Reduced-impact Logging Research Articles

Medium and long term effects of logging systems on forest structure and composition in the tropical rainforest of Suriname

ABSTRACT In this study, we compared the medium and long-term effects of logging on tropical rain forest ecosystems in Suriname using reduced impact logging (RIL), CELOS Management System (CMS), and conventional logging (CL) with forest without logging (NAT). In 18 forest plots with different times since logging ended and 9 unlogged forest plots, assessments were conducted on indicators of forest structure and vegetation composition. These indicators were then modelled as a function of management type and plot age. Multivariate ordination was used to visualize forest structure and vegetation composition differences between plots. High vegetation cover in the tree layer was more related to the older RIL and the CL plots, while high vegetation cover in the sapling and seedling layer related more to the CMS and NAT plots. Canopy openness was more related to the NAT, the CMS and the younger RIL plots. Furthermore, the CMS plots, which were logged 30 years ago, still did not reach the basal area of commercially important primary forest species, compared to unlogged forest. Both RIL and CMS plots have a strong presence of secondary species. This study shows the importance of applying the RIL technique correctly and taking forest characteristics and ecological processes into account when developing forest management principles. Forest management systems should also be flexible enough to adapt to area-specific requirements, for example, increasing the rotation cycle if the timber volume is not on the desired level. In this way, both conservation and economic goals are reached.

Sustainable forest management is vital for the persistence of sun bear Helarctos malayanus populations in Sabah, Malaysian Borneo

With large mammals in Southeast Asia facing a host of threats such as forest loss and poaching, one important strategy in supporting conservation interventions is to understand local trends in species distributions. In Sabah, Malaysian Borneo, selective logging has both historically (conventional selective logging) and currently (reduced impact logging) impacted large areas of forest. Furthermore, the impact of poaching remains difficult to assess, potentially leading forest managers to underestimate its effects on large mammals. Our study focused on the sun bear (Helarctos malayanus), a forest dependent species threatened by poaching across its range. We used camera trap data collected in 2014 and 2018 from 127 camera trap stations in two adjacent forest blocks in Sabah. One of these blocks is a complex of degraded protected areas (formerly logged using conventional selective logging), interspersed with two areas of forest plantations (Tangkulap), while the other block is an active production forest logged using reduced impact logging techniques (Deramakot). We ran single-season occupancy models using covariates representing both habitat quality and poaching pressure to understand what factors most influenced sun bear habitat use and projected their distribution over the study area. We found that sun bear habitat selection was most strongly influenced by a dynamic covariate representing habitat quality, Normalised Difference Moisture Index (NDMI) from Landsat satellite data. Sun bear occupancy marginally increased from 2014 to 2018, in response to forest regeneration in Deramakot and Tangkulap. Overall, sun bear distribution was higher in the reduced impact logged and sustainably managed Deramakot compared to Tangkulap. Our proxies for poaching pressure had very minor effects on sun bear occupancy, suggesting that the population in this study area may not be subject to intense poaching. We highlight that sun bear populations can in fact better persist in well-managed production forests as compared to degraded protected areas.

Forest Management with Reduced-Impact Logging in Amazonia: Estimated Aboveground Volume and Carbon in Commercial Tree Species in Managed Forest in Brazil’s State of Acre

Tropical forest management has both positive and negative effects on climate change, and quantifying these effects is important both to avoid or minimize negative impacts and to reward net positive effects. This study contributes to this effort by estimating the aboveground volume and carbon present in commercial tree species in a managed forest in the forest harvest stage in Brazil’s state of Acre. A total of 12,794 trees of commercial species were measured. Trees were categorized and quantified as: “harvested trees” (“harvest or cut”), which were felled in the harvest stage, and “remaining trees” (“future cutting,” “trees in permanent protection areas or APPs,” “seed trees,” “rare trees” and “trees protected by law”) that remained standing in the forest post-harvest. Aboveground volume and carbon stocks of the 81 commercial species (diameter at breast height [DBH] ≥ 10 cm) totaled 79.19 m³ ha−1 and 21.54 MgC ha−1, respectively. The category “harvested trees” represents 44.48% and “remaining trees” 55.49% of the aboveground volume stocks. In the managed area, the category “harvested trees” is felled; this is composed of the commercial bole that is removed (19.25 m³ ha−1 and 5.32 MgC ha−1) and the stump and crown that remain in the forest as decomposing organic material (15.97 m³ ha−1 and 4.41 MgC ha−1). We can infer that the 21.54 MgC ha−1 carbon stock of standing commercial trees (DBH ≥ 10 cm) represents 13.20% of the total aboveground carbon in the managed area. The commercial boles removed directly from the forest represent 3.26% of the total aboveground carbon, and the stumps and crowns of the harvested trees represent the loss of an additional 2.70%. For sustainability of the management system in terms of carbon balance, growth in the 35-year management cycle must be sufficient to replace not only these amounts (0.27 MgC ha−1 year−1) but also losses to collateral damage and to additional logging-related effects from increased vulnerability to forest fires. Financial viability of future management cycles will depend on replenishment of commercial trees of harvestable size (DBH ≥ 50 cm).

PRODUKTIVITAS PEMANENAN KAYU DENGAN TEKNIK RIL DI PT. WIJAYA SENTOSA, PAPUA BARAT

The aim of this study is to measure the productivity of forest harvesting activities that consist of logging, skidding, loading, hauling, and unloading using the Reduced Impact Logging (RIL) technique in the natural forest of PT. Wijaya Sentosa, West Papua. The location of the study was determined by simple random sampling for 3 days for each activity, at 3 different subplots, also with different logging operator, skidding operator, and hauling driver. The final results of this research are productivity for loading using STIHLL 072 chainsaw is 5.74 m3/hour, productivity of skidding using Caterpillar 527 skidder winch is 11.43 m3/hour, productivity of loading using wheel loader 980F is 117.80 m3/hour, productivity of unloading using wheel loader 980F is 196.14 m3/hour, productivity of hauling using Mercedes Benz Actross logging truck is 15.62 m3/hour, and productivity of hauling using Mercedes Benz 3836AK logging truck is 16.34 m3/hour.Keywords: Productivity; Forest Harvesting; RIL (Reduced Impact Logging) Technique

Pattern of forest biomass recovery and biodiversity loss after reduced impact logging in East Kalimantan

Climate change substantially impacts sustainable ecosystem management, seeking to maintain the maximum amount of biomass and control biodiversity loss. Reduced impact logging (RIL) is a means to reduce the impact of biomass and biodiversity loss on natural production. This practice makes an important contribution to climate change mitigation and adaptation. This study aimed to determine the variation pattern of biomass recovery and biodiversity loss on natural forest stands after using the RIL technique. The approach to assessing the pattern was based on measuring growth patterns and quantitative ecological parameters on stands. The value was constructed based on STREK plots data in the Labanan forest. The total plot was 48 ha with 4 different treatments, using 28 years of data measurements. The biomass recovery values pattern is based on stand structure pattern, basal area, volume and aboveground biomass values. The biodiversity loss indicated by quantitative ecological values included a heterogeneity index, richness index, and similarity index. Fluctuations on stand structure changes followed lognormal curves. This study demonstrates the importance of the RIL technique in maintaining existing stored biomass in standing stock and allowing 99 % recovery 27 years after logging. This technique could curtail biodiversity loss as indicated by maintaining the high heterogeneity and richness indices of species, but the similarity to the primary forest’s initial conditions will reach 60–70 %.

Reduced-Impact Logging Maintain High Moss Diversity in Temperate Forests

Forestry harvesting represents an important economic activity around the world. Habitat degradation due to forest harvesting contributes to biodiversity loss; therefore, it is necessary to implement logging management aimed at reducing its impact. Forest management by reduce-impact logging (RIL) involves cutting trees following regulations focused on diminishing the impact on biodiversity by following harvesting plans based on forestry inventories and participation of trained workers. In Mexico, RIL is applied mainly in temperate habitats and its effectiveness has been assessed based on vascular plants. In this study, we analyzed the diversity and community structure of terrestrial and epiphytic mosses in managed (sites number = 3) and conserved (sites number = 3) sites in the temperate forest of Sierra Juárez, Oaxaca, Mexico. Likewise, we evaluated the potential function of mosses as indicators of habitat degradation. Environmental variables were also quantified at local (canopy coverage, altitude, daily temperature, and light) and regional (total annual rainfall, orientation, and slope) scales to evaluate potential relationships with the community and species diversity. We documented 70 mosses species with a diversity (alfa, beta) and community structure similar between managed and conserved sites. For terrestrial mosses, we found marginal differences in their communities, likely related to species coverture variation in managed sites. The diversity and community structure epiphytic mosses were not statistically different in managed and conserved sites. Only the daily variation in light intensity was positively related to the variation of alpha diversity of epiphytic mosses. The species Dicranum sumichrastii Duby and Leptodontium viticulosoides (P. Beauv.) Wijk & Margad. can be considered as ecological indicators for conserved and managed sites, respectively, likely due to their relationship with light and humidity conditions. Our results suggest that that forest management by RIL could be considered as a promising tool to balance timber production and moss diversity.

Distribuição diamétrica de três espécies de Lecythidaceae após exploração de impacto reduzido na Amazônia Oriental

Eschweilera coriacea, Eschweilera ovata e Lecythis idatimon são espécies arbóreas tropicais potencialmente comerciais, portanto, amplamente colhidas na Amazônia Oriental. O objetivo deste trabalho foi avaliar o desempenho de três funções de densidade probabilísticas no ajuste da distribuição diamétrica de três espécies de Lecythidaceae, para o período entre um e dez anos após a colheita florestal, em área manejada, em Moju, Pará. Foram selecionadas nove clareiras provenientes da exploração de impacto reduzido realizada em 1997. Em torno de cada clareira foram instaladas quatro faixas de 10 x 50 m, começando da bordadura da clareira para dentro da floresta, nas direções norte, sul, leste e oeste, perfazendo uma amostragem de 1,8 ha. Foram mensurados todos os indivíduos com DAP ≥ 5 cm em duas medições (1998 e 2007). Foram testadas as funções de densidade probabilísticas Weibull 3P, Gamma e Weber. Para a seleção do melhor modelo, foram avaliados o valor D de Kolmogorov – Smirnov (α = 0,05), erro padrão da estimativa (Syx%), coeficiente de determinação ajustado (R² ajustado) e análise gráfica das curvas de frequências estimadas sobre o histograma das observadas. Em ambos os anos, a função Weibull 3P apresentou as melhores estatísticas de ajuste e maior precisão para descrever a distribuição diamétrica de Eschweilera coriacea e Lecythis idatimon, e o modelo de Weber para Eschweilera ovata. A função Gamma foi descartada para todas as espécies. Os resultados indicam a não viabilidade do uso de uma mesma função para descrever diferentes espécies, necessitando considerar as características intrínsecas de cada espécie.

Assessing habitat quality on alpha and beta diversity of Odonata larvae (Insect) in logging areas in Amazon forest

The present study evaluated the effects of habitat quality on the alpha and beta diversity of odonate larvae under conditions of reduced impact and conventional logging. We hypothesized that the variation in the abiotic conditions found in areas of conventional logging would result in a greater loss of alpha and beta diversity in these areas in comparison with areas of reduced-impact logging and native forest. The study area was located in northeastern Para state, in northern Brazil. We analyzed data from 10 control streams, located in areas of preserved native forest, 11 streams in forest harvested by reduced-impact logging, and nine streams in areas that had been logged conventionally. Environmental factors such as dissolved oxygen, woody debris, water temperature, sediments, and canopy cover were the principal predictors of the diversity of the Odonata. Our results showed that turnover was the principal component structuring beta diversity in the three areas (native forest, reduced impact and conventional logging). The results of the present study indicate that management initiatives based on reduced-impact logging techniques could be adopted to guarantee socio-economic benefits while minimizing the impacts of logging on local biodiversity.

The effect of different logging regimes on the ecomorphological structure of stream fish assemblages in the Brazilian Amazon

AbstractWe evaluated the effects of changes in habitat structure resulting from conventional logging (CL) and reduced-impact logging (RIL) on the ecomorphological structure of stream fish assemblages in the eastern Brazilian Amazon, investigating (1) which habitat characteristics are influenced by different logging methods, (2) the differences in the ecomorphological structure of fish assemblages between two logging methods and undisturbed forest (UF), and (3) the relationships between ecomorphological traits of fish assemblages and habitat variables. We measured 13 habitat variables and characterized 11 ecomorphological traits of 55 fish species collected in 34 sampling sites (stream stretches) located in CL, RIL, and UF. In logged areas, the streams present a fine sediment increase and reduced vegetation cover, while in the UF a greater amount of natural shelter and more cover vegetation were observed. These environmental changes selected ecomorphological traits of stream fish assemblages, recording a greater relative width of the mouth in UF, higher ventral flattening index in RIL, or greater relative head length, and higher caudal peduncle compression index in CL. In this sense, we recommend that logging practices, including RIL, should be reviewed as a means of adopting better strategies to minimize human disturbances in streams.

Amphibian responses to conventional and reduced impact logging

Unsustainable timber management, primarily via Conventional selective Logging (CL), has resulted in severe habitat degradation and biodiversity loss. Sustainable techniques, principally Reduced Impact Logging (RIL), are therefore essential for conserving highly threatened, physiologically sensitive taxa such as amphibians. In this study we compare the effects of CL and RIL on stream amphibians across a regeneration gradient in two forest reserves in Sabah, Malaysian Borneo. We analysed data from a community of stream amphibians across 29 transects subject to either CL (18 – 21 years since logging) or RIL (4 – 21). Multi-species occupancy modelling determined community and species responses to logging types, regeneration times and logging associated environmental factors. Diversity profiles, which capture representative diversity indices whilst accounting for community evenness, were calculated using occupancy model results. Our results indicated that several species and community occupancy generally responded negatively to logging associated covariates (lower aboveground carbon density and higher siltation). Stream breeding species occupancy was generally lower in CL compared to RIL sites, with higher generalist breeding species occupancy in CL sites. Diversity profiles identified higher anuran diversity and evenness in RIL compared to CL sites of the same age (18 – 21 years). Furthermore, anuran diversity and evenness exhibited a distinct recovery from 4 to 21 years following RIL. Our results provide strong evidence that RIL reduces the negative impacts on stream habitats, amphibian occupancy and diversity compared to CL, and suggest a far quicker recovery after RIL. We believe RIL presents a promising alternative for sustainable tropical timber management and stream amphibian conservation.

The reasons great ape populations are still abundant in logged concessions: Environmental drivers and the influence of management plans

Negative short-term impact of logging on great ape distribution has been investigated but very little information is available on the long-term impacts of selective logging and management plans on ape habitats at a broader scale. We carried out a habitat suitability analysis using the presence data of great apes to identify the spatial relationships between their locations, logging activities patterns, and other human impacts (due to activities in rural communities) in an FSC-certified forestry concession (388,949 ha) in Cameroon. First, field data showed that the two great ape species, gorilla (Gorilla gorilla gorilla) and chimpanzee (Pan troglodytes troglodytes), did not differ in habitat type selection and they particularly appreciated highly dense rainforests which also lie within selectively logged zones. Second, we used MAXENT software to relate ape presence data to a set of predictor variables which represent environmental conditions and human disturbance factors (poaching, farming, trees felled to collect honey, logging roads). The model predicted that half of the whole concession may be suitable for the two primate species. Disturbances stemming from rural communities mainly correspond to poaching in the framework or not of usage rights. Road density was the strongest predictor of habitat choice by the great apes. Whereas the great apes tend to avoid zones with high road density, their population abundance increases in areas with a certain level of old logging roads. The vast majority of the predicted suitable habitats corresponds to areas logged less than 17 years ago (46.8% of the whole suitable area) where sustainable management practices have begun to be implemented, and unlogged areas (34.30%). Unsuitable habitats were mostly located in areas exploited before 2003 (45.46% of the unsuitable zones) with no reduced-impact logging practices, and in unlogged forests (41.51%). Our findings tend to confirm that selective logging combined with reduced-impact logging techniques which were more prevalent in FSC-certified concessions, may maintain or increase the spatial distribution of great ape habitats.

Competition indices after reduced impact logging in the Brazilian Amazon

Competition for resources can affect growth and increase mortality in forest stands. The effects of this process are little known, especially in areas subjected to forest management which show distinct growth dynamics when compared to unmanaged natural areas. This study aimed to estimate, evaluate and select the best fit competition indices (CI) for individual trees in a managed forest in the eastern Amazon. The data used originated from 18 permanent plots of 1 ha each, which were monitored for 12 years after Reduced Impact Logging (RIL). For the competition analysis, 23 indices were tested. CIs were evaluated based on graphical analyses and linear correlation coefficients(r)between each index and variables, basal area growth(ΔG), probability of mortality(Pm), and post-logging period (PLP). In addition, the partial F-test was applied to verify how well the calculated CIs explain the variations observed in ΔGandP(m). The distance-independent index for BAL (Basal Area Larger) performed the best. For the distance-dependent and semi-independent indices, the best interactions with the variables tested occurred in radii of 15 and 20 m away from the object tree. The largest trees presented higher values ofΔG. While P(m)was similar among the three tree classes, larger trees had a slightly higher P(m)due to thier low density of individuals. The results obtained in this study may help to include CIs in growth and production models at the individual tree level in tropical rainforests, especially in those managed in the Brazilian Amazon.

ANALISA BIAYA PEMANENAN KAYU MENGGUNAKAN TEKNIK RIL (Reduced Impact Logging) DI IUPHHK-HA PT. WIJAYA SENTOSA, PAPUA BARAT

The aim of this study is to analyze the cost used of forest harvesting activities which consist of logging, skidding, loading, hauling, and unloading using the Reduced Impact Logging (RIL) technique in the natural forest of PT. Wijaya Sentosa, West Papua. Tools used for this research are stationaries, calculators, and laptop. The data that needs to be collected for cost analysis consists of productivity data as primary data that had taken beforehand, secondary data from the company consist of tree species, tree ages, price of heavy equipment, fuel and lubricant usage data, parts replacement data, and operator wage data. The results obtained from this study are the details of estimated production costs for skidding activities amounting to 101,791 IDR/m3, 5,978 IDR/m3 for loading activities, 3,714 IDR/m3 for unloading activities, and 42,767 IDR/m3 for transportation activities. The total cost of harvesting from logging to transportation is 168,750 IDR/m3.

Stream Amphibian Detectability and Habitat Associations in a Reduced Impact Logging Concession in Malaysian Borneo

Borneo's amphibians face an uncertain future due to high levels of forest degradation and a paucity of data for effective conservation management. Several studies identified strong species–habitat relationships in pristine and conventionally logged forests. However, these studies did not account for detectability or habitat associations in sustainably managed forest reserves. Here, we determined detectability and species habitat relationships in stream amphibians within the Deramakot forest reserve, a reduced impact logging concession in the Malaysian state of Sabah, northern Borneo. We analyzed data for 10 stream species collected along 32 stream transects. An occupancy modeling framework was used to determine the climatological, temporal, and environmental covariates associated with detection and occupancy probabilities. We identified high variability in detection probability between species, including significant associations with moon phase (six species), time since sunset (five species), humidity (five species), rainfall (four species), and temperature (three species). Stream slope and volume provided by far the best predictors of occurrence, with significant positive or negative associations with the occupancy of six species each. These associations were more similar to those found in pristine compared with conventionally logged habitats. The highly variable detectability associations within our amphibian community suggest a level of temporal separation in regard to activity and breeding phenology in these species. This stresses the importance of accounting for detection probability via surveying sites across varying climatic/temporal conditions to obtain a representative sample of amphibian communities in pristine and disturbed tropical forests.

Soil greenhouse gas fluxes following conventional selective and reduced-impact logging in a Congo Basin rainforest

Selective logging is among the main causes of tropical forest degradation, but little is known about its effects on greenhouse gas (GHG) fluxes from highly weathered Ferralsol soils in Africa. We measured soil CO2, N2O, and CH4 fluxes, and their soil controlling factors at two forests that had undergone conventional selective logging and reduced-impact logging in Cameroon. Each logging system had four replicate plots, each included the disturbed strata (road, logging deck, skidding trail, and felling gap) and an undisturbed reference area. Measurements were conducted monthly from September 2016 to October 2017. Annual GHG fluxes ranged from 4.9 to 18.6 Mg CO2–C, from 1.5 to 79 kg N2O–N, and from − 4.3 to 71.1 kg CH4–C ha−1 year−1. Compared to undisturbed areas, soil CO2 emissions were reduced and soil CH4 emissions increased in skidding trails, logging decks and roads (P < 0.01) whereas soil N2O emissions increased in skidding trails (P = 0.03–0.05). The combined disturbed strata had 28% decrease in soil CO2 emissions, 83% increase in soil N2O emissions, and seven times higher soil CH4 emissions compared to undisturbed area (P ≤ 0.01). However, the disturbed strata represented only 4–5% of the area impacted in both logging systems, which reduced considerably the changes in soil GHG fluxes at the landscape level. Across all strata, soil GHG fluxes were regulated by soil bulk density and water-filled pore space, indicating the influence of soil aeration and gas diffusion, and by soil organic carbon and nitrogen, suggesting the control of substrate availability on microbial processes of these GHG.

Resilience of terrestrial mammals to logging in an active concession in Sarawak, Borneo

Abstract Selective logging is very widespread across the tropics and can alter the habitat for myriad wildlife species. But while many studies have assessed the impacts of past logging on forest animals, far fewer have investigated how species respond to logging while the timber operations are actually going on. This is an important knowledge gap because, considering the prevalence of logging across the world, numerous areas will be undergoing active extraction at any given time. We compared the occurrence and diel activity patterns of individual species of medium- to large-bodied terrestrial mammals, as well as the richness of the entire assemblage, among sites that were either unlogged, had been logged historically, or had ongoing ‘reduced impact’ timber extraction in the Kapit Region of Sarawak, Malaysian Borneo. We found no significant differences in estimated occupancy or activity patterns of particular species, or in overall species richness, among logging treatments. Across sites, species richness in this area appeared to be as high as or higher than in many other parts of the state, including some protected areas. Though monitoring is needed to assess potential long-term impacts, our results suggest that reduced-impact logging could allow economic development that is sustainable for many wildlife populations.

Tropical forest management altered abundances of individual tree species but not diversity

In this study, we evaluated tree species abundance distribution in a Brazilian Amazon forest following reduced-impact logging and unlogged plots over two decades. A large body of literature shows changes in species abundance distributions (SADs) after disturbances in many systems and suggests that SADs are a useful tool for monitoring community recovery after impacts. We used Poisson lognormal distribution to described SAD, which has sigma and correlation parameters, for comparing species diversity before logging and over time. Sigma is a measure of alpha diversity and correlation expresses similarity between communities. Contrary to our expectations, we detected no changes for sigma, but correlation showed a decreasing trend, particularly after the first decade. Those changes were due to the increase in the abundance of pioneers and little recruitment of commercial species. Mortality was mainly attributed to the accidental death of non-targeted individuals. Because of the dominant role of random accidental tree deaths, the signature of logging operations was not detected immediately after logging using sigma. Forest management did not cause changes in the abundance structure of communities if we disregard species composition. However, considering species composition, and particularly, which species were common or rare, we did detect changes over time. The trends we have detected might be an early warning of a stronger long term tendency, which could affect abundance distributions. Due to the long lifecycle of trees and their slow growth rate, longer-term studies are needed to understand the impacts of logging on tropical forests.

Responses in growth and dynamics of the shade-tolerant species Theobroma subincanum to logging gaps in the Eastern Amazon

Aim of study: To assess responses of the shade-tolerant species Theobroma subincanum in relation to canopy gaps created by reduced impact logging (RIL).Materials and methods: A managed forest in themunicipality ofMoju, Pará state,Brazil, harvested in 1997 through RIL was monitored during 12 years (1998-2010). Nine logging gaps were selected and classified in small, medium, and large. Four 10 m x 50 m strips starting from the gap’s border towards the forest and following the directions of cardinal points were installed. Each strip was divided in five 10 m x 10 m plots. Density, diameter distribution (DBH ≥ 5 cm with intervals = 5 cm), and diameter growth were measured.Main results: No significant changes in seedling density of T. subincanum were found, and its diameter distribution followed the reverse “J” shape during all monitoring time. T. subincanum presented diameter growth of 0.15 cm year-1 with highest Periodic Annual Increment in diameter up to three years, and stabilization in nine years after RIL. The species responded to a growth gradient inversely proportional to the gap’s border distance (p = 0.001) but not to gap size and plots direction in cardinal points around the gap.Research highlights: Shade-tolerant species such as T. subincanum have sensible and positive growth responses to disturbances caused by RIL even when seedlings received low amounts of indirect sunlight. These positive responses should be considered in the management of production forests.Keywords: Ecological group; forest management; diameter distribution; reduced impact logging (RIL).

Impacts of selective logging on avian phylogenetic and functional diversity in the Amazon

AbstractSelective logging is the most extensive land use threatening tropical forests worldwide, making our understanding of its impacts on biodiversity vital for the conservation of these hyperdiverse ecosystems. We investigated the effects of selective logging on the phylogenetic and functional diversity (FD) of Amazonian forest avifauna. We sampled the overall and the understorey bird community using point counts and mist‐netting, respectively, in logged and unlogged forests. The overall bird community showed significantly lower phylogenetic diversity (PD) and FD in logged than unlogged forests, but significantly higher standard effect size (ses) of PD, ses mean pairwise distance (MPD), mean nearest taxon distance (MNTD), ses functional MPD (sesfMPD) and functional MNTD (fMNTD) in logged sites. Evolutionary distinctiveness, sesMPD, sesFD, fMNTD and sesfMNTD of understorey birds were significantly higher in logged forests. Higher logging intensity affected some metrics negatively and others positively, although the relatively small changes in phylogenetic and FD suggest limited overall effects from logging at lower intensity. In combination, selective logging brings deleterious effects for some components of phylogenetic and FD of the overall Amazonian bird community. Nonetheless, selectively logged forests retain important avian diversity, underscoring their high conservation value. Relatively, low impacts of selective logging may be due to the reduced‐impact logging (RIL) techniques used, high connectivity with well‐preserved unlogged forests or focus on short‐term logging impacts. Because even low‐intensity RIL causes some losses in evolutionary history and functional traits, an optimal approach for conserving phylogenetic and FD requires incorporating some land‐sparing logging to protect more old‐growth patches within the logged matrix.

Comparison of Statistical Modelling Approaches for Estimating Tropical Forest Aboveground Biomass Stock and Reporting Their Changes in Low-Intensity Logging Areas Using Multi-Temporal LiDAR Data

Accurately quantifying forest aboveground biomass (AGB) is one of the most significant challenges in remote sensing, and is critical for understanding global carbon sequestration. Here, we evaluate the effectiveness of airborne LiDAR (Light Detection and Ranging) for monitoring AGB stocks and change (ΔAGB) in a selectively logged tropical forest in eastern Amazonia. Specifically, we compare results from a suite of different modelling methods with extensive field data. The calibration AGB values were derived from 85 square field plots sized 50 × 50 m field plots established in 2014 and which were estimated using airborne LiDAR data acquired in 2012, 2014, and 2017. LiDAR-derived metrics were selected based upon Principal Component Analysis (PCA) and used to estimate AGB stock and change. The statistical approaches were: ordinary least squares regression (OLS), and nine machine learning approaches: random forest (RF), several variations of k-nearest neighbour (k-NN), support vector machine (SVM), and artificial neural networks (ANN). Leave-one-out cross-validation (LOOCV) was used to compare performance based upon root mean square error (RMSE) and mean difference (MD). The results show that OLS had the best performance with an RMSE of 46.94 Mg/ha (19.7%) and R² = 0.70. RF, SVM, and ANN were adequate, and all approaches showed RMSE ≤54.48 Mg/ha (22.89%). Models derived from k-NN variations all showed RMSE ≥64.61 Mg/ha (27.09%). The OLS model was thus selected to map AGB across the time-series. The mean (±sd—standard deviation) predicted AGB stock at the landscape level was 229.10 (±232.13) Mg/ha in 2012, 258.18 (±106.53) in 2014, and 240.34 (sd ± 177.00) Mg/ha in 2017, showing the effect of forest growth in the first period and logging in the second period. In most cases, unlogged areas showed higher AGB stocks than logged areas. Our methods showed an increase in AGB in unlogged areas and detected small changes from reduced-impact logging (RIL) activities occurring after 2012. We also detected that the AGB increase in areas logged before 2012 was higher than in unlogged areas. Based on our findings, we expect our study could serve as a basis for programs such as REDD+ and assist in detecting and understanding AGB changes caused by selective logging activities in tropical forests.