA boundary as well as a 1 km buffer within the PA boundary. We analysed 3376 tropical and subtropical moist forest PAs in 56 countries over 4 continents. We found that 73% of PAs experienced substantial deforestation pressure, with >0.1% a−1 forest loss in the outer 1 km buffer. Forest loss within PAs was greatest in Asia (0.25% a−1) compared to Africa (0.1% a−1), the Neotropics (0.1% a−1) and Australasia (Australia and Papua New Guinea; 0.03% a−1). We defined performance (P) of a PA as the ratio of forest loss in the inner 1 km buffer compared to the loss that would have occurred in the absence of the PA, calculated as the loss in the outer 1 km buffer corrected for any difference in deforestation pressure between the two buffers. To remove the potential bias due to terrain, we analysed a subset of PAs (n = 1804) where slope and elevation in inner and outer 1 km buffers were similar (within 1° and 100 m, respectively). We found 41% of PAs in this subset reduced forest loss in the inner buffer by at least 25% compared to the expected inner buffer forest loss (P<0.75). Median performance () of subset reserves was 0.87, meaning a reduction in forest loss within the PA of 13%. We found PAs were most effective in Australasia (), moderately successful in the Neotropics () and Africa (), but ineffective in Asia (). We found many countries have PAs that give little or no protection to forest loss, particularly in parts of Asia, west Africa and central America. Across the tropics, the median effectiveness of PAs at the national level improved with gross domestic product per capita. Whilst tropical and subtropical moist forest PAs do reduce forest loss, widely varying performance suggests substantial opportunities for improved protection, particularly in Asia.

Old-growth forests are an important, rare and endangered habitat in Europe. The ability to identify old-growth forests through remote sensing would be helpful for both conservation and forest management. We used data on beech, Norway spruce and mountain pine old-growth forests in the Ukrainian Carpathians to test whether Sentinel-2 satellite images could be used to correctly identify these forests. We used summer and autumn 2017 Sentinel-2 satellite images comprising 10 and 20 m resolution bands to create 6 vegetation indices and 9 textural features. We used a Random Forest classification model to discriminate between dominant tree species within old-growth forests and between old-growth and other forest types. Beech and Norway spruce were identified with an overall accuracy of around 90%, with a lower performance for mountain pine (70%) and mixed forest (40%). Old-growth forests were identified with an overall classification accuracy of 85%. Adding textural features, band standard deviations and elevation data improved accuracies by 3.3%, 2.1% and 1.8% respectively, while using combined summer and autumn images increased accuracy by 1.2%. We conclude that Random Forest classification combined with Sentinel-2 images can provide an effective option for identifying old-growth forests in Europe.

No abstract

Many remote sensing studies do not distinguish between natural and planted forests. We combine C-Band Synthetic Aperture Radar (Sentinel-1, S-1) and optical satellite imagery (Sentinel-2, S-2) and examine Random Forest (RF) classification of acacia plantations and natural forest in North-Central Vietnam. We demonstrate an ability to distinguish plantation from natural forest, with overall classification accuracies of 87% for S-1, and 92.5% and 92.3% for S-2 and for S-1 and S-2 combined respectively. We found that the ratio of the Short-Wave Infrared Band to the Red Band proved most effective in distinguishing acacia from natural forest. We used RF on S-2 imagery to classify acacia plantations into 6 age classes with an overall accuracy of 70%, with young plantation consistently separated from older. However, accuracy was lower at distinguishing between the older age classes. For both distinguishing plantation and natural forest, and determining plantation age, a combination of radar and optical imagery did nothing to improve classification accuracy.