Fusing Landsat 8 and Sentinel-2 data for 10-m dense time-series imagery using a degradation-term constrained deep network

Wu, Jingan; Lin, Liupeng; Li, Tongwen; Cheng, Qing; Zhang, Chi; Shen, Huanfeng

doi:10.1016/j.jag.2022.102738

Cited by 14 publications

(16 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…By integrating additional data sources, we believe that the accuracy and applicability of these indices will be significantly improved. Furthermore, using Harmonized Landsat-Sentinel (HLS) data and downscaled Landsat data based on the Sentinel 2 10 m resolution images (Wu et al, 2022), could further help to increase the usability of these indices, at least for the years covered by Sentinel 2 (+2016).…”

Section: Future Workmentioning

confidence: 99%

Time-series of Landsat-based spectral indices for continental Europe for 2000--2022 to support soil health monitoring

Tian,

Hengl,

Consoli

et al. 2024

Preprint

View full text Add to dashboard Cite

As the importance of soil health in supporting European ecosystems and agriculture becomes increasingly critical, effective monitoring is essential for its protection and restoration. This paper describes production and quality assessment of a data cube (17TB in size) derived from Landsat Analysis Ready Data version 2 (ARD V2) dataset and tailored for soil health monitoring, featuring multiple spectral indices, long time span, high resolution, and analysis readiness. The focus was on indices of importance for monitoring soil health: Normalized Difference Vegetation Index (NDVI), Soil Adjusted Vegetation Index (SAVI), Fraction of Absorbed Photosynthetically Active Radiation (FAPAR), Normalized Difference Snow Index (NDSI), Normalized Difference Water Index (NDWI), Normalized Difference Tillage Index (NDTI), minimum Normalized Difference Tillage Index (minNDTI), Bare Soil Fraction (BSF), Number of Seasons (NOS) and Crop Duration Ratio (CDR). The data set is available with a resolution of 30~m, with bimonthly (i.e. one image available per two months), annual, and long-term analysis throughout continental Europe, including Ukraine, the UK, and Turkey, covering 2000 to 2022. The results of the quality assessment, including both visual examination and plausibility check with ground survey data, show that these indices can effectively capture environmental processes and offer insight into soil health through aspects such as vegetation, crop status, tillage practices, and soil exposure directly or serve as covariates in digital soil mapping. In particular, the BSF shows a strong negative correlation of -0.73 with crop coverage data from 2006 to 2017, suggesting an effective detection of soil exposure. The minNDTI shows a moderate positive correlation of 0.57 with the Eurostat tillage practices survey data, indicating that it provides valuable information on tillage intensity, although not definitively. This data set is analysis-ready and cloud-optimized, making it suitable for applications in soil property mapping and the development of comprehensive soil health assessment approaches throughout Europe.

show abstract

Section: Future Workmentioning

confidence: 99%

Time-series of Landsat-based spectral indices for continental Europe for 2000--2022 to support soil health monitoring

Tian,

Hengl,

Consoli

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…In southern China, smallholder-operated croplands are often less than 0.07 ha in size, smaller than the size of a single Landsat pixel, making it challenging to differentiate between rapeseed and other crops 67 . One potential solution to this issue is to utilize Sentinel-2 data to create 10 m Landsat image products 74 , 75 . The availability of Landsat observations can also have an impact on CARM30 accuracy in some regions with heavy cloud cover.…”

Section: Usage Notesmentioning

confidence: 99%

CARM30: China annual rapeseed maps at 30 m spatial resolution from 2000 to 2022 using multi-source data

Liu,

Li,

Tao

et al. 2024

Sci Data

View full text Add to dashboard Cite

Rapeseed is a critical cash crop globally, and understanding its distribution can assist in refined agricultural management, ensuring a sustainable vegetable oil supply, and informing government decisions. China is the leading consumer and third-largest producer of rapeseed. However, there is a lack of widely available, long-term, and large-scale remotely sensed maps on rapeseed cultivation in China. Here this study utilizes multi-source data such as satellite images, GLDAS environmental variables, land cover maps, and terrain data to create the China annual rapeseed maps at 30 m spatial resolution from 2000 to 2022 (CARM30). Our product was validated using independent samples and showed average F1 scores of 0.869 and 0.971 for winter and spring rapeseed. The CARM30 has high spatial consistency with existing 10 m and 20 m rapeseed maps. Additionally, the CARM30-derived rapeseed planted area was significantly correlated with agricultural statistics (R2 = 0.65–0.86; p < 0.001). The obtained rapeseed distribution information can serve as a reference for stakeholders such as farmers, scientific communities, and decision-makers.

show abstract

“…Additionally, since band 11-SWIR 1 had a spatial resolution of 20 m, an interpolation and resampling process was carried out at 10 m, supported by the 10 m resolution bands. For this process, a convolutional neural network (Wu et al 2022) was optimized and applied, implemented in the free Python programming language using the TensorFlow and Keras libraries (Brownlee 2016).…”

Section: Acquisition Of Multi-band Satellite Images With High Tempora...mentioning

confidence: 99%

Population Dynamics and Estimation of Damage of the Spittlebug Aeneolamia varia on Sugarcane in Colombia by Using remote Sensing and Machine Learning Tools

et al. 2023

View full text Add to dashboard Cite

The spittlebug (Aeneolamia varia) is one of the most important sugarcane pests in Colombia, where a recent increase in population and distribution specially in southwestern Colombia have led to the need for new technologies for integrated pest management. The objectives of this study were to determine the spatial distribution of this pest in commercial sugarcane fields and to validate machine learning (ML) tools for indirect injury detection and impact on yield (damage) using satellite images. This study was carried out in fields grown with the CC 01-1940 variety in El Cerrito, Valle del Cauca, Colombia, where systematic sampling of the populations (number of adults and nymphs per stem) was carried out. The spatial aggregation and distribution were determined using Moran’s index and point patterns, sequence observations, and analysis with distance indicators (Sadie). The indirect injury detection and quantification of the impact on production were carried out with a ML approach using satellite image products with 10 m spatial and five days temporal resolutions, obtained from a Sentinel-2 sensor using Google Earth Engine. The results indicated that spittlebug populations had an aggregate spatial behavior and high spatial dependence. In addition, the ML algorithms predicted spittlebug injury, and the effect on production was estimated at 26.4 tons of cane per hectare, which represented a 17% reduction in the expected yield. The use of spatial analysis and remote sensing tools are an alternative for indirect detection of injury and for understanding population dynamics of the pest in sugarcane, so they can become instrumental for decision-making on an integrated pest management program.

show abstract

Fusing Landsat 8 and Sentinel-2 data for 10-m dense time-series imagery using a degradation-term constrained deep network

Cited by 14 publications

References 32 publications

Time-series of Landsat-based spectral indices for continental Europe for 2000--2022 to support soil health monitoring

Time-series of Landsat-based spectral indices for continental Europe for 2000--2022 to support soil health monitoring

CARM30: China annual rapeseed maps at 30 m spatial resolution from 2000 to 2022 using multi-source data

Population Dynamics and Estimation of Damage of the Spittlebug Aeneolamia varia on Sugarcane in Colombia by Using remote Sensing and Machine Learning Tools

Contact Info

Product

Resources

About