We outline the principles of the natural capital approach to decision making and apply these to the contemporary challenge of very significantly expanding woodlands as contribution to attaining net zero emissions of greenhouse gases. Drawing on the case of the UK, we argue that a single focus upon carbon storage alone is likely to overlook the other ‘net zero plus’ benefits which woodlands can deliver. A review of the literature considers the wide variety of potential benefits which woodlands can provide, together with costs such as foregone alternative land uses. We argue that decision making must consider all of these potential benefits and costs for the right locations to be planted with the right trees. The paper closes by reviewing the decision support systems necessary to incorporate this information into policy and decision making. Read the free Plain Language Summary for this article on the Journal blog.

To enable an evidence-based management of ecosystems to adapt to the climate crisis, we require fine spatiotemporal resolution estimates of carbon, water, and energy fluxes at the field scale. To overcome the limitations resulting from the coarse spatial resolution of existing flux products, e.g. 500 m (Running et al. 2015), and the challenges in matching eddy covariance (EC) footprints with land use field scales, we for the first time investigate the influence of satellite resolution on flux estimation, which is to support the fine-scale extrapolation of EC fluxes from the tower footprint to the field scale. We validate the extrapolation at 206 FLUXNET2015 tower sites to pave the way for estimating field-scale fluxes extrapolated from three towers in a managed European grazing pasture on a fine-scale, 30 m spatial resolution. The findings suggest that (a) tower-level flux estimates from 30 m satellites were in agreement with fluxes estimated from moderate-resolution satellites, which are extensively employed in literature (R2 difference ≪ 0.1); (b) flux estimates were in reasonable agreement with EC measurements (R2: 0.7 and annual bias < 2 Mg ha−1 yr−1 for carbon fluxes); (c) Sentinel-2 was advantageous in capturing land-use variability over other satellites in European pastures; (d) the machine-learning extrapolation algorithm was resistant to livestock grazing.

Eddy covariance serves as one the most effective techniques for long-term monitoring of ecosystem fluxes, however long-term data integrations rely on complete timeseries, meaning that any gaps due to missing data must be reliably filled. To date, many gap-filling approaches have been proposed and extensively evaluated for mature and/or less actively managed ecosystems. Random forest regression (RFR) has been shown to be stable and perform better in these systems than alternative approaches, particularly when filling longer gaps. However, the performance of RFR gap filling remains less certain in more challenging ecosystems, e.g. actively managed agri-ecosystems and following recent land-use change due to management disturbances, ecosystems with relatively low fluxes due to low signal to noise ratios, or for trace gases other than carbon dioxide (e.g. methane). In an extension to earlier work on gap filling global carbon dioxide, water, and energy fluxes, we assess the RFR approach for gap filling methane fluxes globally. We then investigate a range of gap-filling methodologies for carbon dioxide, water, energy, and methane fluxes in challenging ecosystems, including European managed pastures, Southeast Asian converted peatlands, and North American drylands. Our findings indicate that RFR is a competent alternative to existing research standard gap-filling algorithms. The marginal distribution sampling (MDS) is still suggested for filling short (< 12 days) gaps in carbon dioxide fluxes, but RFR is better for filling longer (> 30 days) gaps in carbon dioxide fluxes and also for gap filling other fluxes (e.g. sensible heat, latent energy and methane). In addition, using RFR with globally available reanalysis environmental drivers is effective when measured drivers are unavailable. Crucially, RFR was able to reliably fill cumulative fluxes for gaps > 3 moths and, unlike other common approaches, key environment-flux responses were preserved in the gap-filled data.

Drylands cover ca. 40% of the land surface and are hypothesised to play a major role in the global carbon cycle, controlling both long-term trends and interannual variation. These insights originate from land surface models (LSMs) that have not been extensively calibrated and evaluated for water-limited ecosystems. We need to learn more about dryland carbon dynamics, particularly as the transitory response and rapid turnover rates of semi-arid systems may limit their function as a carbon sink over multi-decadal scales. We quantified aboveground biomass carbon (AGC; inferred from SMOS L-band vegetation optical depth) and gross primary productivity (GPP; from PML-v2 inferred from MODIS observations) and tested their spatial and temporal correspondence with estimates from the TRENDY ensemble of LSMs. We found strong correspondence in GPP between LSMs and PML-v2 both in spatial patterns (Pearson’s r = 0.9 for TRENDY-mean) and in inter-annual variability, but not in trends. Conversely, for AGC we found lesser correspondence in space (Pearson’s r = 0.75 for TRENDY-mean, strong biases for individual models) and in the magnitude of inter-annual variability compared to satellite retrievals. These disagreements likely arise from limited representation of ecosystem responses to plant water availability, fire, and photodegradation that drive dryland carbon dynamics. We assessed inter-model agreement and drivers of long-term change in carbon stocks over centennial timescales. This analysis suggested that the simulated trend of increasing carbon stocks in drylands is in soils and primarily driven by increased productivity due to CO2 enrichment. However, there is limited empirical evidence of this 50-year sink in dryland soils. Our findings highlight important uncertainties in simulations of dryland ecosystems by current LSMs, suggesting a need for continued model refinements and for greater caution when interpreting LSM estimates with regards to current and future carbon dynamics in drylands and by extension the global carbon cycle.

. There is high potential for ecosystem restoration across tropical savannah-dominated regions, but the benefits that could be gained from this restoration are rarely assessed. This study focuses on the Brazilian Cerrado, a highly species-rich savannah-dominated region, as an exemplar to review potential restoration benefits using three metrics: net biomass gains, plant species richness and ability to connect restored and native vegetation. Localized estimates of the most appropriate restoration vegetation type (grassland, savannah, woodland/forest) for pasturelands are produced. Carbon sequestration potential is significant for savannah and woodland/forest restoration in the seasonally dry tropics (net biomass gains of 58.2 ± 37.7 and 130.0 ± 69.4 Mg ha
. −1
. ). Modelled restoration species richness gains were highest in the central and south-east of the Cerrado for savannahs and grasslands, and in the west and north-west for woodlands/forests. The potential to initiate restoration projects across the whole of the Cerrado is high and four hotspot areas are identified. We demonstrate that landscape restoration across all vegetation types within heterogeneous tropical savannah-dominated regions can maximize biodiversity and carbon gains. However, conservation of existing vegetation is essential to minimizing the cost and improving the chances of restoration success.
.
. This article is part of the theme issue ‘Understanding forest landscape restoration: reinforcing scientific foundations for the UN Decade on Ecosystem Restoration’.

Native vegetation across the Brazilian Cerrado is highly heterogeneous and biodiverse and provides important ecosystem services, including carbon and water balance regulation, however, land-use changes have been extensive. Conservation and restoration of native vegetation is essential and could be facilitated by detailed landcover maps. Here, across a large case study region in Goiás State, Brazil (1.1 Mha), we produced physiognomy level maps of native vegetation (n = 8) and other landcover types (n = 5). Seven different classification schemes using different combinations of input satellite imagery were used, with a Random Forest classifier and 2-stage approach implemented within Google Earth Engine. Overall classification accuracies ranged from 88.6-92.6% for native and non-native vegetation at the formation level (stage-1), and 70.7-77.9% for native vegetation at the physiognomy level (stage-2), across the seven different classifications schemes. The differences in classification accuracy resulting from varying the input imagery combination and quality control procedures used were small. However, a combination of seasonal Sentinel-1 (C-band synthetic aperture radar) and Sentinel-2 (surface reflectance) imagery resulted in the most accurate classification at a spatial resolution of 20 m. Classification accuracies when using Landsat-8 imagery were marginally lower, but still reasonable. Quality control procedures that account for vegetation burning when selecting vegetation reference data may also improve classification accuracy for some native vegetation types. Detailed landcover maps, produced using freely available satellite imagery and upscalable techniques, will be important tools for understanding vegetation functioning at the landscape scale and for implementing restoration projects.

Continuous time-series of CO2, water, and energy fluxes are useful for evaluating the impacts of climate-change and management on ecosystems. The eddy covariance (EC) technique can provide continuous, direct measurements of ecosystem fluxes, but to achieve this gaps in data must be filled. Research-standard methods of gap-filling fluxes have tended to focus on CO2 fluxes in temperate forests and relatively short gaps of less than two weeks. A gap-filling method applicable to other fluxes and capable of filling longer gaps is needed. To address this challenge, we propose a novel gap-filling approach, Random Forest Robust (RFR). RFR can accommodate a wide range of data gap sizes, multiple flux types (i.e. CO2, water and energy fluxes). We configured RFR using either three (RFR3) or ten (RFR10) driving variables. RFR was tested globally on fluxes of CO2, latent heat (LE), and sensible heat (H) from 94 suitable FLUXNET2015 sites by using artificial gaps (from 1 to 30 days in length) and benchmarked against the standard marginal distribution sampling (MDS) method. In general, RFR improved on MDS's R2 by 15% (RFR3) and by 30% (RFR10) and reduced uncertainty by 70%. RFR's improvements in R2 for H and LE were more than twice the improvement observed for CO2 fluxes. Unlike MDS, RFR performed well for longer gaps; for example, the R2 of RFR methods in filling 30-day gaps dropped less than 4% relative to 1-day gaps, while the R2 of MDS dropped by 21%. Our results indicate that the RFR method can provide improved gap-filling of CO2, H and LE flux timeseries. Such improved continuous flux measurements, with low bias, can enhance our understanding of the impacts of climate-change and management on ecosystems globally.

Estimates of peatland carbon fluxes based on remote sensing data are a useful addition to monitoring methods in these remote and precious ecosystems, but there are questions as to whether large-scale estimates are reliable given the small-scale heterogeneity of many peatlands. Our objective was to consider the reliability of models based on Earth Observations for estimating ecosystem photosynthesis at different scales using the Forsinard Flows RSPB reserve in Northern Scotland as our study site. Three sites across the reserve were monitored during the growing season of 2017. One site is near-natural blanket bog, and the other two are at different stages of the restoration process after removal of commercial conifer forestry. At each site we measured small (flux chamber) and landscape scale (eddy covariance) CO2 fluxes, small scale spectral data using a handheld spectrometer, and obtained corresponding satellite data from MODIS. The variables influencing GPP at small scale, including microforms and dominant vegetation species, were assessed using exploratory factor analysis. A GPP model using land surface temperature and a measure of greenness from remote sensing data was tested and compared to chamber and eddy covariance CO2 fluxes; this model returned good results at all scales (Pearson's correlations of 0.57 to 0.71 at small scale, 0.76 to 0.86 at large scale). We found that the effect of microtopography on GPP fluxes at the study sites was spatially and temporally inconsistent, although connected to water content and vegetation species. The GPP fluxes measured using EC were larger than those using chambers at all sites, and the reliability of the TG model at different scales was dependent on the measurement methods used for calibration and validation. This suggests that GPP measurements from remote sensing are robust at all scales, but that the methods used for calibration and validation will impact accuracy.

Need for regional economic development and global demand for agro-industrial commodities have resulted in large-scale conversion of forested landscapes to industrial agriculture across South East Asia. However, net emissions of CO2 from tropical peatland conversions may be significant and remain poorly quantified, resulting in controversy around the magnitude of carbon release following conversion. Here we present long-term, whole ecosystem monitoring of carbon exchange from two oil palm plantations on converted tropical peat swamp forest. Our sites compare a newly converted oil palm plantation (OPnew) to a mature oil palm plantation (OPmature) and combine them in the context of existing emission factors. Mean annual net emission (NEE) of CO2 measured at OPnew during the conversion period (137.8 Mg CO2  ha-1  year-1 ) was an order of magnitude lower during the measurement period at OPmature (17.5 Mg CO2  ha-1  year-1 ). However, mean water table depth (WTD) was shallower (0.26 m) than a typical drainage target of 0.6 m suggesting our emissions may be a conservative estimate for mature plantations, mean WTD at OPnew was more typical at 0.54 m. Reductions in net emissions were primarily driven by increasing biomass accumulation into highly productive palms. Further analysis suggested annual peat carbon losses of 24.9 Mg CO2 -C ha-1  year-1 over the first 6 years, lower than previous estimates for this early period from subsidence studies, losses reduced to 12.8 Mg CO2 -C ha-1  year-1 in the later, mature phase. Despite reductions in NEE and carbon loss over time, the system remained a large net source of carbon to the atmosphere after 12 years with the remaining 8 years of a typical plantation's rotation unlikely to recoup losses. These results emphasize the need for effective protection of tropical peatlands globally and strengthening of legislative enforcement where moratoria on peatland conversion already exist.

Physical and chemical stabilisation mechanisms are now known to play a critical role in controlling carbon (C) storage in mineral soils, leading to suggestions that climate warming-induced C losses may be lower than previously predicted. By analysing > 9,000 soil profiles, here we show that, overall, C storage declines strongly with mean annual temperature. However, the reduction in C storage with temperature was more than three times greater in coarse-textured soils, with limited capacities for stabilising organic matter, than in fine-textured soils with greater stabilisation capacities. This pattern was observed independently in cool and warm regions, and after accounting for potentially confounding factors (plant productivity, precipitation, aridity, cation exchange capacity, and pH). The results could not, however, be represented by an established Earth system model (ESM). We conclude that warming will promote substantial soil C losses, but ESMs may not be predicting these losses accurately or which stocks are most vulnerable.

The recent expansion of oil palm (OP, Elaeis guineensis) plantations into tropical forest peatlands has resulted in ecosystem carbon emissions. However, estimates of net carbon flux from biomass changes require accurate estimates of the above ground biomass (AGB) accumulation rate of OP on peat. We quantify the AGB stocks of an OP plantation on drained peat in Malaysia from 3 to 12 years after planting using destructive harvests supported by non-destructive surveys of a further 902 palms. Peat specific allometric equations for palm (R2 = 0.92) and frond biomass are developed and contrasted to existing allometries for OP on mineral soils. Allometries are used to upscale AGB estimates to the plantation block-level. Aboveground biomass stocks on peat accumulated at ~6.39 ± 1.12 Mg ha-1 per year in the first 12 years after planting, increasing to ~7.99 ± 0.95 Mg ha-1 yr-1 when a 'perfect' plantation was modelled. High inter-palm and inter-block AGB variability was observed in mature classes as a result of variations in palm leaning and mortality. Validation of the allometries defined and expansion of non-destructive inventories across alternative plantations and age classes on peat would further strengthen our understanding of peat OP AGB accumulation rates.

AbstractPotential land–climate feedbacks in subarctic regions, where rapid warming is driving forest expansion into the tundra, may be mediated by differences in transpiration of different plant functional types. Here, we assess the environmental controls of overstorey transpiration and its relevance for ecosystem evapotranspiration in subarctic deciduous woodlands. We measured overstorey transpiration of mountain birch canopies and ecosystem evapotranspiration in two locations in northern Fennoscandia, having dense (Abisko) and sparse (Kevo) overstories. For Kevo, we also upscale chamber‐measured understorey evapotranspiration from shrubs and lichen using a detailed land cover map. Subdaily evaporative fluxes were not affected by soil moisture and showed similar controls by vapour pressure deficit and radiation across sites. At the daily timescale, increases in evaporative demand led to proportionally higher contributions of overstorey transpiration to ecosystem evapotranspiration. For the entire growing season, the overstorey transpired 33% of ecosystem evapotranspiration in Abisko and only 16% in Kevo. At this latter site, the understorey had a higher leaf area index and contributed more to ecosystem evapotranspiration compared with the overstorey birch canopy. In Abisko, growing season evapotranspiration was 27% higher than precipitation, consistent with a gradual soil moisture depletion over the summer. Our results show that overstorey canopy transpiration in subarctic deciduous woodlands is not the dominant evaporative flux. However, given the observed environmental sensitivity of evapotranspiration components, the role of deciduous trees in driving ecosystem evapotranspiration may increase with the predicted increases in tree cover and evaporative demand across subarctic regions.

Tropical ecosystems harbour the highest concentrations of biodiversity on Earth and play a pivotal role in the global carbon cycle, yet deforestation and degradation continue unabated in many regions, with net forest loss at 5.5 million ha yr-1 between 2010 and 2015. Protected areas offer a partial solution to this problem, with a growing body of evidence demonstrating their effectiveness for habitat conservation in the dense forests of Amazonia, Central Africa and Southeast Asia. Despite containing over a quarter of global biodiversity hotspots and being low density but significant carbon stores, tropical drylands have received far less attention in conservation terms, and research into protected areas in these ecosystems is far more limited. The overall effectiveness of protected areas in different dryland regions, and the factors influencing performance, are less understood. By measuring protected area performance as a function of aboveground biomass change, this study investigated the effectiveness of protected areas in the savannah belt of Nigeria, a country with a long history of environmental degradation. L-band Synthetic Aperture Radar (SAR), a form of remote sensing that penetrates the vegetation canopy, provided a means of consistently monitoring aboveground biomass change over time. Twenty-one areas, ranging in size from 117,000 ha to 608,410 ha, and offering varying levels of protection according to IUCN designations, were selected, with aboveground biomass changes between 2007 and 2017 determined by subjecting L-band SAR data to a novel approach called ‘Biomass Matching’. The combination of SAR and Biomass Matching allowed aboveground biomass changes within these protected areas to be detected and estimated without the need for supplementary field data, which is usually required to calibrate such remote sensing data. All but four protected areas experienced increases in aboveground biomass over the study period, with mean change being +1.22 Mg ha-1, compared to +0.26 Mg ha-1 for a set of twelve similar unprotected areas. Furthermore, their performance was affected by an array of factors, though accessibility and management efficacy were deemed the most influential. These results suggest that, with appropriate monitoring and resourcing, protected areas in Nigerian dry forests and savannahs can provide effective habitat conservation, though more inaccessible areas will inherently perform better.

Aerated soils represent an important sink for atmospheric methane (CH4), due to the effect of methanotrophic bacteria, thus mitigating current atmospheric CH4 increases. Whilst rates of CH4 oxidation have been linked to types of vegetation cover, there has been no systematic investigation of the interaction between plants and soil in relation to the strength of the soil CH4 sink. We used quasi-continuous automated chamber measurements of soil CH4 and CO2 flux from soil collar treatments that selectively include root and ectomycorrhizal (ECM) mycelium to investigate the role of rhizosphere activity as well as the effects of other environmental drivers on CH4 uptake in a temperate coniferous forest soil. We also assessed the potential impact of measurement bias from sporadic chamber measurements in altering estimates of soil CO2 efflux and CH4 uptake. Results show a clear effect of the presence of live roots and ECM mycelium on soil CO2 efflux and CH4 uptake. The presence of ECM hyphae alone (without plant roots) showed intermediate fluxes of both CO2 and CH4 relative to soils that either contained roots and ECM mycelium, or soil lacking root- and ECM mycelium. Regression analysis confirmed a significant influence of soil moisture as well as temperature on flux dynamics of both CH4 and CO2 flux. We further found a surprising increase in soil CH4 uptake during the night, and discuss diurnal fluctuations in atmospheric CH4 (with higher concentrations during stable atmospheric conditions at night) as a potential driver of CH4 oxidation rates. Using the high temporal resolution of our data set, we show that low-frequency sampling results in systematic bias of up-scaled flux estimates, resulting in under-estimates of up to 20% at our study site, due to fluctuations in flux dynamics on diurnal as well as longer time scales.

Background

The success of satellites in mapping deforestation has been invaluable for improving our understanding of the impacts and nature of land cover change and carbon balance. However, current satellite approaches struggle to quantify the intensity of forest disturbance, i.e. whether the average rate of biomass loss for a region arises from heavy disturbance focused in a few locations, or the less severe disturbance of a wider area. The ability to distinguish between these, very different, disturbance regimes remains critical for forest managers and ecologists.
Results

We put forward a framework for describing all intensities of forest disturbance, from deforestation, to widespread low intensity disturbance. By grouping satellite observations into ensembles with a common disturbance regime, the framework is able to mitigate the impacts of poor signal-to-noise ratio that limits current satellite observations. Using an observation system simulation experiment we demonstrate that the framework can be applied to provide estimates of the mean biomass loss rate, as well as distinguish the intensity of the disturbance. The approach is robust despite the large random and systematic errors typical of biomass maps derived from radar. The best accuracies are achieved with ensembles of ≥1600 pixels (≥1 km 2 with 25 by 25 m pixels).
Summary

The framework we describe provides a novel way to describe and quantify the intensity of forest disturbance, which could help to provide information on the causes of both natural and anthropogenic forest loss—such information is vital for effective forest and climate policy formulation.

Quantifying landscape-scale methane (CH4) fluxes from boreal and arctic regions, and determining how they are controlled, is critical for predicting the magnitude of any CH4 emission feedback to climate change. Furthermore, there remains uncertainty regarding the relative importance of small areas of strong methanogenic activity, vs. larger areas with net CH4 uptake, in controlling landscape-level fluxes. We measured CH4 fluxes from multiple microtopographical subunits (sedge-dominated lawns, interhummocks and hummocks) within an aapa mire in subarctic Finland, as well as in drier ecosystems present in the wider landscape, lichen heath and mountain birch forest. An intercomparison was carried out between fluxes measured using static chambers, up-scaled using a high-resolution landcover map derived from aerial photography and eddy covariance. Strong agreement was observed between the two methodologies, with emission rates greatest in lawns. CH4 fluxes from lawns were strongly related to seasonal fluctuations in temperature, but their floating nature meant that water-table depth was not a key factor in controlling CH4 release. In contrast, chamber measurements identified net CH4 uptake in birch forest soils. An intercomparison between the aerial photography and satellite remote sensing demonstrated that quantifying the distribution of the key CH4 emitting and consuming plant communities was possible from satellite, allowing fluxes to be scaled up to a 100 km2 area. For the full growing season (May to October), ~ 1.1-1.4 g CH4 m-2 was released across the 100 km2 area. This was based on up-scaled lawn emissions of 1.2-1.5 g CH4 m-2, vs. an up-scaled uptake of 0.07-0.15 g CH4 m-2 by the wider landscape. Given the strong temperature sensitivity of the dominant lawn fluxes, and the fact that lawns are unlikely to dry out, climate warming may substantially increase CH4 emissions in northern Finland, and in aapa mire regions in general.

Quantifying landscape-scale methane (CH4 ) fluxes from boreal and arctic regions, and determining how they are controlled, is critical for predicting the magnitude of any CH4 emission feedback to climate change. Furthermore, there remains uncertainty regarding the relative importance of small areas of strong methanogenic activity, vs. larger areas with net CH4 uptake, in controlling landscape-level fluxes. We measured CH4 fluxes from multiple microtopographical subunits (sedge-dominated lawns, interhummocks and hummocks) within an aapa mire in subarctic Finland, as well as in drier ecosystems present in the wider landscape, lichen heath and mountain birch forest. An intercomparison was carried out between fluxes measured using static chambers, up-scaled using a high-resolution landcover map derived from aerial photography and eddy covariance. Strong agreement was observed between the two methodologies, with emission rates greatest in lawns. CH4 fluxes from lawns were strongly related to seasonal fluctuations in temperature, but their floating nature meant that water-table depth was not a key factor in controlling CH4 release. In contrast, chamber measurements identified net CH4 uptake in birch forest soils. An intercomparison between the aerial photography and satellite remote sensing demonstrated that quantifying the distribution of the key CH4 emitting and consuming plant communities was possible from satellite, allowing fluxes to be scaled up to a 100 km(2) area. For the full growing season (May to October), ~ 1.1-1.4 g CH4  m(-2) was released across the 100 km(2) area. This was based on up-scaled lawn emissions of 1.2-1.5 g CH4  m(-2) , vs. an up-scaled uptake of 0.07-0.15 g CH4  m(-2) by the wider landscape. Given the strong temperature sensitivity of the dominant lawn fluxes, and the fact that lawns are unlikely to dry out, climate warming may substantially increase CH4 emissions in northern Finland, and in aapa mire regions in general.

AbstractCarbon emissions from tropical land‐use change are a major uncertainty in the global carbon cycle. In African woodlands, small‐scale farming and the need for fuel are thought to be reducing vegetation carbon stocks, but quantification of these processes is hindered by the limitations of optical remote sensing and a lack of ground data. Here, we present a method for mapping vegetation carbon stocks and their changes over a 3‐year period in a &gt; 1000 km2 region in central Mozambique at 0.06 ha resolution. L‐band synthetic aperture radar imagery and an inventory of 96 plots are combined using regression and bootstrapping to generate biomass maps with known uncertainties. The resultant maps have sufficient accuracy to be capable of detecting changes in forest carbon stocks of as little as 12 MgC ha−1 over 3 years with 95% confidence. This allows characterization of biomass loss from deforestation and forest degradation at a new level of detail. Total aboveground biomass in the study area was reduced by 6.9 ± 4.6% over 3 years: from 2.13 ± 0.12 TgC in 2007 to 1.98 ± 0.11 TgC in 2010, a loss of 0.15 ± 0.10 TgC. Degradation probably contributed 67% (96.9 ± 91.0 GgC) of the net loss of biomass, but is associated with high uncertainty. The detailed mapping of carbon stock changes quantifies the nature of small‐scale farming. New clearances were on average small (median 0.2 ha) and were often additions to already cleared land. Deforestation events reduced biomass from 33.5 to 11.9 MgC ha−1 on average. Contrary to expectations, we did not find evidence that clearances were targeted towards areas of high biomass. Our method is scalable and suitable for monitoring land cover change and vegetation carbon stocks in woodland ecosystems, and can support policy approaches towards reducing emissions from deforestation and degradation (REDD).

AbstractData assimilation (DA) is increasingly being employed to estimate the parameters and states of terrestrial ecosystem models from eddy covariance measurements of net carbon (C) fluxes. The length of the observation time series used varies for each study. The impact of these differences has not been quantified explicitly. Therefore, in this study, we investigate the importance of the time series length relative to observation noise and data gaps. Different length synthetic time series are used to determine the parameter and C stocks of a simple ecosystem C model. Two commonly used DA schemes are tested: the sequential Ensemble Kalman Filter (EnKF) and a batch Metropolis Markov chain Monte Carlo algorithm. Longer time series improve both the parameter and C pool estimates of the EnKF, while adversely affecting those of the Metropolis algorithm. For both DA approaches, the length of the time series has more influence on the parameter and pool estimates than the level of random noise or amount of data. In this study, the EnKF provides more robust parameter and C pool estimates than the Metropolis algorithm. Optimized parameters and states are often used as the basis for forecasting future responses. Despite having better parameter and C pool estimates, EnKF forecasts estimates have much larger uncertainties than the Metropolis algorithm forecast estimates. Finally, we suggest that the structure of simple box models, as used in this study, introduces a large degree of equifinality into DA. Neither DA scheme correctly accounts for the equifinality, but our results suggest that it is particularly problematic for the batch Metropolis algorithm.