Estimates Of Carbon Stocks Research Articles

A Blueprint for the Estimation of Seagrass Carbon Stock Using Remote Sensing-Enabled Proxies

Seagrass ecosystems sequester carbon at disproportionately high rates compared to terrestrial ecosystems and represent a powerful potential contributor to climate change mitigation and adaptation projects. However, at a local scale, rich heterogeneity in seagrass ecosystems may lead to variability in carbon sequestration. Differences in carbon sequestration rates, both within and between seagrass meadows, are related to a wide range of interrelated biophysical and environmental variables that are difficult to measure holistically using traditional field surveys. Improved methods for producing robust, spatially explicit estimates of seagrass carbon storage across large areas would be highly valuable, but must capture complex biophysical heterogeneity and variability to be accurate and useful. Here, we review the current and emerging literature on biophysical processes which shape carbon storage in seagrass beds, alongside studies that map seagrass characteristics using satellite remote sensing data, to create a blueprint for the development of remote sensing-enabled proxies for seagrass carbon stock and sequestration. Applications of satellite remote sensing included measuring seagrass meadow extent, estimating above-ground biomass, mapping species composition, quantifying patchiness and patch connectivity, determining broader landscape environmental contexts, and characterising seagrass life cycles. All of these characteristics may contribute to variability in seagrass carbon storage. As such, remote sensing methods are uniquely placed to enable proxy-based estimates of seagrass carbon stock by capturing their biophysical characteristics, in addition to the spatiotemporal heterogeneity and variability of these characteristics. Though the outlined approach is complex, it is suitable for accurately and efficiently producing a full picture of seagrass carbon stock. This review has drawn links between the processes of seagrass carbon sequestration and the capabilities of remote sensing to detect and characterise these processes. These links will facilitate the development of remote sensing-enabled proxies and support spatially explicit estimates of carbon stock, ensuring climate change mitigation and adaptation projects involving seagrass are accounted for with increased accuracy and reliability.

Direct use of large-footprint lidar waveforms to estimate aboveground biomass

Many studies have established the strong connections between aboveground biomass and lidar height metrics; however, these relationships are site-specific. Field data required to derive these relationships are not readily available in many cases. We developed a model to estimate plot-level aboveground biomass density (AGBD) directly from large-footprint lidar waveform measurements. An individual tree-based aboveground biomass (AGB)-height allometric relationship was scaled up to the plot level using lidar-waveform sensed tree height and crown size distribution characteristics. The AGBD was estimated based on a waveform/foliage profile-weighted height-based allometric equation. The AGBD-height scaling exponent was then built on the allometric relationships of tree height with stem diameter and crown volume with tree height. Global vegetation structure data analysis demonstrated that one general model (scaling exponent ~ 1.6–1.8) works reasonably well across all global forest biomes except boreal forests (scaling exponent ~ 0.9). We applied the model to estimate aboveground biomass in two distinct geographic regions: temperate deciduous/conifer forests in the northeastern USA and a montane conifer forest in Sierra National Forest in California. Local vegetation structural data analysis leads to a consistent height scaling exponent for these two distinct biomes, slightly different from the global data analysis results. This model produced optimal AGBD estimates using the local height scaling exponent value. Adequate AGBD estimates with the general height scaling exponent value were also provided by our model. Our analysis suggests one general allometric relationship between plot-level AGBD and large-footprint lidar waveforms. Integrating local structure allometric relationships improve the predictive accuracy of the model. Our model outperformed the lidar height metrics-based approach for AGBD estimates and overcame the biomass underestimation problem using height metrics for high biomass regions. This model could potentially serve as a general and robust model for monitoring forest carbon stocks using large-footprint lidar waveform measurements such as the Global Ecosystem Dynamics Investigation (GEDI) mission at the continental and global scales. The model could be a framework for integrating a demography-based terrestrial ecosystem model and GEDI global mission measurements to improve global carbon stock and flux estimates.

Contributions of three types of biocrusts to soil carbon stock and annual efflux in a small watershed of Northern Chinese Loess Plateau

Accurate estimation of soil carbon (C) stock and efflux in drylands is vital for understanding the C cycling processes and predicting global climate change. As a “living skin”, biocrusts strongly change soil C fixation and respiration, but till now it is still challenging to accurately estimate and assess biocrust's contributions to soil C stock and efflux, especially taking different types of biocrusts and annual scale into consideration. In a small representative watershed on the northern Chinese Loess Plateau, cyanobacterial crust (cyano crust), cyanobacteria and moss mixed crust (mixed crust), moss crust, and bare soil were sampled. The organic C content of the samples were determined in laboratory, and subsequently their C stocks were calculated. Moreover, the respiration rate (Rs) of biocrusts and bare soil were repetitively measured in field for a four-month period, respectively, and an optimal regression model was established based on the relationship between Rs and corresponding temperature and moisture of soil. Lastly, the daily Rs was extended to a whole year through model extrapolation, and the seasonal and annual C efflux was obtained through summing the daily respiration together. Our results showed that (i) the organic C content of three types of biocrusts were 69 %–159 % higher than that of bare soil, and the soil C stock at depth of 0–5 cm was 0.35, 0.51, 0.81, and 0.61 kg m−2 for bare soil, cyano crust, mixed crust, and moss crust, respectively. (ii) During the measurement period, the mean Rs of the four habitats were 1.13, 2.21, 2.86, and 2.56 μmol m−2 s−1, respectively, showing that the biocrusts increased Rs by 96 %–152 % in contrast to bare soil. (iii) The variations of Rs were best explained by soil temperature and moisture through the Gaussian model, which had the lowest AIC and RMSE. (iv) Based on model extrapolation, the estimated annual C efflux from bare soil, cyano crust, mixed crust, and moss crust were 228.0, 378.0, 894.0, and 679.0 g m−2, respectively, indicating that in annual scale biocrusts increased C efflux by 66 %–292 % compared with bare soil. The seasonal variations also implied that the C efflux of biocrusts in summer and autumn were significantly higher than that in winter and spring, but they all accounted for a considerable proportion and should not be neglected. In conclusion, different types of biocrusts are the dominant contributors of soil C fixation and respiration in drylands. More importantly, the increased soil C efflux from biocrusts is highly overwhelmed by their elevated photosynthetic capacity, which make biocrusts sequestrate a considerable amount of C. These effects on soil C are highly associated with biocrust's roles in restoring degraded soil and mitigating global climate warming.

Size‐dependent intraspecific variation in wood traits has little impact on aboveground carbon estimates in a tropical forest landscape

Abstract There is increasing evidence that intraspecific trait variation plays a role in governing rates of ecosystem functioning. While wood traits such as wood specific gravity (WSG) and wood carbon concentration (WCC) are key drivers of forest aboveground carbon (AGC) stocks, the sources of intraspecific variation in these wood traits and the consequences of this variation on AGC are poorly known, especially in the tropics. Here, we investigated intraspecific variation in wood specific gravity (WSG) and wood carbon concentration (WCC) from 556 individual trees belonging to 15 species that well characterize different successional stages of seasonal evergreen forests in Southeast Asia. Specifically, we tested the contribution of individual or species characteristics (tree size, growth rate and regeneration guilds) and local environmental conditions (topographic wetness index and successional stages) to intraspecific variation in WSG and WCC, and assessed the consequences of intraspecific variation in these wood traits on AGC estimates in 14 permanent forest plots established along a successional gradient in Khao Yai National park, Thailand. We found that tree size was the main driver of intraspecific variation in WSG and WCC as tree sizes increased from 10–100 cm in diameter, WSG increased by 7.3%, while WCC increased by 2.4% in heartwood, 1.6% and 2.7% in sapwood without and with volatile carbon included. There was no effect of the topographic wetness and other local environment condition in wood traits led to a slight overestimation of AGC in young secondary forests (+0.09% to +1.29%) and a small underestimation in older forests (−0.86% to −2.87%), but overall AGC estimates (13 of 14 forest plots) remained within error margins (the 95% interval). Our study provides evidence that tree size variation translates into intraspecific variability in wood traits, whereas local environmental conditions related to topography successional stages had no effect on wood trait variability. While size‐dependent variation in WSG and WCC have largely been undocumented and thus ignored in forest carbon assessment approaches, we highlight that it has a limited impact on AGC estimates, indicating that it does not invalidate current forest carbon stock estimation approaches. Read the free Plain Language Summary for this article on the Journal blog.

Global stocks and capacity of mineral-associated soil organic carbon

Soil is the largest terrestrial reservoir of organic carbon and is central for climate change mitigation and carbon-climate feedbacks. Chemical and physical associations of soil carbon with minerals play a critical role in carbon storage, but the amount and global capacity for storage in this form remain unquantified. Here, we produce spatially-resolved global estimates of mineral-associated organic carbon stocks and carbon-storage capacity by analyzing 1144 globally-distributed soil profiles. We show that current stocks total 899 Pg C to a depth of 1 m in non-permafrost mineral soils. Although this constitutes 66% and 70% of soil carbon in surface and deeper layers, respectively, it is only 42% and 21% of the mineralogical capacity. Regions under agricultural management and deeper soil layers show the largest undersaturation of mineral-associated carbon. Critically, the degree of undersaturation indicates sequestration efficiency over years to decades. We show that, across 103 carbon-accrual measurements spanning management interventions globally, soils furthest from their mineralogical capacity are more effective at accruing carbon; sequestration rates average 3-times higher in soils at one tenth of their capacity compared to soils at one half of their capacity. Our findings provide insights into the world’s soils, their capacity to store carbon, and priority regions and actions for soil carbon management.

입목수간재적표 기반의 입목재적 추정의 불확실성과 이에 따른 탄소축적 추정의 편향: 강원도 홍천군을 대상으로

The purpose of this study was to quantify the uncertainty in the estimation of forest growing stock and carbon stock that occurs as DBH(Diameter at Breast Height) and HT (Height) are rounded when locating stem volumes in a stem volume table, and to present a case study using the Hongcheon-gun NFI (National Forest Inventory) data. We quantified the uncertainty of the estimated value of the stem volume obtained from the stem volume table, assuming that the stem volume calculated by the stem taper equation is the true value, and the resulting bias in carbon stock estimation, using two indicators. In all the indices, as DBH and HT increased, a trend was observed in which the absolute values of these indicators increased. BIAS appeared as a positive value in all sections, which means that the stem volume table with rounding DBH and HT underestimated stem volumes. BIAS was found to range from 0.0134m³ to 0.0178m³ depending on the species. RMSE (Root Mean Square Error) was about 4 ~ 7 times greater than BIAS with the same tree sizes. The increase rate of stem volume was almost constant regardless of HT, but increased as DBH increased. This indicates that DBH is more effective in reducing rounding-induced uncertainty when using stem volume tables. Note that rounding DBH combined with this characteristic led to systematic underestimation of stem volume. As a result of estimating the growing stock in the forests of Hongcheon-gun, Gangwon-do using the stem taper equation and the stem volume table, it was confirmed that the stem volume table underestimated the growing stock by 1,475,606m³, and accordingly, did the carbon stock by 860,232 tC. Even if the figure was limited to Hongcheon-gun, it was difficult to consider it a small value. As projects related to greenhouse gas reduction are expected to become more active in the future compared to the past, uncertainty in such estimates may affect the reliability of such projects. Therefore, it is necessary to provide a tool that can calculate stem volume and carbon stock by directly inputting DBH or HT values.

A mid-20th century inventory-based estimate of global terrestrial vegetation carbon stocks

A mid-20th century inventory-based estimate of global terrestrial vegetation carbon stocks

ReforesTree: A Dataset for Estimating Tropical Forest Carbon Stock with Deep Learning and Aerial Imagery

Forest biomass is a key influence for future climate, and the world urgently needs highly scalable financing schemes, such as carbon offsetting certifications, to protect and restore forests. Current manual forest carbon stock inventory methods of measuring single trees by hand are time, labour, and cost intensive and have been shown to be subjective. They can lead to substantial overestimation of the carbon stock and ultimately distrust in forest financing. The potential for impact and scale of leveraging advancements in machine learning and remote sensing technologies is promising, but needs to be of high quality in order to replace the current forest stock protocols for certifications. In this paper, we present ReforesTree, a benchmark dataset of forest carbon stock in six agro-forestry carbon offsetting sites in Ecuador. Furthermore, we show that a deep learning-based end-to-end model using individual tree detection from low cost RGB-only drone imagery is accurately estimating forest carbon stock within official carbon offsetting certification standards. Additionally, our baseline CNN model outperforms state-of-the-art satellite-based forest biomass and carbon stock estimates for this type of small-scale, tropical agro-forestry sites. We present this dataset to encourage machine learning research in this area to increase accountability and transparency of monitoring, verification and reporting (MVR) in carbon offsetting projects, as well as scaling global reforestation financing through accurate remote sensing.

ACE R-CNN: An Attention Complementary and Edge Detection-Based Instance Segmentation Algorithm for Individual Tree Species Identification Using UAV RGB Images and LiDAR Data

Accurate and automatic identification of tree species information at the individual tree scale is of great significance for fine-scale investigation and management of forest resources and scientific assessment of forest ecosystems. Despite the fact that numerous studies have been conducted on the delineation of individual tree crown and species classification using drone high-resolution red, green and blue (RGB) images, and Light Detection and Ranging (LiDAR) data, performing the above tasks simultaneously has rarely been explored, especially in complex forest environments. In this study, we improve upon the state of the Mask region-based convolution neural network (Mask R-CNN) with our proposed attention complementary network (ACNet) and edge detection R-CNN (ACE R-CNN) for individual tree species identification in high-density and complex forest environments. First, we propose ACNet as the feature extraction backbone network to fuse the weighted features extracted from RGB images and canopy height model (CHM) data through an attention complementary module, which is able to selectively fuse weighted features extracted from RGB and CHM data at different scales, and enables the network to focus on more effective information. Second, edge loss is added to the loss function to improve the edge accuracy of the segmentation, which is calculated through the edge detection filter introduced in the Mask branch of Mask R-CNN. We demonstrate the performance of ACE R-CNN for individual tree species identification in three experimental areas of different tree species in southern China with precision (P), recall (R), F1-score, and average precision (AP) above 0.9. Our proposed ACNet–the backbone network for feature extraction–has better performance in individual tree species identification compared with the ResNet50-FPN (feature pyramid network). The addition of the edge loss obtained by the Sobel filter further improves the identification accuracy of individual tree species and accelerates the convergence speed of the model training. This work demonstrates the improved performance of ACE R-CNN for individual tree species identification and provides a new solution for tree-level species identification in complex forest environments, which can support carbon stock estimation and biodiversity assessment.

Allometric equations for biomass and carbon stock estimation of small diameter woody species from tropical dry deciduous forests: Support to REDD+

In order to assess the contribution to the overall carbon stock and generate carbon credits under REDD+, it is essential to have an accurate estimation of biomass of different forest components. Forest cover in India is gradually increasing due to the active restoration of degraded land and plantation on waste and barren lands leading to an increase in the abundance of small diameter woody species. However, these were not included in biomass studies due to the non- availability of adequate allometric equations and low carbon stocks compared to mature individuals. We have harvested 589 individuals belonging to 23 woody species at the seedling and sapling stage from a tropical dry deciduous forest and developed species specific allometric equation and general allometric equation for aboveground biomass estimation. Further, the belowground biomass equation of 9 species were also developed using above ground biomass and root to shoot ratio as predictor variable. In the case of general equation, the combination of diameter with height and diameter, height and wood specific gravity exhibited highest adjusted R2 value. In case of species-specific allometric equations, combination of diameter with height predicts above ground biomass more precisely as compared to the diameter and wood specific gravity. Since the estimation of wood specific gravity requires destruction of lower diameter individuals, usage of diameter and height for biomass estimation would help protecting regeneration as both methods yield same results. All the equations developed in the present study for below ground biomass predicts biomass precisely. We suggest use of species specific allometric model developed with diameter and height for estimation of biomass. Further, general models consisting of height and diameter may be used for biomass estimation in case of non-availability of species specific equations without destroying the regeneration.

Estimating height-diameter relations for structure groups in the natural forests of Northeastern China

Tree height-diameter (H-D) models are essential for estimating tree volume and forest production. The development of such models can be especially challenging in multi-species natural forests. We present new generalized mixed-effects models based on a set of 456 sample plots and 77,908 individual tree measurements. Tree species were assigned to one of four structure groups according to their life forms, regeneration modes and stand structure characteristics: (1) Understory tree species with an L-shaped D distribution (UL); (2) Sub-canopy tree species with an L-shaped D distribution (SL); (3) Sub-canopy tree species with a bell-shaped D distribution (SB); and (4) Canopy tree species with a bell-shaped D distribution (CB). A distance-dependent competition index (DCI) was included in the base model of the structure groups, and a mixed-effects model for each group with tree species as random-effect. The four models represent unique allometric relationships for the four structure groups. Model accuracy was improved for all groups after adding DCI. Two sets of mixed-effects H-D models (with or without DCI) are presented to improve the estimates of productivity and carbon stock in the temperate natural forests of Northeastern China. The general approach adopted in this study may serve as an example for similar studies in other multi-species and multi-layered forest stands.

Research on monitoring overground carbon stock of forest vegetation communities based on remote sensing technology

The monitoring of forest carbon stocks provides a better understanding of the carbon sequestration capacity of forests. This paper took forest vegetation communities in Henan as an example and extracted waveband information, texture features and vegetation features from Landsat8 remote sensing image data as modeling factors to establish a multiple regression model and a back-propagation neural network (BPNN) model for overground carbon stock. It was found that the BPNN model had a $${R}^{2}$$ value value 0.678, a relative error of − 1.89%, and a root-mean-square error value of 13.21 t/hm2, indicating higher accuracy and more accurate estimation of overground carbon stock. The results verify the reliability of the remote sensing technology and the BPNN model in monitoring overground carbon stock, providing some theoretical bases for research on the carbon fixation capability of forest vegetation communities.

Aboveground Carbon Stock Estimation Model Using Sentinel-2A Imagery in Mbeliling Lanscape in Nusa Tenggara Timur, Indonesia

To determine emission levels, information on carbon stocks and changes in each carbon pool is required. Aboveground biomass, particularly on dry land, is one carbon pool that contributes significantly to carbon storage. The goal of this study was to develop a model for estimating aboveground carbon stocks in the Mbeliling landscape, in Nusa Tenggara Timur, using a vegetation index that was correlated with field carbon stocks. The best model was then used to create a map of the distribution of carbon stocks as the final result. Simple linear regression analysis and multiple linear regression analysis were used in the study. Google Earth Engine was used to process the images on a cloud system. When comparing the RGI index for measuring field carbon stocks to other indexes, the correlation test revealed a perfect correlation. The linear regression model for aboveground biomass = 14.046 + 272.496 RGI (R-sq = 0.86) was found to be the best model for aboveground biomass. In the multiple linear regression model, there were signs of multicollinearity. With an overall accuracy of 68% and a cappa accuracy of 54.23%, the best model was able to be used to create a carbon stock map in Mbeliling landscape.
 Keywords: Carbon stock estimation model, Above Ground Biomass, Sentinel 2A

Crop Production and Carbon Sequestration Potential of Grewia oppositifolia-Based Traditional Agroforestry Systems in Indian Himalayan Region

Bhimal (Grewia oppositifolia) is the most important agroforestry tree species used for fodder, fuel and fiber in the Himalayan region. In the present study, G. oppositifolia-based traditional agroforestry systems were selected for the estimation of carbon stock and the production potential of barnyard millet (Echinochloa frumentacea) and finger millet (Eleusine coracana), with two elevational ranges, i.e., 1000–1400 and 1400–1800 m amsl, in Garhwal Himalaya, India. The results of the investigation showed a decline in the growth and yield attributes of both the millet crops under the G. oppositifolia-based agroforestry system at both elevations as compared to their respective control sites (sole crops). Among the elevations, the total number of tillers per plant (2.70 and 2.48), the number of active tillers per plant (2.18 and 2.25), panicle length (17.63 cm and 6.95 cm), 1000-seed weight (5.49 g and 4.33 g), grain yield (10.77 q ha−1 and 11.35 q ha−1), straw yield (37.43 q ha−1 and 30.15 q ha−1), biological yield (48.21 q ha−1 and 41.51 q ha−1) and the harvest index (22.53% and 27.78%) were recorded as higher in the lower elevation in both E. frumentacea and E. coracana, respectively, while plant population per m2 (18.64 and 25.26, respectively) was recorded as higher in the upper elevation. Plant height for E. frumentacea (180.40 cm) was also observed to be higher in the upper elevation, while for E. coracana (98.04 cm), it was recorded as higher in the lower elevation. Tree carbon stock was reported negatively with an increase in altitude. The maximum amount of sequestered carbon in the tree biomass for G. oppositifolia was 23.29 Mg ha−1 at the lower elevation and 18.09 Mg ha−1 at the upper elevation. Total carbon stock in the tree biomass was reported to be the highest (15.15 Mg ha−1) in the 10–20 cm diameter class, followed by 20–30 cm (6.99 Mg ha−1), >30 cm (2.75 Mg ha−1) and the lowest (2.32 Mg ha−1) in the <10 cm diameter class. The results show that the yield of E. frumentacea and E. coracana was not reduced so severely under the G. oppositifolia system; however, keeping in mind the other benefits of this multipurpose tree, i.e., carbon sequestration and socioecological relevance, farmers can get benefit from adopting these crops under G. oppositifolia-based agroforestry systems.

Spatial patterns and driving factors of carbon stocks in mangrove forests on Hainan Island, China

AbstractAimMangrove forests are important coastal wetlands for the blue carbon budget and play a significant role in mitigating global climate change. However, spatial patterns of carbon stocks in mangrove forests on an island scale have not been quantified owing to methodological limitations and lack of understanding of controlling factors. We took the entire Hainan Island as a case study and aimed to carry out a comprehensive investigation of the spatial patterns and driving factors of carbon stocks in mangrove forests.LocationSouthern China.Time period2017–2020.Major taxa studiedMangrove forest.MethodsThe upscaling method combined with field surveys and Sentinel‐2 imagery analysis were used to compare different models for optimization of mangrove ecosystem carbon stock estimations. We also used structural equation modelling (SEM) to evaluate the factors driving the distributional patterns of mangrove carbon stocks on an island scale.ResultsThe current total mangrove carbon stock of the entire Hainan Island was estimated to be 703,181 Mg C (with a mean density of 192 Mg C/ha), with the above‐ and below‐ground carbon stocks averaging at 44.7 and 147.3 Mg C/ha, respectively. The mangrove carbon storage in the north‐eastern region of the island was the highest and in the west region the lowest. Sediment nitrogen content and plant species diversity had the most positive driving effects on the distribution of total carbon stock for Hainan Island mangroves.Main conclusionsThe combination of field surveys and Sentinel‐2 imagery analysis can be applied to regional‐scale estimations of carbon stocks for mangrove forests. Spatial pattens of mangrove carbon stocks vary among locations on Hainan Island, and soil nutrient (especially nitrogen) availability is the dominant factor regulating carbon stock variations along the land‐to‐sea gradient. Our findings have significant implications for better understanding the distribution of mangrove carbon stocks and quantification of the global blue carbon budget.

Assessing intraspecific wood density variations of Syzgium sp. in tropical forest of Southwest Sabah

Wood density (WD) is a critical determinant of estimating forest above-ground biomass (AGB) and carbon stock. Thus, heterogeneity in WD on individuals within species trees needs to be scrutinized, and acquisition of fixed WD value is essential to estimate carbon stock with confidence. This study investigated intraspecific variation in WD of Syzgium sp., also known as “Jambu” or “Kelat”. It is the most occurring species in study areas, and is regarded as an economically important species. Firstly, one half-diameter drilling from bark-to-pith measurement was taken per tree using Rinntech Resistograph R650-ED at breast height. Meanwhile, 5.15 mm-diameter core was sampled at 1.30 m above-ground, with DeWalt DCF899HP2 20V impact wrench 950 Nm and Haglöf increment borer. WD was estimated for each core sample using a dimensional method. Drilling resistance (DR) profiles were processed using DECOM 2.38m1 Scientific (c), and several independent variables were extracted from the resistogram. All resistogram-derived variables were positively correlated with field WD (R: 0.2 – 0.70). In addition, variability on WD in Syzgium sp. population is predominantly explained by the Resistograph amplitude, expressed as mean raw scale of adjusted DR (DR adj.RawSC) in a regression model. Given that intraspecific variation in WD is a crucial conjecture in forest AGB estimation, it is recommended to analyze with larger samples, and in-depth exploration on Resistograph-based variables is deemed to improve the accuracy of WD prediction models.

Digital mapping of soil organic carbon hotspots in nature conservation areas in the region of Flanders, Belgium

Climate-smart land management offers an opportunity to reconcile biodiversity goals with climate change mitigation. An important first step towards this, is an accurate estimate of the soil organic carbon (SOC) stock and the identification of locations with a large share in storage. General-purpose, regional assessments are not always useful to identify such hotspots, since they tend to even out extremes of SOC stocks in the landscape. The objectives of this study were to (i) develop and validate a digital soil mapping procedure to assess the SOC stock of nature conservation areas in Flanders, N. Belgium, and (ii) identify soil carbon hotspots in these areas. To this end, different soil inventories, encompassing 864 soil profiles, were combined in a digital soil mapping approach.The resulting boosted regression trees model was able to explain 75% of the variability in the training dataset (R2 = 0.75, RMSE = 9.83 kg m−2) and 59% after 10-fold cross-validation (R2 = 0.59, RMSE = 12.38 kg m−2). Reference Soil Group according to World Reference Base (WRB), vegetation type, highest groundwater level and clay fraction were identified as key predictors of the SOC stock in the upper 100 cm. The total SOC stock stored under 86,828 ha of nature conservation area was estimated to be 21,565 kton OC, or 24.8 kg m−2 on average. According to the presented hotspot definition, taking the SOC stock of the surrounding landscape into account, 10.5–11.3% of the nature conservation areas was classified as a hotspot. However, validation of the results in five study sites showed the local variability in predictive performance, with relatively accurate predictions in the two data-rich sites in the eastern part of the territory and poor predictions in the three remaining sites. Local, site-specific variations, together with uncertainties in the input data and unmapped elements can hamper the accurate estimation of SOC stocks and hotspots.

Uncertainties of soil organic carbon stock estimation caused by paleoclimate and human footprint on the Qinghai Plateau

BackgroundQuantifying the stock of soil organic carbon (SOC) and evaluating its potential impact factors is important to evaluating global climate change. Human disturbances and past climate are known to influence the rates of carbon fixation, soil physiochemical properties, soil microbial diversity and plant functional traits, which ultimately affect the current SOC storage. However, whether and how the paleoclimate and human disturbances affect the distribution of SOC storage on the high-altitude Tibetan Plateau remain largely unknown. Here, we took the Qinghai Plateau, the main component of the Tibetan Plateau, as our study region and applied three machine learning models (random forest, gradient boosting machine and support vector machine) to estimate the spatial and vertical distributions of the SOC stock and then evaluated the effects of the paleoclimate during the Last Glacial Maximum and the mid-Holocene periods as well as the human footprint on SOC stock at 0 to 200 cm depth by synthesizing 827 soil observations and 71 environmental factors.ResultsOur results indicate that the vegetation and modern climate are the determinant factors of SOC stocks, while paleoclimate (i.e., paleotemperature and paleoprecipitation) is more important than modern temperature, modern precipitation and the human footprint in shaping current SOC stock distributions. Specifically, the SOC stock was deeply underestimated in near natural ecosystems and overestimated in the strongly human disturbance ecosystems if the model did not consider the paleoclimate. Overall, the total SOC stock of the Qinghai Plateau was underestimated by 4.69%, 12.25% and 6.67% at depths of 0 to 100 cm, 100 to 200 cm and 0 to 200 cm, respectively. In addition, the human footprint had a weak influence on the distributions of the SOC stock. We finally estimated that the total and mean SOC stock at 200 cm depth by including the paleoclimate effects was 11.36 Pg C and 16.31 kg C m−2, respectively, and nearly 40% SOC was distributed in the top 30 cm.ConclusionThe paleoclimate is relatively important for the accurate modeling of current SOC stocks. Overall, our study provides a benchmark for predicting SOC stock patterns at depth and emphasizes that terrestrial carbon cycle models should incorporate information on how the paleoclimate has influenced SOC stocks.

Estimation of carbon stocks in harvested wood products of buildings in Japan: flux-data method and direct inventory method

In this study, carbon stocks in harvested wood products (HWPs) of buildings in Japan were estimated using the direct inventory method, which is highly accurate, and the flux-data method, which was proposed by the Intergovernmental Panel on Climate Change (IPCC) and is commonly used worldwide. We analyzed the differences between the estimated results and the respective reasons. The results indicate that the flux-data method greatly underestimated the carbon stocks in HWPs of buildings in Japan. In 2019, the values estimated by the flux-data method were only approximately 64% of those estimated by the direct inventory method. The half-lives of HWPs and the estimated continuous rate of change in industrial roundwood consumption proposed by the IPCC were likely the main causes of this difference. As for the decay function, the first-order decay, which is a default function proposed by the IPCC, was considered reliable for the estimations, because the decay function was not the main cause of the obtained difference.

Improving the multiple linear regression method of biomass estimation using plant water-based spectrum correction

ABSTRACT Accurate observation of plant biomass is crucial for estimation of global carbon stocks and ecosystem productivity, and optical spectroscopy technology represents a potential solution. However, the plant water content limits biomass estimation accuracy, due to its contribution to the plant spectral characteristics. In this study, we used a plant water-based spectrum correction (PSC) method to decrease the influence of the plant water content, and biomass models were developed by combining the PSC and multiple linear regression (MLR) methods. The spectral data of 387 tree leaves were acquired using an ASD FieldSpec 3 spectrometer, and their water content values were determined by the oven-drying method. Based on the plant water values of the leaves, the optimal model for the PSC-MLR method was determined and was treated as the ultimate PSC-MLR model (coefficient of determination R 2 of validation = 0.6959, root-mean-square error of validation = 0.0220 kg m−2, mean relative error of validation = 20.60%, ratio of performance to deviation of validation = 1.3492), which produced a better performance than the standard MLR model. This work shows the great potential of combining PSC with MLR for improved plant biomass estimation accuracy.