Uma versão mais recente desse conjunto de dados com dados de 2015 a 2021 pode ser encontrada em JAXA/ALOS/PALSAR/YEARLY/SAR_EPOCH. O mosaico global de 25 m do PALSAR/PALSAR-2 é uma imagem global de SAR contínua criada por faixas de mosaico de imagens de SAR do PALSAR/PALSAR-2. Para cada ano e local, os dados de faixa foram selecionados por inspeção visual do …

Global Satellite Mapping of Precipitation (GSMaP) provides a global hourly rain rate with a 0.1 x 0.1 degree resolution. GSMaP is a product of the Global Precipitation Measurement (GPM) mission, which provides global precipitation observations at three hour intervals. Values are estimated using multi-band passive microwave and infrared radiometers from the GPM Core Observatory satellite and with the assistance of a constellation of other satellites. GPM's precipitation rate retrieval algorithm is based on a radiative transfer model. The gauge-adjusted rate is calculated based on the optimization of the 24h accumulation of GSMaP hourly rain rate to daily precipitation by NOAA/CPC gauge measurement. This dataset is processed by GSMaP algorithm version 6 (product version 3). See GSMaP Technical Documentation for more details. This dataset contains provisional products GSMaP_NRT that are regularly replaced with updated versions when the GSMaP_MVK data become available. The products are marked with a metadata property called ''status''. When a product is initially made available, the property value is ''provisional''. Once a provisional product has been updated with the final version, this value is updated to ''permanent''. For more information please refer General Documentation

This dataset is contains orthorectified imagery from the Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-2) sensor on-board the Advanced Land Observing Satellite (ALOS) "DAICHI". The AVNIR-2 ORI product was created from AVNIR-2 1B1 data after stereo matching with reference to ALOS's Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM)-derived …

The global forest/non-forest map (FNF) is generated by classifying the SAR image (backscattering coefficient) in the global 25m resolution PALSAR-2/PALSAR SAR mosaic so that strong and low backscatter pixels are assigned as "forest" and "non-forest", respectively. Here, "forest" is defined as the natural forest with the area larger than 0.5 …

The SEN12TP dataset (Sentinel-1 and -2 imagery, timely paired) contains 2319 scenes of Sentinel-1 radar and Sentinel-2 optical imagery together with elevation and land cover information of 1236 distinct ROIs taken between 28 March 2017 and 31 December 2020. Each scene has a size of 20km x 20km at 10m pixel spacing. The time difference between optical and radar images is at most 12h, but for almost all scenes it is around 6h since the orbits of Sentinel-1 and -2 are shifted like that. Next to the \(\sigma^\circ\) radar backscatter also the radiometric terrain corrected \(\gamma^\circ\) radar backscatter is calculated and included. \(\gamma^\circ\) values are calculated using the volumetric model presented by Vollrath et. al 2020.

The uncompressed dataset has a size of 222 GB and is split spatially into a train (~90%) and a test set (~10%). For easier download the train set is split into four separate zip archives.

Please cite the following paper when using the dataset, in which the design and creation is detailed:
T. Roßberg and M. Schmitt. A globally applicable method for NDVI estimation from Sentinel-1 SAR backscatter using a deep neural network and the SEN12TP dataset. PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science, 2023. https://doi.org/10.1007/s41064-023-00238-y.

The file sen12tp-metadata.json includes metadata of the selected scenes. It includes for each scene the geometry, an ID for the ROI and the scene, the climate and land cover information used when sampling the central point, the timestamps (in ms) when the Sentinel-1 and -2 image was taken, the month of the year, and the EPSG code of the local UTM Grid (e.g. EPSG:32643 - WGS 84 / UTM zone 43N).

Naming scheme: The images are contained in directories called {roi_id}_{scene_id}, as for some unique regions image pairs of multiple dates are included. In each directory are six files for the different modalities with the naming {scene_id}_{modality}.tif. Multiple modalities are included: radar backscatter and multispectral optical images, the elevation as DSM (digital surface model) and different land cover maps.

Data modalities

name	Modality	GEE collection
s1	Sentinel-1 radar backscatter	COPERNICUS/S1_GRD
s2	Sentinel-2 Level-2A (Bottom of atmosphere, BOA) multispectral optical data with added cloud probability band	COPERNICUS/S2_SR COPERNICUS/S2_CLOUD_PROBABILITY
dsm	30m digital surface model	JAXA/ALOS/AW3D30/V3_2
worldcover	land cover, 10m resolution	ESA/WorldCover/v100

The following bands are included in the tif files, for an further explanation see the documentation on GEE. All bands are resampled to 10m resolution and reprojected to the coordinate reference system of the Sentinel-2 image.

Modality Bands

Modality	Band count	Band names in tif file	Notes
s1	5	VV_sigma0, VH_sigma0, VV_gamma0flat, VH_gamma0flat, incAngle	VV/VH_sigma0 are the \(\sigma^\circ\) values, VV/VH_gamma0flat are the radiometric terrain corrected \(\gamma^\circ\) backscatter values incAngle is the incident angle
s2	13	B1, B2, B3, B4, B5, B7, B7, B8, B8A, B9, B11, B12, cloud_probability	multispectral optical bands and the probability that a pixel is cloudy, calculated with the sentinel2-cloud-detector library optical reflectances are bottom of atmosphere (BOA) reflectances calculated using sen2cor
dsm	1	DSM	Height above sea level. Signed 16 bits. Elevation (in meter) converted from the ellipsoidal height based on ITRF97 and GRS80, using EGM96†1 geoid model.
worldcover	1	Map	Landcover class

Checking the file integrity
After downloading and decompression the file integrity can be checked using the provided file of md5 checksum.
Under Linux: md5sum --check --quiet md5sums.txt

References:

Vollrath, Andreas, Adugna Mullissa, Johannes Reiche (2020). "Angular-Based Radiometric Slope Correction for Sentinel-1 on Google Earth Engine". In: Remote Sensing 12.1, Art no. 1867. https://doi.org/10.3390/rs12111867.

Global Precipitation Measurement (GPM) is an international satellite mission to provide next-generation observations of rain and snow worldwide every three hours. The Integrated Multi-satellitE Retrievals for GPM (IMERG) is the unified algorithm that provides rainfall estimates combining data from all passive-microwave instruments in the GPM Constellation. This algorithm is intended …

This product is the sum of the one-sided green leaf area per unit ground area. For data after 2021-11-28, see the V3 dataset. GCOM-C conducts long-term and continuous global observation and data collection to elucidate the mechanism behind fluctuations in radiation budget and carbon cycle needed to make accurate projections regarding future temperature rise. At the same time, cooperating with research institutions having a climate numerical model, it contributes to reduction of errors in temperature rise prediction derived from the climate numerical model and improvement of accuracy of prediction of various environmental changes. SGLI mounted on GCOM-C is the succession sensor of the Global Imager (GLI) mounted on ADEOS-II (MIDORI II) and is the Imaging Radiometer which measures the radiation from near-ultraviolet to thermal infrared region (380 nm-12 um) in 19 channels. Global observation of once for approximately every two days is possible at mid-latitude near Japan by observation width at ground greater than 1,000 km. In addition, SGLI realizes high resolution than the similar global sensor and has a polarized observation function and a multi-angle observation function.

This data publication contains an optimized mosaic of PALSAR-2 L-band dual-polarized radar backscatter summer composite for the year 2020 across Canada (excluding the Arctic Archipelago). Its primary purpose is to offer the best possible L-band radar summer-like composite mosaic mostly tailored for i) classifying natural treed or shrubby vegetation covers, and ii) estimating their structural attributes, such as height and biomass. ## Methodology: This product is based on the freely available and open dataset of yearly JAXA Global PALSAR-2/PALSAR Mosaics ver. 1 (hereafter JAXA GPM v1). They were generated by the Japanese space agency (JAXA) using PALSAR L-band synthetic aperture radar sensors aboard the Advanced Land Observing Satellites (ALOS): ALOS-2 PALSAR-2 (2015 to 2020) and ALOS PALSAR (2007 to 2010). JAXA GPM v1 provide yearly mosaics orthorectified and slope-corrected L-band HH- and HV-polarized gamma naught (γ°) backscatter amplitude with 25-m pixel size and scaled as 16-bit data (Shimada et al. 2014). JAXA GPM v1 are accessible as a Google Earth Engine image collection at https://developers.google.com/earth-engine/datasets/catalog/JAXA_ALOS_PALSAR_YEARLY_SAR. The yearly 2007 to 2020 JAXA GPM v1 dataset across Canada underwent a post-processing and compositing methodology implemented in Google Earth Engine, as detailed in Pontone et al. 2024 and summarized in a pdf “Readme” file provided with the dataset. In summary, the method involves these three steps: Post-processing of yearly γ° HH and HV datasets: handling data gaps, filtering speckle noise, and generating two radar vegetation indices, the HV/HH ratio (HVHH) and the radar forest degradation index (RFDI). Temporal compositing from 2015 to 2020 of post-processed yearly γ° PALSAR-2 HH, HV, HVHH, and RFDI backscatter data aimed to i) address data gaps and ii) mitigate detrimental backscatter fluctuations across ALOS-2 orbits resulting from numerous out-of-summer acquisitions. Generating the final PALSAR-2 L-band γ° radar backscatter summer composite circa 2020 raster files. ## Performance et limitations: The resulting Canada-wide, excluding the Arctic Archipelago, gap-free and radiometrically optimized mosaic of circa 2020 PALSAR-2 L-band backscatter summer composite was found significantly improved compared to the single-year 2020 JAXA GPM v1 mosaic, particularly in northern boreal Canada (Pontone et al. 2024). However, this product should be considered as a summer-like composite and users should be mindful of the following known limitations: • In northwestern Canada, there were often minimal to no summer PALSAR-2 acquisitions, resulting in residual backscatter fluctuations across ALOS-2 orbits. • The composite may exhibit patchy radiometric noise in areas that experienced major disturbances (fires, harvesting) between 2015 and 2020 despite they were accounted for in our compositing methodology. • This product is deemed less performant, or possibly not suitable, for i) characterizing highly dynamic land cover types such as grasslands, croplands, and water bodies, or for ii) estimating soil and/or vegetation moisture content for the year 2020. As a final note, JAXA released an improved GPM ver. 2 that was not available at the time of this study. A preliminary analysis shows that the circa 2020 PALSAR-2 composite product still seems to outperform the 2020 JAXA GPM v2 in northern Canada. ## Additional Information on Dataset: This dataset comprises four raster geotiff files of circa 2020 L-band PALSAR-2 summer temporal composites as mosaics of orthorectified and radiometrically slope corrected dual-polarized HH and HV gamma naught (γ°) backscatter amplitude, along with two radar vegetation indices (HVHH, RFDI), all scaled as 16-bit Digital Number (DN) values with 30-m pixel size in Lambert conformal conic projection. An additional 8-bit RGB quick-view file is also provided. A companion pdf ”Readme” file provides further details about these geotiff files and equations to convert DN values to γ° absolute intensity values. ## Dataset Citation: Beaudoin, A., Villemaire, P., Gignac, C., Tolszczuk, S., Guindon, L., Pontone, N., Millard, C. (2024). Canada’s PALSAR-2 dual-polarized L-band radar summer backscatter composite, circa 2020. Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, Quebec, Canada. https://doi.org/10.23687/8ec4ee78-9240-4bd0-9c97-d3a27829e209 In addition, please provide credits to the Japanese space agency JAXA with the mention “Original Global PALSAR-2/PALSAR Mosaics v1 provided by JAXA (©JAXA)” ## Publication Reference for Product Development and Use in Wetland Mapping: Pontone, N., Millard, K., Thompson, D., Guindon, L., Beaudoin, A. (2024). A hierarchical, Multi-Sensor Framework for Peatland Sub-Class and Vegetation Mapping Throughout the Canadian Boreal Forest. Remote Sensing for Ecology and Conservation (accepted for publication). ## Cited reference: Shimada, M., Itoh, T., Motooka, T., Watanabe, M., Tomohiro, S., Thapa, T., Lucas, R. (2014). New Global Forest/Non-Forest Maps from ALOS PALSAR Data (2007-2010). Remote Sensing of Environment, 155, pp. 13-31. https://doi.org/ 10.1016/j.rse.2014.04.014

This product is the concentration of the photosynthetic pigment (chlorophyll-a) in phytoplankton in the sea surface layer. This is an ongoing dataset with a latency of 3-4 days. GCOM-C conducts long-term and continuous global observation and data collection to elucidate the mechanism behind fluctuations in radiation budget and carbon cycle …

This product is the temperature of terrestrial land surface. This is an ongoing dataset with a latency of 3-4 days. GCOM-C conducts long-term and continuous global observation and data collection to elucidate the mechanism behind fluctuations in radiation budget and carbon cycle needed to make accurate projections regarding future temperature rise. At the same time, cooperating with research institutions having a climate numerical model, it contributes to reduction of errors in temperature rise prediction derived from the climate numerical model and improvement of accuracy of prediction of various environmental changes. SGLI mounted on GCOM-C is the succession sensor of the Global Imager (GLI) mounted on ADEOS-II (MIDORI II) and is the Imaging Radiometer which measures the radiation from near-ultraviolet to thermal infrared region (380 nm-12 um) in 19 channels. Global observation of once for approximately every two days is possible at mid-latitude near Japan by observation width at ground greater than 1,000 km. In addition, SGLI realizes high resolution than the similar global sensor and has a polarized observation function and a multi-angle observation function.

ALOS World 3D - 30m (AW3D30) is a global digital surface model (DSM) dataset with a horizontal resolution of approximately 30 meters (1 arcsec mesh). The dataset is based on the DSM dataset (5-meter mesh version) of the World 3D Topographic Data. More details are available in the dataset documentation. This ingested dataset combines data from versions 3.1, 4.0, and 4.1. Version 4.1 (April 2024): This major update released 19,051 tiles covering global regions (excluding Antarctica and Japan). It incorporates new supplementary data for void filling and corrects partial anomalies found in versions 3.1 and 3.2, along with re-filling voids. For specific tile updates in v4.1, please use the v4.1 filter on map tiles or consult the latest format description. Version 4.0 (April 2023): This update released 1,886 tiles, improving low and mid-latitude regions and areas south of 60 degrees latitude. Key changes include: 1. New supplementary data for void filling. 2. Correction of partial anomalies and re-filling of voids (2 tiles). 3. Updated coastlines for regions south of 60 degrees latitude (44 tiles). 4. Disabled Caspian Sea water mask and supplemented with elevation data (54 tiles). 5. Extracted and corrected new partial anomaly areas in South America (1,786 tiles). 6. For detailed tile information for v4.0, please use the v4.0 filter on map tiles or refer to the format description. Version 3.2, released in January 2021, is an improved version created by reconsidering the format in the high latitude area, auxiliary data, and processing method. Different pixel spacing for each latitude zone was adopted at high latitude area. Coastline data, which is one of the auxiliary datasets, was changed, and new supplementary data was used. In addition, as a source data for Japan, AW3D version 3 was also used. Furthermore, the method of detecting anomalous values in the process was improved. Note: See the code example for the recommended way of computing slope. Unlike most DEMs in Earth Engine, this is an image collection due to multiple resolutions of source files that make it impossible to mosaic them into a single asset, so the slope computations need a reprojection. The AW3D DSM elevation is calculated by an image matching process that uses a stereo pair of optical images. Clouds, snow, and ice are automatically identified during processing and applied the mask information. However, mismatched points sometimes remain especially surrounding (or at the edges of) clouds, snow, and ice areas, which cause some height errors in the final DSM.

In this study, the NDVI data were obtained from Landsat 7 ETM+ and Landsat 8 OLI/TIRS imagery, which underwent atmospheric corrections, cloud removal, and cloud processing using Google Earth Engine (GEE) (https://code.earthengine.google.com). Field-measured MAGT generally at a depth of 15 m were collected for four periods (2010, 2013, 2016, and 2019), primarily in July or August each year, from 16 borehole sites on the northern (K445, MDB, YNG-1, YNG-2, YNG-3, CLP-2, CLP-3, CLP-4, and BSK) and southern (BSKN, CLQ-2, K364-1, K364-2, QSH-1, QSH-2, and QSH-3) flanks of the Bayan Har Mountains. These measurements were taken at the DZAA (Luo et al., 2018b). The digital elevation model (DEM) was sourced from the Japan Aerospace Exploration Agency (JSXA) (https://www.eorc.jaxa.jp/ALOS/en/aw3d30/data/index.htm; accessed on 12 May 2024), and land use and cover change (LUCC) data were obtained from the GlobeLand30 dataset (https://www.globallandcover.com/; accessed on 6 August 2021). Annual precipitation (AP) and land surface temperature (LST) data were retrieved from the Big Earth Data Platform for Three Poles (https://poles.tpdc.ac.cn/zh-hans/; accessed on 16 May 2024) and MAGT data were sourced from the European Space Agency (ESA) climate office (https://catalogue.ceda.ac.uk/uuid/6ebcb73158b14cd5a321b7c0bc6ed393/; accessed on 16 May 2024). MODIS snow cover and evapotranspiration (ET) data were downloaded from the National Tibetan Plateau Data Center (TPDC) (https://data.tpdc.ac.cn; accessed on 03 November 2023). Soil type data were provided by the Chinese Academy of Sciences Resource and Environment Science Data Center (https://www.resdc.cn; accessed on 12 May 2024), while Human Footprint (HFP) data, from (https://doi.org/10.6084/m9.figshare.16571064; accessed on 28 June 2023).

La misión Tropical Rainfall Measuring Mission (TRMM) es una misión conjunta entre la NASA y la Agencia de Exploración Aeroespacial de Japón (JAXA) diseñada para monitorear y estudiar las lluvias tropicales. El producto 34B2 contiene una estimación de la precipitación infrarroja combinada y ajustada por TRMM (mm/h) y del error cuadrático medio de la precipitación, con una resolución temporal de 3 horas y una resolución espacial de 0.25 grados. Consulta la descripción del algoritmo y la especificación del archivo para obtener más detalles. Documentación: Documentación de la PI Especificación de archivos para productos de TRMM Comparación entre las versiones 6 y 7 de TRMM Readme Detalles del algoritmo de TMPA que se usa en este producto TRMM Data Gaps (Vacíos de datos del TRMM) Transición de TMPA a IMERG

This product is the temperature of sea surface. This is an ongoing dataset with a latency of 3-4 days. GCOM-C conducts long-term and continuous global observation and data collection to elucidate the mechanism behind fluctuations in radiation budget and carbon cycle needed to make accurate projections regarding future temperature rise. At the same time, cooperating with research institutions having a climate numerical model, it contributes to reduction of errors in temperature rise prediction derived from the climate numerical model and improvement of accuracy of prediction of various environmental changes. SGLI mounted on GCOM-C is the succession sensor of the Global Imager (GLI) mounted on ADEOS-II (MIDORI II) and is the Imaging Radiometer which measures the radiation from near-ultraviolet to thermal infrared region (380 nm-12 um) in 19 channels. Global observation of once for approximately every two days is possible at mid-latitude near Japan by observation width at ground greater than 1,000 km. In addition, SGLI realizes high resolution than the similar global sensor and has a polarized observation function and a multi-angle observation function.

Puedes encontrar una versión más reciente de este conjunto de datos con datos del período 2015-2021 en JAXA/ALOS/PALSAR/YEARLY/SAR_EPOCH. El mosaico global de 25 m de PALSAR/PALSAR-2 es una imagen SAR global sin interrupciones creada a partir de mosaicos de bandas de imágenes SAR de PALSAR/PALSAR-2. Para cada año y ubicación, los datos de las bandas se seleccionaron a través de la inspección visual de los mosaicos de exploración disponibles durante el período, y se utilizaron preferentemente aquellos que mostraban una respuesta mínima a la humedad de la superficie. En los casos en que la disponibilidad era limitada (p.ej., debido al requisito de observaciones durante emergencias específicas), los datos se seleccionaron necesariamente del año anterior o posterior, incluido el 2006. Shimada et al., 2014 No hay datos disponibles para el período 2011-2014 debido a la brecha entre la cobertura temporal de ALOS y ALOS-2. Las imágenes de SAR se ortorrectificaron y se corrigieron según la pendiente con el Modelo Digital de Elevación SRTM de 90 m. Se aplicó un proceso de eliminación de bandas (Shimada & Isoguchi, 2002, 2010) para igualar las diferencias de intensidad entre las bandas vecinas, que se producen principalmente debido a las diferencias estacionales y diarias en las condiciones de humedad de la superficie. Los datos de polarización se almacenan como números digitales (DN) de 16 bits. Los valores de DN se pueden convertir en valores de gamma naught en unidades de decibelios (dB) con la siguiente ecuación: γ₀ = 10log₁₀(DN²) – 83.0 dB Atención: Los valores de dispersión pueden variar significativamente de una ruta a otra en las áreas forestales de latitudes altas. Esto se debe al cambio en la intensidad de la retrodispersión causada por los árboles congelados en invierno. Puedes encontrar más información en la Descripción del conjunto de datos del proveedor.

Die 25 m PALSAR-2 ScanSAR-Daten sind normalisierte Rückstreudaten des PALSAR-2-Beobachtungsmodus für große Gebiete mit einer Beobachtungsbreite von 350 km. Die SAR-Bilder wurden mithilfe des digitalen Oberflächenmodells ALOS World 3D – 30 m (AW3D30) orthorektifiziert und neigungskorrektiert. Polarisationsdaten werden als 16-Bit-Digitalzahlen (DN) gespeichert. Die DN-Werte können mit der folgenden Gleichung in Gamma-Naught-Werte in Dezibel (dB) umgerechnet werden: γ0 = 10*log10(DN2) – 83,0 dB Daten der Stufe 2.2 sind orthorektifiziert und radiometrisch geländekorrekt. Dieses Dataset ist mit dem Committee on Earth Observation (CEOS)-Standard Analysis Ready Data for LAND (CARD4L) kompatibel.

Les données de rétrodiffusion normalisées de la couverture ScanSAR PALSAR-2 de 25 mètres sont issues du mode d'observation à large zone de PALSAR-2, avec une largeur d'observation de 350 km. L'imagerie SAR a été orthorectifiée et corrigée en fonction de la pente à l'aide du modèle numérique de surface ALOS World 3D - 30 m (AW3D30). Les données de polarisation sont stockées sous forme de nombres numériques de 16 bits. Les valeurs DN peuvent être converties en valeurs gamma naught en décibels (dB) à l'aide de l'équation suivante : γ0 = 10*log10(DN2) - 83,0 dB Les données de niveau 2.2 sont orthorectifiées et corrigées radiométriquement en fonction du terrain. Cet ensemble de données est compatible avec la norme Committee on Earth Observation (CEOS) Analysis Ready Data for LAND (CARD4L).

PALSAR-2 ScanSAR 25 m là dữ liệu tán xạ ngược được chuẩn hoá của chế độ quan sát diện rộng PALSAR-2 với chiều rộng quan sát là 350 km. Hình ảnh SAR được chỉnh sửa theo phương pháp chỉnh hình và độ dốc bằng Mô hình bề mặt kỹ thuật số ALOS World 3D – 30 m (AW3D30). Dữ liệu phân cực được lưu trữ dưới dạng số kỹ thuật số (DN) 16 bit. Bạn có thể chuyển đổi các giá trị DN thành giá trị gamma naught theo đơn vị decibel (dB) bằng phương trình sau: γ0 = 10*log10(DN2) – 83,0 dB Dữ liệu cấp 2.2 được điều chỉnh về hình học và điều chỉnh về địa hình theo phương pháp đo bức xạ. Tập dữ liệu này tương thích với tiêu chuẩn Uỷ ban Quan sát Trái đất (CEOS) Dữ liệu sẵn sàng phân tích cho LAND (CARD4L).

El mosaico global de 25 m de PALSAR/PALSAR-2 es una imagen SAR global continua creada a partir de mosaicos de bandas de imágenes SAR de PALSAR/PALSAR-2. Para cada año y ubicación, los datos de las tiras se seleccionaron a través de la inspección visual de los mosaicos de exploración disponibles durante el período, y se prefirieron aquellos que mostraban una respuesta mínima a la humedad de la superficie…

Bạn có thể tìm thấy phiên bản mới hơn của tập dữ liệu này có dữ liệu từ năm 2015 đến năm 2021 trong JAXA/ALOS/PALSAR/YEARLY/SAR_EPOCH Bức ảnh ghép PALSAR/PALSAR-2 25m trên toàn cầu là một bức ảnh SAR liền mạch trên toàn cầu, được tạo bằng cách ghép các dải hình ảnh SAR từ PALSAR/PALSAR-2. Đối với mỗi năm và vị trí, dữ liệu dải được chọn thông qua kiểm tra trực quan các hình ảnh thu nhỏ duyệt xem có sẵn trong khoảng thời gian đó, trong đó những hình ảnh cho thấy phản ứng tối thiểu đối với độ ẩm bề mặt được ưu tiên sử dụng. Trong trường hợp dữ liệu có hạn (ví dụ: do yêu cầu quan sát trong các trường hợp khẩn cấp cụ thể), dữ liệu bắt buộc phải được chọn từ năm trước hoặc năm sau, kể cả từ năm 2006. Shimada và cộng sự, 2014 Không có dữ liệu cho giai đoạn 2011 – 2014 do khoảng trống giữa phạm vi thời gian của ALOS và ALOS-2. Hình ảnh SAR được chỉnh sửa theo phương pháp chỉnh hình và độ dốc bằng Mô hình độ cao kỹ thuật số SRTM 90m. Quy trình loại bỏ sọc (Shimada & Isoguchi, 2002, 2010) được áp dụng để cân bằng sự khác biệt về cường độ giữa các dải lân cận, chủ yếu là do sự khác biệt theo mùa và hằng ngày về điều kiện độ ẩm bề mặt. Dữ liệu phân cực được lưu trữ dưới dạng số kỹ thuật số (DN) 16 bit. Bạn có thể chuyển đổi các giá trị DN thành giá trị gamma naught theo đơn vị decibel (dB) bằng phương trình sau: γ₀ = 10log₁₀(DN²) – 83,0 dB Chú ý: Giá trị tán xạ ngược có thể thay đổi đáng kể từ đường dẫn này sang đường dẫn khác trên các khu vực rừng ở vĩ độ cao. Nguyên nhân là do sự thay đổi về cường độ tán xạ ngược do cây bị đóng băng vào mùa đông. Bạn có thể xem thêm thông tin trong phần Mô tả tập dữ liệu của nhà cung cấp.