In this study, a novel approach to map forest attributes has been investigated for boreal forests in Sweden. The methodology relies on machine learning, utilizing a combination of remote sensing data and field data for both training and evaluating the proposed models. To ensure the accuracy in estimating forest attributes at any given time, the approach incorporates a broad range of available remote sensing data including airborne laser scanning (ALS) data, weekly satellite data from Sentinel-1 and Sentinel-2, and global forest map data. However, in this study focus has been on utilizing ALS data. The field data utilized in the study are derived from the Swedish National Forest Inventory and encompass measurements of key forest variables such as above-ground biomass, stem volume, basal area-weighted mean tree height, basal area-weighted mean diameter at breast height, and basal area. The potential of exporting knowledge gained from mapping Sweden to other forested landscapes such as in Latvia, using model updating with limited reference data from the new targeted area will be the next step to investigate. Here, data from Sweden were used to take the first steps towards developing a mapping methodology. The results demonstrate a promising potential of the proposed approach that will showcase new possibilities to share knowledge of updated forest mapping using the increasing flow of high-precision remote sensing data.

This paper describes time-series backscatter measurements using BorealScat-2, which is a multi-polarization, multi-frequency tomographic radar. It is installed on a 50-m tower at a boreal forest site in northern Sweden. We describe the system and show measurements related to forest water dynamics during summer conditions. The measurements include data at P-, UHF- and L-band. Results show periodic diurnal variations for all frequency bands but with different phases and amplitudes depending on frequency band and polarization. The results indicate the potential of estimating forest water dynamics from radar data.

This paper presents the first results of the ongoing development of new forest mapping methods for the Swedish national forest mapping case using Airborne Laser Scanning (ALS) data, utilizing the recent findings in machine learning (ML) and Artificial Intelligence (AI) techniques. We used Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) as ML models. In addition, Neural networks (NN) based approaches were utilized in this study. ALS derived features were used to estimate the stem volume (V), above-ground biomass (AGB), basal area (B), tree height (H), stem diameter (D), and forest stand age (A). XGBoost ML algorithm outperformed RF 1 % to 3 % in the R² metric. NN model performed similar to ML model, however it is superior in the estimation of V, AGB, and B parameters.

This paper presents the project ForestMap – a project aiming to develop and distribute new methods, which provide the benefits of accurate forest maps to a global audience. Using the recent developments in remote sensing, machine learning, and Artificial Intelligence (AI) the goal is to export the Scandinavian success stories to a wide range of stakeholders in the world.

Pairs of fully polarimetric ALOS-2 PALSAR-2 L-band SAR images were used to model biomass on backscatter change over seven growth seasons in a hemi-boreal forest. The biomass change was related to backscatter change via consecutive field surveys of 263 field plots with a 10 m radius. To correct for differences in backscatter not related to biomass abundance, a HV-VV polarization ratio based correction, previously used on airborne L-band data, was applied to the data. The uncertainty of obtained predictions (lowest model mean RMSE 65.1 t/ha, lowest model mean bias 7.1 t/ha) was almost identical whether model fitting and prediction used data from the same scene pair, or different scene pairs. This could possibly attest to the feasibility of the backscatter correction for PALSAR-2 data, but no large backscatter offsets were observed for uncorrected data, and significant variance in predictions, due to the inherent noise in the data and the comparatively small area of evaluation plots, inhibit the analysis.

This paper describes the development and initial evaluation of the BorealScat-2 radar. It is installed on a 50 m tall tower at a boreal forest site in northern Sweden and acquires tomographic and multi-polarization measurements at P- and L-band every 5 min. It is of the same design as its predecessor BorealScat except that the antenna frame can be moved along a 4 m horizontal track. A moveable antenna position is motivated by the need to acquire independent samples to reduce backscatter fluctuations. It also enables focused 3D imaging by combing tomographic and SAR processing when the backscatter fluctuations are temporally stable. We show results from data acquired in 2022, which demonstrates 3D imaging of a forest as well as the response from a trihedral.

Detecting forest insect damage before the visible discoloration (green attacks) using remote sensing data is challenging, but important for damage control. In recent years, the European spruce bark beetle (Ips typographus, L.) has damaged large amounts of forest in Europe. However, it is still debatable how early the infestations can be detected with remote sensing data. Some studies showed a spectral difference between healthy and green-attacked spruce trees at the plot level, while others showed that spectral differences existed before attacks. Therefore, a hypothesis is proposed that no spectral difference can be identified between green-attacked forests compared to healthy forests if the differences do not exist before the attacks. In this study, we tested this hypothesis using Sentinel-2 and WorldView-3 SWIR images on 24 healthy plots and 24 plots with mild, moderate, and severe attacks. In the results, the severely attacked plots did not show significant spectral differences in the Sentinel-2 images until August, and the sensitivity was found in the blue, red, red-edge, and SWIR band. Only the red band showed a significant difference between the healthy and moderately attacked plots in August, and only the blue, red, and SWIR band showed significant differences in September, October, and November. No significant differences were observed in the WorldView-3 images at the plot or individual tree level. We accepted the hypothesis that green attacks do not show spectral differences with the healthy forests when the differences do not exist before the attacks. We concluded that the SWIR bands were sensitive to attacks in the Sentinel-2 images with 10 m resolution, but not in the WorldView-3 images with 3.7 m resolution. Further studies are needed to explore the methodology of using WorldView-3 SWIR images for the early detection of forest infestation.

This study investigated the potential of utilizing time series of TanDEM-X phase heights, corrected for penetration depth, to detect silvicultural treatments in hemi-boreal forest. In total, 34 field plots with 40 m radius were used in conjunction with detailed forest management records to construct a reliable data set of treatments. The study area is situated in Remningstorp, a forest test site in southern Sweden. In the analysis, the temporal mean corrected phase heights were compared before and after a silvicultural treatment in order to quantify the effects of thinnings and clear-cuts on the phase height. As expected clear-cuts were highly distinguishable, but thinnings, while exhibiting a negative change in phase height on average, were not individually distinct from all untreated plots. Moreover, the results regarding the utility of applying penetration depth correction for the task were inconclusive. Overall, the results look very promising for using time series of phase height from TanDEM-X to map thinnings and clear-cuts, especially when several observations are available before and after the silvicultural treatment.

This study investigated the potential of using polarimetric synthetic aperture radar (PolSAR) data acquired at C- and X-band to estimate forest aboveground biomass (biomass) at a test site in southern Sweden. The SAR data were acquired with RADARSAT-2 and TerraSAR-X in September 2015, and the biomass estimations were cross-validated with 48 field inventoried plots of 40 m radius (0.5 ha), covering dominantly coniferous hemi-boreal forest. The quad-pol SAR data (HH, HV, VH, VV) were decomposed using the Yamaguchi model and the mean modelled scatter components for the plots were extracted and used as predictors in linear regression models to estimate the biomass. The model performances were quantified using the root mean square error (RMSE) and adjusted coefficient of determination, R2adj. The leave-one-out cross-validated RMSEs were 58.1 (37.1%) and 60.6 (38.6%) tons/ha for C-and X-band, respectively, and the R2adjwere 0.52 and 0.47, respectively. The regression model for C-band data used the volume and helix scattering components in the Yamaguchi decomposition as predictors, while the corresponding X-band model used the double bounce and surface components. The differences were likely due to the different penetrations at C-and X-band. We noticed a tendency to saturation at about 300 tons/ha in the X-band predictions, while no such tendency was noticed for the C-band model. We conclude that the Yamaguchi polarimetric decomposition technique is a useful approach when estimating biomass in a hemi-boreal forest. The strengths are not comparable to approaches where height information are available (e.g., single-pass interferometry from TanDEM-X), but this polarimetric approach based on single satellite images showed sensitivity to the entire biomass range (0 - 400 tons/ha) in our dataset. The potential of using quad-pol SAR data is, therefore, important to investigate further.

When modeling forest variables from InSAR phase height data, phase noise is a nuisance that can be mitigated by spatial averaging. In this empirical study, based on data from a hemi-boreal forest, the relation between field measured forest variables and TanDEM-X phase height and its dependence on the size of the field plots was investigated. Also, the impact of extending the region of phase height extraction beyond the actual plot size was investigated in order to determine the usefulness of this approach when small plot sizes cause significant noise in the estimation of phase height. For fixed field plot sizes (7 m and 10 m radii), the maximum correlation between Lorey's height and phase height occurred for phase height extraction regions multiple times larger in area than the field plots, which were located in homogenous stands.

SLU Forest Map are maps of the Swedish forest state, produced by the Swedish University of Agricultural Sciences (SLU) from satellite images using the Swedish National Forest Inventory sample plots as reference data. Until now, four maps have been produced, in raster format (12.5 × 12.5 m2to 25 × 25 m2cell sizes), with estimates of basal area-weighted mean tree height, basal area-weighted mean stem diameter, stand age, total as well as species-specific stem volume, for the years 2000, 2005, 2010, and 2015. These maps provide publicly available data, free of charge, supporting a wide range of applications; scientific research as well as operational uses in forest management planning, biodiversity assessment, and monitoring. This paper presents SLU Forest Map, the data and methods utilised in the production, and a new consistent evaluation of the estimation accuracy for each variable and mapped year.

In this study, Synthetic Aperture Radar (SAR) backscatter data from the Swedish airborne CARABAS-II and LORA systems were used to estimate stem density. The analysis were performed at the test site Remingstorp, located in the south of Sweden, consisting mainly of coniferous in hemi-boreal forests. In total, ten 80 m × 80 m forested areas, where all trees were measured in situ, with stem densities in the range of 278-552 stems and stem volumes in the range of 70-550 m3ha-1were analysed. SAR data from CARABAS-II and LORA were acquired in 2006 with nine unique flight headings. Local maxima of the backscatter were used to estimate stem density. The results were compared with in situ data and the accuracy in terms of root mean square error (RMSE) for CARABAS-II and LORA were found to be 81 stems (20.9%) and 82 stems (20.2%), respectively, in the best cases. The accuracy assessment of the stem density were performed after averaging images from each SAR system. This was done in order to suppress noise to a greater extent compared to using single images. The results show a potential to estimate stem density with low-freqency SAR data for the benefit of forest management.

In this study, stem volume measured by the Swedish National Forest Inventory were modelled using the k nearest neighbor (kNN) algorithm, with k=1, 3, or 5 neighbors. As independent variables, the combination of two satellite sensors were used: the active radar sensor TanDEM-X and the passive optical sensor Sentinel-2. The results indicate that stem volume per species can be predicted relatively accurately, mainly due to the inclusion of Sentinel-2 data, while the total stem volume is largely predicted well due to inclusion of the TanDEM-X phase height. The prediction of total stem volume was, however, not significantly improved with the additional spectral information from Sentinel-2 about the tree species. The kNN method is somewhat limited in the highest range of volumes, since no extrapolation is supported. Thus, it is important to have a reference dataset representing the entire range of the population for a successful application. The main advantage of combining the two data sources is the convenient procedure of obtaining both the tree species classification and volumes (divided per species) in a single method. It is concluded, that when sufficient reference data are available, the kNN approach with a combination of radar and optical data provides additional information about the stem volumes (in terms of tree species), but without improving the prediction of the total stem volume accuracy.

In this study, mapping of tree growth was performed using data from the two nation-wide acquisitions of airborne laser scanning in Sweden. Following the successful first national acquisition performed in 2009 - 2015, a new, repeated, scanning is now launched and ongoing. The first scanning provided new, accurate (in accuracy as well as in spatial resolution) data about the forest and quickly found wide-spread use in the forest industry. It outperformed previous methods and provided a new standard of data capture for forest management planning. The addition of a second scanning provide information also about changes, where forest tree growth is of high interest in the industry. This study presents the first results from large-scale assessment of growth for basal area-weighted mean tree height (H) and mean stem volume (V), using the bi-temporal scannings and sample-plot data from the National Forest Inventory. Growth was most accurately assessed by the direct change metrics of the scannings, although the accuracies were moderate. The accuracy of forecasts, i.e. only utilizing the predicted forest state at the first scanning, were similar for H but inferior for V, though.

This study applies the two-level model to predict stem volume (VOL), presented as wall-to-wall rasters. The SAR data were acquired with the TanDEM-X system and 518 scenes covered the entire Sweden. For comparison, a multiple linear regression model is also evaluated. Compared to earlier studies, the model parameters are fitted separately for each satellite scene. The prediction accuracy at the stand-level is evaluated using field inventoried reference stands within one scene, located in Northern Sweden and provided by a Swedish forest company. The results from the two models were similar, with an RMSE of 34.8 m3/ha and 32.9 m3/ha at the stand-level, respectively, and the corresponding biases were 14.3 m3/ha and 12.1 m3/ha. The error is significantly lower, compared to a previous study (52-65 m3/ha) where a universal multiple linear regression model was used for all scenes. It can be concluded, that using model parameters fitted at the local scene appears to improve the prediction performance in terms of RMSE, but no significant difference could be determined between predictions based on the two- level model or multiple linear regression, evaluated in this study.

In this study we present results of above-ground biomass (AGB) estimation for Sweden based on ALOS PALSAR backscatter data. The retrieval was performed within the framework of the BIOMASAR algorithm, and the validation was based on the Swedish National Forest Inventory (NFI) plots. The main focus, however, is about the modelling framework on the retrieval of AGB by integrating PALSAR backscatter, ICESAT GLAS metrics and allometric functions to improve the estimation of AGB. To evaluate the AGB retrieval, the average stem volume for each Swedish county from the NFI plots and the corresponding pixels covered by the inventory plots was computed. The result show an high agreement in terms of an RMSE of 8.2%, a bias of -1.9 m3/ha and a R2value of 0.82. The results indicate that stem volume derived from ALOS PALSAR data can be used to support statistics for Sweden and spatially explicit give information on stem volume distribution across the country, albeit local fluctuations due to the still simplified modelling framework here implemented.

The development of drone technology has been rapid in the last decade, providing highly competent and economical platforms for applied remote sensing. Mapping forest using drones is not an economical alternative for most operational applications in forestry and environmental monitoring, due to the limited area covered. However, the potential of drone-based remote sensing is expected to be very large in research applications. In this study, a standard, small, four-rotor drone is used as platform for a multispectral camera to accurately collect 3D-data about the forest canopy, data used to produce maps of forest variables. These maps proved to be very valuable input for development of new models for the exchange of green-house gases in boreal and hemi-boreal forests. The task was performed within the ICOS initiative at three sites in Sweden where each has a 150 m high tower monitoring green-house gas fluxes at many levels in the atmosphere.

Three-year forest above-ground biomass change were measured using L- and P-band Synthetic Aperture Radar (SAR) backscatter. The SAR data were collected in the airborne BioSAR 2007 and BioSAR 2010 campaigns over the hemiboreal Remningstorp test site in southern Sweden. Regression models for biomass were developed using reference biomass maps created using airborne laser scanning data and field measurements. The results from regression analysis show that using HV backscatter (or VH) in a model with above-ground biomass and backscatter change on either natural logarithmic or square root, and decibel scale, respectively, explained most of the variation in the biomass change, both for L- and P-band. In the case of L-band, the two best cases showed R2values of 66%, when comparing two SAR images acquired 2007 and 2010. For P-band using the same models, the best cases showed R2values of 62%. In summary, the results look promising using L- and P-band backscattering for mapping biomass change.

A C-band convoy mission with a receive-only interferometer using Sentinel-1 as the active sensor could cover the needs for an operational satellite mission for forest monitoring, that can clarify the magnitude and geography of the residual land sink of carbon in the global carbon budget; routinely map forest carbon changes in the tropics for REDD+, and routinely provide updated forest resource maps for operational forestry planning. We are about to carry out an airborne campaign in order to demonstrate the technical feasibility of this, and supplement this with analyses of Tandem-X.

A first evaluation of ALOS-2 PALSAR-2 data for forest stem volume estimation has been performed at a coniferous dominated test site in southern Sweden. Both the Fine Beam Dual (FBD) polarization and the Quad-polarimetric mode were investigated. Forest plots with stem volume reaching up to a maximum of about 620 m3ha−1(corresponding to 370 tons ha−1) were analyzed by relating backscatter intensity to field data using an exponential model derived from the Water Cloud Model. The estimation accuracy of stem volume at plot level (0.5 ha) was calculated in terms of Root Mean Square Error (RMSE). For the best case investigated an RMSE of 39.8% was obtained using one of the FBD HV-polarized images. The corresponding RMSE for the FBD HH-polarized images was 43.9%. In the Quad-polarimetric mode the lowest RMSE at HV- and HH-polarization was found to be 43.1% and 66.1%, respectively.

Interferometric TanDEM-X data from 2011 and 2014 were used to create biomass maps over the Swedish test site Remningstorp. These maps were used to compute the biomass change for four classes; pre-commercial thinning, thinning, clear-cutting, and untouched forest. Field inventory and ALS data from the corresponding years were used as reference data. The biomass change was compared on 12 subjectively chosen plots with 40 m radius for each class. It was found, that pre-commercial thinning was difficult to detect, as the biomass loss was less than the biomass growth during the four vegetation seasons investigated. Thinning could be detected from the biomass change being about zero or slightly negative, while clear-cut plots were obvious to notice, with the biomass withdrawal being several hundreds of tons ha-1. The untouched plots had a biomass growth of about 4 to 6 tons ha-1year-1. It was concluded, that annual TanDEM-X images can be used to detect also smaller silviculture activities such as thinning, but further research with shorter time periods would be desired.

In this paper, the potential of mono- and bistatic HH-polarized UHF-band SAR imagery for mapping of a hemi-boreal forest is studied. SAR data have been acquired with the two airborne SAR sensors LORA and SETHI during a joint FOI-ONERA campaign conducted in 2010. Three acquisition modes are compared: monostatic, quasi-monostatic (difference in elevation angle around 0°) and bistatic (difference in elevation angle around 6°). Images acquired at two perpendicular flight headings are used, i.e. 178° and 268°, to evaluate the influence of topography, which often has an aggravating effect on forest variable estimation. It is observed that for the quasi-monostatic and bistatic acquisitions, the influence of ground topography is lower compared to the monostatic acquisition. A linear regression model is used to explain the dependence of the backscattering coefficient on the logarithm of biomass, and it is observed that the estimated intercept and slope are similar for the two headings only in the case of quasi-monostatic and bistatic acquisitions, of which the latter features the lowest error and the highest coefficient of determination (0.73 and 0.70, respectively, for headings 178° and 268°).

Three-dimensional (3D) data about forest captured by airborne laser scanning (ALS) have revolutionized forest management planning. Accurate, updated large-scale maps of forest variables produced with low costs today support greatly improved decisions about silvicultural treatments compared to the past practice based on field surveyed data only. These maps usually lack important information about forest age and site productivity, as this cannot be accurately assessed from the available ALS data. In Sweden, ALS has recently been performed nation-wide, except the mountainous area, to produce a new and accurate digital terrain model (DTM). This DTM enables extremely cost-efficient extraction of 3D data about the forest from other sources than ALS, such as automatic stereo-matching of aerial images as well as from single-pass spaceborne interferometric synthetic aperture radar (InSAR). In contrast to ALS, these data sources can provide low-cost time-series of 3D data. Aerial images of Sweden are often available in archives back to approximately 1960, and the TanDEM-X SAR system has the potential to provide new data every second week over large areas. These data have a potentially high value for forest management planning, since they may provide missing and highly important information - forest site productivity, Site Index (SI) and forest age. This pilot study explores a least-squares minimization approach to estimate forest age and SI from time series of 3D data produced by 1) image matching of DMC aerial images, and 2) TanDEM-X SAR data.

The paper presents advances in regional mapping of forest growing stock volume (GSV, unit: m3/ha) using multitemporal ALOS PALSAR images. For this, an approach based on the Water Cloud Model has been used. Fine Beam Dual (FBD) and Wide Beam (WB) mode backscatter data acquired during 2010 were inverted to GSV; the individual GSV estimates were then combined to form an improved estimate. Examples of regional mapping for the boreal, temperate and tropical zone are here presented. Estimates were obtained at 25 m resolution using FBD data only and at 75 m combining multi-looked FBD estimates and WB estimates. The inclusion of multi-temporal WB data served to improve the spatial characterization of GSV. At pixel level, the discrepancy with respect to other estimates from Earth Observation data was large (relative RMSE ~ 90%). Aggregation to lower resolution (e.g., 1 km) increased the agreement and a relative RMSE mostly below 30% was obtained.

Methods to estimate forest biomass change have been investigated using experimental P-band synthetic aperture radar (SAR) data from the recent airborne campaigns BioSAR 2007 and BioSAR 2010 conducted over a hemiboreal test site in southern Sweden. Regression models based on backscatter change were developed using reference biomass change maps derived from high-density laser scanning data. Different regression models were developed for linear, square root, and logarithmic biomass change scales. The models were compared to the change maps based on laser data using twofold cross-validation, and estimation errors were evaluated using six 80 m by 80 m plots with detailed in situ measurements. The results indicate that the root-mean-square error of biomass change estimates based on P-band SAR backscatter data is about 15% or 20 t/ha. This suggests that not only clear-cuts but also growth and thinning can be measured. Simulations were performed in order to evaluate the possibility of using a spaceborne P-band SAR for measurements of forest biomass change. The simulations show that, with 64 equivalent number of looks (ENL) and a 50% change in biomass, it is possible to correctly indicate whether the forest has gained or lost biomass. Similarly, for a biomass loss of more than 75%, a correct indication of the sign of biomass change can be achieved with only 8 ENL.

Synthetic Aperture Radar (SAR) backscatter data from the Swedish airborne CARABAS-II and LORA systems were used to estimate stem volume at stand level. The study was performed in hemi-boreal forests at the Remningstorp test site, located in southern Sweden. In total, ten 80 m × 80 m stands, where all trees were measured in situ, with stem volumes in the range of 70-530 m3ha-1(on average 347 m3ha-1) were analyzed. SAR data from CARABAS-II and LORA were acquired from two different years, with nine unique flight headings that were repeated for each system and year. Regression analysis was used to estimate stem volume and the accuracy was assessed in terms of Root Mean Square Error (RMSE). As a first step, stem volume was estimated for each flight heading separately. The accuracy assessment was then performed by weighting the separate estimates for each system and year inversely proportionally to the variance about the regression function. The best results for CARABAS-II and LORA showed a relative RMSE of 7% and 24% of the mean stem volume, respectively. In a previous study, stem volume was estimated using LiDAR data and the same forest stands, resulting in an RMSE of about 12%. In conclusion, the estimation accuracy of stem volume using combined low-frequency CARABAS SAR images was found to be superior to that from using LiDAR data for the stands investigated.

Two field experiments have been conducted in Sweden to allow an evaluation of the backscatter signatures of wind-thrown forest from L-band, C-band and X-band Synthetic Aperture Radar. When the trees are felled the backscattered signal from TerraSAR-X (X-band) increase with about 1.5 dB, while for ALOS PALSAR (L-band) a decrease with the same amount is observed. Radar images with fine spatial resolution also show shadowing effects that should be possible to use for identification of storm felled forest.

In this study, dense hemi-boreal forest biomass and height estimation was investigated based on optical satellite data and a high quality Digital Elevation Model (DEM) from airborne laser scanning. This analysis was carried out on data collected 2008-2010 over the test site Remningstorp in southern Sweden. The optical sensors SPOT-5 HRS and ASTER were tested to process a Digital Surface Model (DSM), i.e. the vegetation height above mean sea level, that is used together with the DEM (derived from laser data) to calculate a Canopy Height Model (CHM) as the difference between the former ones. By modeling biomass and height using regression analysis on spectral data from SPOT-5 HRG and height metrics from the CHM an improved Root Mean Squared Error (RMSE) and adjusted R2is expected, compared to using the single data sources alone. The best results showed a relative RMSE for standwise prediction of mean biomass and height of 30.3% and 23.3%, respectively. Adding CHM data to a spectral based (HRG) prediction model improved the mapping accuracy roughly 3%. In conclusion, the estimation accuracy did not improve significantly by adding height metrics to spectral data.

Using data from the recent BioSAR 2007 and BioSAR 2010 campaigns, it has for the first time been possible to measure forest biomass change using P-band SAR data. Regression models based on backscatter change have been developed using reference data derived from high density laser scanning data. The models were evaluated for six areas with detailed in-situ measurements, for which the maximum biomass loss and growth was 30% and 20%, respectively. For the best model the coefficients of determination was 55-89%. This result suggests that not only clear cuts but also forest growth and thinning can be measured using P-band SAR backscatter.

The increasing availability of accurate Digital Elevation Models (DEMs) of nation-wide cover has opened new possibilities to produce accurate forest variable estimation using 3D data acquired from aerial imagery. Such data can be produced by automatic matching of stereo images and photogrammetric modeling of the forest canopy height. Using existing accurate DEM information, the forest canopy height above ground is then easily assessed. Today, Airborne Laser Scanning (ALS) is frequently used to capture data for accurate estimation of variables to be used in forest management planning. Recent studies in Scandinavia show estimation accuracies almost as accurate as ALS, using 3D data obtained from standard aerial imagery, at least for the most important forest variables. So far mainly area-based estimation methods at field plot or raster cell level have been studied. This paper reports early results from applying a single-tree modeling approach, corresponding to the Semi-ITC (Individual Tree Crown) method, commonly used in ALS-based applications, using 3D data acquired from aerial DMC imagery. Here, a simplified Semi-ITC method was used to estimate tree height at segment level. The Root Mean Square Error of estimating the maximum tree height was 34% (of the true mean maximum tree height). Clearly, the methodology used shows promising results and has potential to be used in forest management planning.

Biomass estimation performance using polarimetric interferometric synthetic aperture radar (PolInSAR) data is evaluated at L- and P-band frequencies over boreal forest. PolInSAR data are decomposed into ground and volume contributions, retrieving vertical forest structure and polarimetric layer characteristics. The sensitivity of biomass to the obtained parameters is analyzed, and a set of these parameters is used for biomass estimation, evaluating one parametric and two non-parametric methodologies: multiple linear regression, support vector machine, and random forest. The methodology is applied to airborne SAR data over the Krycklan Catchment, a boreal forest test site in northern Sweden. The average forest biomass is 94 tons/ha and goes up to 183 tons/ha at forest stand level (317 tons/ha at plot level). The results indicate that the intensity at HH-VV is more sensitive to biomass than any other polarization at L-band. At P-band, polarimetric scattering mechanism type indicators are the most correlated with biomass. The combination of polarimetric indicators and estimated structure information, which consists of forest height and ground-volume ratio, improved the root mean square error (rmse) of biomass estimation by 17%-25% at L-band and 5%-27% at P-band, depending on the used parameter set. Together with additional ground and volume polarimetric characteristics, the rmse was improved up to 27% at L-band and 43% at P-band. The cross-validated biomass rmse was reduced to 20 tons/ha in the best case. Non-parametric estimation methods did not improve the cross-validated rmse of biomass estimation, but could provide a more realistic distribution of biomass values.

Biomass estimation performance from model-based polarimetric SAR interferometry (PolInSAR) using generic parametric and non-parametric regression methods is evaluated at Land P-band frequencies over boreal forest. PolInSAR data is decomposed into ground and volume contributions, estimating vertical forest structure, and using a set of obtained parameters for biomass regression. The considered estimation methods include multiple linear regression, support vector machine and random forest. The biomass estimation performance is evaluated on DLR's airborne SAR data at L- and P-bands over Krycklan Catchment, a boreal forest test site in Northern Sweden. The combination of polarimetric indicators and estimated structure information has improved the root mean square error (RMSE) of biomass estimation up to 28% at L-band and up to 46% at P-band. The cross-validated biomass RMSE was reduced to 20 tons/ha.

The ESA funded campaign BioSAR 2010 was carried out at the forestry test site Remningstorp in southern Sweden, in support to the BIOMASS satellite mission under study. Fully polarimetric SAR data were successfully acquired at Land P-band using ONERA's multi-frequency system SETHI. In addition with other data types gathered, e.g. LiDAR and in-situ measurements, the compiled data set will be used for analyses and comparisons with biomass estimation results obtained at the same test site in the campaign BioSAR 2007, in which DLR's E-SAR made the SAR imaging. Detection of forest changes, robustness of biomass retrieval algorithms and long-term P-band coherence will be in focus as well as cross-validations between the two SAR sensors.

The study focuses on investigation and evaluation of wind-thrown forest mapping using satellite remotely sensed data from three synthetic aperture radar (SAR) sensors. The study is carried out at Remningstorp, a test site in the south of Sweden dominated by coniferous forest, where trees were manual felled to simulate wind-thrown forest. The satellite data consisted of time series of HH polarized SAR images acquired by the Advanced Land Observing Satellite (ALOS) Phased Array type L-band Synthetic Aperture Radar (PALSAR), Radarsat-2 (C-band) and TerraSAR-X (X-band). The results from visual interpretation of SAR images acquired before and after the simulated wind-throw together with corresponding ratio images show that ALOS PALSAR HH polarized intensity images are not able to detect wind-thrown forest, probably due to too coarse spatial resolution. In contrast, the wind-thrown forest is clearly visible in the Radarsat-2 and TerraSAR-X HH polarized images, implying that it may be possible to develop a new application using these SAR data for mapping of wind-thrown forests.

In this study, we assess the efficiency of using L-band satellite SAR for forest change monitoring, by applying a combination of change detection techniques to SAR backscatter intensity images acquired over Swedish forest. We use a bi-temporal change detection approach based on image rationing. Histogram based techniques are used both for radiometric normalization and thresholding. For a final classification step, we evaluate a data fusion based change detection method that exploits the spatial and spectral information from one or multiple SAR channels. Pre-applied filters are also evaluated as alternatives to tackle low resolution and speckle. HH and HV polarized Fine Beam Dual images, acquired 34 degrees off nadir (FBD34) by the Advanced Land Observing Satellite Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR), are used to find clear-cuts in Swedish boreal forest. Our results show that clear-cuts can be clearly extracted from L-band SAR data.

The nation-wide Airborne Laser Scanning (ALS) currently performed by the Swedish National Land Survey will provide a new and accurate Digital Elevation Model (DEM). These data will enable new and cost-efficient assessments of vegetation height using Canopy Height Models (CHMs) derived as the difference between a Digital Surface Model (DSM) and the DEM. In this context, the High Resolution Stereoscopic (HRS) sensor onboard SPOT-5 and the airborne Z/I Digital Mapping Camera (DMC) used for operational aerial photography by the Swedish National Land Survey are of main interest. Previous research has shown that reliable tree height data are a powerful source of information for forest management planning. This study investigated the possibilities to map forest variables using CHMs derived from either the SPOT-5 HRS or Z/I DMC sensor together with ALS DEM data, in combination with spectral data from the SPOT-5 High Resolution Geometric (HRG) sensor. The results when using the Z/I DMC CHM in combination with SPOT-5 HRG data showed Root Mean Square Errors for standwise prediction of mean tree height, stem diameter, and stem volume of 7.3%, 9.0%, and 19%, respectively. The SPOT-5 HRS CHM in combination with SPOT-5 HRG data improved the SPOT HRG based estimates from 13% to 10%, 15% to 13%, and 31% to 23%, for tree height, stem diameter, and stem volume, respectively. Adding CHM data to a SPOT-5 HRG based prediction model improved the mapping accuracy between 13% to 44%. In conclusion, the obtained accuracies may be sufficient for operational forest management planning.

This study presents results for observing forest changes m Sweden using multi-temporal L-band satellite data and is a part of the JAXA's ALOS Kyoto and Carbon Initiative. An extensive dataset of images acquired by the Advanced Land Observing Satellite Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR) is investigated for clear-cut detection m boreal forests m northern Sweden (Lat. 64°14' N, Long. 19°50' E). Strong forest/non-forest contrast and temporal consistency were found for the Fine Beam Dual HV-polanzed backscatter during unfrozen conditions. Thus, a simple thresholding algorithm that exploits the temporal consistency of pair-wise HV-backscatter measurements has been developed for clear-cut detection. When applied to an image pair acquired during favorable weather conditions, the detection algorithm identified 76% of the clear-cut pixels within a reference layer, with zero erroneously detected pixels. With further refinement the developed methodology can be an option to present operational alternatives for clear-cut detection.

With the continued threat of global warming, the need to obtain consistent and accurate measurements of the carbon stored in forests is strong. L- and P-band SAR backscatter data have shown to be sensitive to forest biomass, which in turn is coupled to the stored carbon. In this paper a biomass retrieval method is developed for L- and P-band using data from the BioSAR campaign conducted in Sweden during the spring 2007 over hemi-boreal forest. The results show that the use of L-band data gives an underestimation of biomass for stands with high biomass; while for P-band no such underestimation is seen. RMSEs are found to be 30-40% of the mean biomass for L-band and about 25% for P-band for stands with biomass ranging from 10 to 290 tons/ha.

The Phased Array type L-band Synthetic Aperture Radar onboard the Advanced Land Observing Satellite has, since its launch, been acquiring an extensive data set of images over two forest test sites in Sweden. The sites of Remningstorp (Lat. 58deg30' N, Long. 13deg40' E) in the south and Krycklan (Lat. 64deg14' N, Long. 19deg50' E) in the north of Sweden are characterized by hemiboreal and boreal forests, respectively. In this paper, we have investigated the signatures of standwise backscatter measurements from forests with different growth stages in relation to polarization, environmental conditions, image viewing geometry, and spatial resolution. The HV backscatter presented stronger sensitivity to the forest growth stage than the HH and VV backscatter. Under unfrozen conditions, the dynamic range of fine-beam data acquired at 34.3deg was 8-9 dB for the HV polarization and 6-7 dB for the HH polarization. At 21.5deg, in the polarimetric mode, the dynamic range was 6, 7, and 9 dB at VV, HH, and HV polarizations, respectively. Regardless of the specific polarization, the backscatter was temporally consistent under unfrozen conditions, with a small increase of backscatter in regrowing young forest for wet conditions. Under thawing and frozen conditions, repeated measurements were available only for the HH backscatter at 34.3deg. For thawing conditions, the backscatter level was similar to the unfrozen conditions even though the signatures differed depending on temperature dynamics, snow-cover properties, and precipitation. Under frozen conditions, the signatures varied depending on temperature. For images acquired when the temperature was well below the freezing point, the backscatter was low, and the dynamic range was small (2-4 dB); nonetheless, the measurements were consistent. Images acquired when temperature was close to the freezing point presented a behavior similar to unfrozen conditions. The sensitivity of the backscatter to the image viewing geometry for different growth stages was studied for data acquired under dry unfrozen conditions. The backscatter difference increased for increasing look angle because of the increase in volume scattering and the decrease of ground-surface backscatter. The largest difference was observed at 41.5deg with 2.5-4-dB difference for the HH and 4-5-dB difference for the HV case. Loss of spatial resolution (20-50 m) did not have any effect on the backscatter signatures in Krycklan, whereas in Remningstorp, the smallest stands were affected.

Interferometric coherence from nine image pairs acquired by the Japanese satellite ALOS over a forest area in northern Sweden have been analyzed. Coherence levels for sparse and dense forest are compared. The results are in line with previous studies for JERS-1 and confirm that in most cases the temporal decorrelation is a severe problem for L-band SAR with repeat cycles on the order of six weeks. However, frozen conditions can give relatively high coherence for sparse forest and thereby allow good separation between dense and sparse forest.

In 2006 calibration activities for ALOS PALSAR were conducted in Sweden. Four five-metre trihedral corner reflectors and three smaller dihedral reflectors were deployed and operated during eight months. 23 PALSAR scenes were acquired over the calibration site allowing an evaluation of the quality and temporal stability of the data. Results show that the co-polarized data have been stable during the whole calibration period with variations in the trihedral responses lower than 0.7 dB. The measured resolution in azimuth was 4.4 m and in slant range 4.7 m for single polarization images and 9.5 m for polarimetric data. For the cross-polarized data large variations in the dihedral responses were found. It is assumed that this is caused by a larger sensitivity to pointing errors. For the polarimetric data, estimation of Faraday rotation gave values ranging from 0.1deg to 3deg.

SAR images from the Swedish airborne CARABAS-II and LORA systems have been visually analyzed over simulated wind-thrown forest at both single tree and stand level. In ideal conditions, the results show that LORA is more accurate than CARABAS-II at detecting wind-thrown trees, regardless of tree size and direction of the fallen trees relative to flight heading. Furthermore, the visible single trees in the LORA images appeared more distinct than in the CARABAS-II images, which could be explained by the high resolution in the LORA images. Based on visual interpretation, it is likely that the detection of wind-thrown forests could be improved using VHF/UHF SAR images acquired both prior to and after a storm event.

A controlled experiment has been performed to quantify the ability to detect clear-cuts using ALOS PALSAR data. The experiment consisted of 8 old spruce dominated stands, each with a size of about 1.5 ha, located at a test site in southern Sweden. Four of the stands were clear-felled and the remaining stands were left untreated for reference. A time series of PALSAR images was acquired prior to, during, and after treatment, including 7 fine beam single polarization (FBS, look angle 34.3deg, HH-polarization) SAR images. The results clearly show that the clear-felled stands could be separated from the reference stands. The drop in backscattering coefficient between the reference and the clear-felled stands was on average 2.1 dB. This implies that ALOS PALSAR data potentially can be used for large-scale mapping of changes in forest cover.

A first evaluation of ALOS PALSAR data for forest stem volume estimation has been performed at a coniferous dominated test site in Sweden. In total, 7 Fine Beam Single polarization (FBS, look angle 34.3°, HH-polarization) and 7 Polarimetric (PLR, look angle 21.5°, HH-, HV-, VH-, and VVpolarization) SAR images were used. In total, 56 forest stands with stem volume in the range of 45-650 m3 ha-1 (average 325 m3 ha‒1) were analyzed by relating backscatter intensity to field data. The estimation accuracy of stem volume at stand level was calculated in terms of root mean square error (RMSE). For the best case investigated an RMSE of 30% was obtained using one of the FBS images acquired in the winter season. The corresponding RMSEs for the PLR images with HH-, HV-, VH-, and VV-polarization were 65%, 65%, 62%, and 81%, respectively. The better results for FBS compared to PLR mode could be explained by particularly favorable weather conditions at image acquisition, or by the higher ground resolution in the FBS images, which makes small stands less sensitive to errors in geo- coding.

This paper describes the Swedish activities to support calibration and validation of the Japanese satellite ALOS. Data over three test sites from the instruments PALSAR and AVNIR-2 will be provided by JAXA for evaluation purposes. The main activity during 2006 is a PALSAR calibration experiment using four 5-m radar reflectors, which will be deployed at the test site in southern Sweden. The ALOS data will also be used for developing forest applications, e.g. detection of clear-cuts and storm damages.

Space- and airborne SAR have a potential for providing timely information of forest storm damage. In this paper, we compare Envisat, Radarsat and CARABAS with aerial photography and field observations after a severe storm event. The spaceborne C-band images were not able to detect the storm- damaged areas due to unfavorable frequency band and coarse resolution. The airborne CARABAS VHF SAR, on the other hand, detects most storm-damaged forest areas and even partly damaged which are often not detected in the aerial photography.

In this paper, the accuracy of forest stem volume estimation using a combination of optical SPOT-5 satellite and TopEye laser scanner data is investigated, at stand level. It is anticipated that the accuracy will be improved for the combined stem volume estimate compared to using SPOT-5 data only. The test site is located in the south of Sweden and consists mainly of coniferous forest. The stem volume for the selected stands was in the range of 30-620 m/sup 3/ ha/sup -1/ with an average stem volume of 288 m/sup 3/ ha/sup -1/ and an average size of 2.9 ha. Regression analysis has been used to develop stem volume functions for each sensor and for the combination. In the combined stem volume function the horizontal forest structure is captured by the optical satellite data whereas the vertical structure is represented by the laser derived tree height data. The accuracy in terms of relative root mean square error was 30.8% of the average stem volume for SPOT-5 and 15.7% for the combination. Thus, compared to using only SPOT-5 data the improvement was found to be 49% The result implies that the combination of multi-spectral optical satellite and laser derived tree height data can be used for standwise stem volume estimation in forestry applications.

The accuracy of forest stem volume estimation at stand level using a combination of airborne synthetic aperture radar (SAR) and optical satellite data is investigated in this paper. It is anticipated that the accuracy will be improved for the combined stem volume estimate compared to using single sensor data. The test site is located in the south of Sweden and consists mainly of coniferous forest. The stem volume for the selected stands was in the range of 15-585 m/sup 3/ ha/sup -1/ with an average stem volume of 266 m/sup 3/ ha/sup -1/ and an average size of 3.5 ha. Remotely sensed data have been collected with the airborne CARABAS-II VHF SAR sensor and the multi-spectral optical Landsat TM satellite sensor. Regression analysis has been used to develop stem volume functions for each sensor and for the combination. The accuracy in terms of root mean square error was 49 m/sup 3/ ha/sup -1/ (corresponding to a relative error of 18.5% of the average stem volume) for CARABAS-II, 66 m/sup 3/ ha/sup -1/ (24.8%) for Landsat TM, and 38 m/sup 3/ ha/sup -1/ (14.3%) for the combination. Thus, compared to using only CARABAS-II data the improvement was found to be 23% and using only Landsat TM data 42%. For high stem volumes CARABAS-II gave the best result, while for lower stem volumes Landsat TM was more accurate. Hence, the combination of the two techniques provided significantly better results over the full range of stem volumes. The result implies that the combination of low-frequency SAR data and multi-spectral optical satellite data can be used for standwise stem volume estimation in forestry applications.

Retrieval of forest stem volume has been the main civilian application investigated for VHF SAR data during recent years. Based on promising results obtained from previous studies, a project was defined with the objective to develop a fully integrated system for retrieval of stem volume at high spatial resolution from CARABAS-II VHF SAR images. Using the developed production line, stem volume can easily be retrieved within predefined forest stands or within homogeneous areas defined by segmentation of optical satellite images. In order to evaluate the accuracy and efficiency of the production line, CARABAS-II images have been collected and analyzed over a test area covering 25/spl times/25 km/sup 2/. The major part of the area is owned by the Swedish forest company Holmen Skog AB. For coniferous forest stands, exhibiting normal forest production conditions, the accuracy in terms of root mean square error (RMSE) for estimation of forest stem volume is about 44 m/sup 3/ ha/sup -1/ (22% of the average stem volume) in the range of 70-400 m/sup 3/ ha/sup -1/, using the developed production line. In comparison to earlier research tools, the developed production line speeds up the analyses approximately 10 times.

Multitemporal European Remote Sensing satellites 1 and 2 (ERS-1/2) and the Japanese Earth Resources Satellite 1 (JERS-1) interferometric synthetic aperture radar (InSAR) data from a boreal forest test site in Sweden (stem volumes up to 335 m/sup 3//ha, equivalent to above-ground dry biomass of /spl sim/200 tons/ha) are studied in order to estimate stem volume using coherence and backscatter. The changes of JERS-1 backscatter and ERS-1/2 tandem coherence between images are consistent over the area studied, in contrast to ERS-1/2 backscatter. A model-based regression analysis has been performed, and the use of the model for inversion is discussed and compared with other approaches found in the literature. The model parameters are discussed in terms of their relation to wind speed and temperature. Results from the different acquisitions are combined to improve the stem volume estimation. The accuracy in terms of rms error (RMSE) for standwise estimated stem volume is /spl ap/10 m/sup 3//ha using ERS-1/2 coherence. Using backscatter and coherence from JERS-1 we obtain an RMSE of /spl ap/30-35 m/sup 3//ha. Finally, conditions for accurate retrieval of stem volume using multitemporal InSAR observations are discussed. We conclude that C- and L-band repeat-pass InSAR can provide stem volume estimates in boreal forests with accuracies similar to those of standard in situ measurements.

Presents a comparison of remote sensing of forests using two complementary, high-resolution sensors; namely the TopEye lidar and CARABAS VHF SAR systems. The lidar data are from a small footprint configuration, allowing discrimination between pulses reflected from the tree crowns, and those penetrating through gaps in the canopy. From these data, measurements of tree height and crown size can be made for individual trees, and the tree volume calculated through empirical relationships. The CARABAS VHF SAR provides a more direct measurement of tree volume, since the long wavelengths penetrate the forest canopy and are scattered by the trunk-ground dihedral. Results of stem volume retrieval for individual trees using data from the two systems are presented, and the possibility of using the complementary measurements discussed.

Thinning cuttings constitute a considerable part of the annually harvested forest stem volume in Scandinavia. The interest to detect thinning cuttings is in the areas of global biomass monitoring in accordance with international treaties, annual local scale measurements of thinnings, forestry law enforcement, and commercial planning of the forest resource. This paper presents results from a study using the airborne CARABAS-II VHF SAR system for detection of thinning cuttings in coniferous forest stands. SAR data were acquired prior to and after thinning cuttings over 25 stands located at a test site in the southwest of Sweden. For the analysis another 24 stands were selected for reference measurements. The results from the statistical analysis show a significant difference between the reference stands and the thinning cuttings for one out of three flight paths examined. A decrease in backscatter due to thinning was found for all image pairs analyzed. Factors influencing the results such as variations in radio frequency environment, image noise variations, and removal of image artifacts have to be further investigated. Nevertheless, the proposed change detection method indicates a possibility to map thinning cuttings using CARABAS-II VHF SAR data.

Strong winds cause severe damage worldwide to forested land every year. The devastating storms that struck large parts of Europe in late 1999 destroyed the equivalent of several years of normal forest harvesting, amounting to very large economical sums. Therefore, rapid mapping of damaged areas is of major importance for assessment of short-term actions as well as for long-term reforestation purposes. In this paper, the use of airborne CARABAS-II very high frequency (VHF) (20-90 MHz) synthetic aperture radar (SAR) imagery for high spatial resolution mapping of wind-thrown forests has been investigated and evaluated. The investigation was performed at a test site located in southern Sweden and dominated by Norway spruce forests. A regression model estimating radar backscattering amplitude prior to the storm was developed. The estimated amplitudes were compared to measured amplitudes after the storm. The results clearly show that the backscattering amplitude, at a given stem volume, is considerably higher for wind-thrown forests than for unaffected forests. Furthermore, the backscattering from fully harvested storm-damaged areas was, as expected, significantly lower than from unaffected stands. These findings imply that VHF SAR imagery has potential for mapping wind-thrown forests. However, to prevent ambiguities in increased backscattering caused by normal stem volume growth or wind-fellings, multitemporal change detection techniques using VHF SAR images acquired prior to and after wind-fellings would be preferable.

The use of airborne CARABASII VHF (20-90 MHz) SAR data for retrieval of forest parameters has been investigated. The investigation was performed at a test site located in the southwest of Sweden consisting mainly of Norway spruce forests. Regression models predicting forest parameters from radar backscattering amplitude were developed and evaluated. The results showed a linear relationship between backscattering amplitude and forest stem volume, stem diameter, and tree height. The analysis also showed that the radar signal is strongly affected by ground slope conditions. The root mean square errors from the regression analysis, restricted to forest stands on near-horizontal ground, were found to be 66 m/sup 3/ ha/sup -1/, 3.2 cm, and 2.3 m for stem volume, stem diameter, and tree height respectively. No saturation of the backscattered signal was observed up to the maximum stem volume of 625 m/sup 3/ ha/sup -1/, corresponding to a biomass of 375 tons ha/sup -1/. The results imply that VHF SAR data have significant potential for operational use in forestry.

The ability to retrieve forest stem volume using CARABAS (coherent all radio band sensing) SAR images (28-60 MHz) has been investigated. The test site is a deciduous mixed forest on the island of Oland in southern Sweden. The images have been radiometrically calibrated using an array of horizontal dipoles. The images exhibit a clear discrimination between the forest and open fields. The results show that the dynamic range of the backscattering coefficient among the forest stands is higher than what has been found with conventional SAR using microwave frequencies. The backscatter increases with increasing radar frequency. This work shows an advantage compared to higher frequencies for stem volume estimation in dense forests.