Dynamic monitoring of the Poyang Lake wetland by integrating Landsat and MODIS observations

Chen, Bin; Chen, Lifan; Huang, Bo; Michishita, Ryo; Xu, Bing

doi:10.1016/j.isprsjprs.2018.02.021

Cited by 118 publications

(42 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By using TPS, residual distribution not only ensures that the re-aggregated fused fine-resolution image exactly matches the original coarse resolution image, but also help to improve accuracy of individual subpixels. Previous research also reported better accuracy of FSDAF compared to STARFM and UBDF in various application scenarios [7,[50][51][52]. Please note that FSDAF has some shortcomings.…”

Section: Model Characteristics and Applicable Situationsmentioning

confidence: 83%

Comparison of Five Spatio-Temporal Satellite Image Fusion Models over Landscapes with Various Spatial Heterogeneity and Temporal Variation

Liu

Yin

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

In recent years, many spatial and temporal satellite image fusion (STIF) methods have been developed to solve the problems of trade-off between spatial and temporal resolution of satellite sensors. This study, for the first time, conducted both scene-level and local-level comparison of five state-of-art STIF methods from four categories over landscapes with various spatial heterogeneity and temporal variation. The five STIF methods include the spatial and temporal adaptive reflectance fusion model (STARFM) and Fit-FC model from the weight function-based category, an unmixing-based data fusion (UBDF) method from the unmixing-based category, the one-pair learning method from the learning-based category, and the Flexible Spatiotemporal DAta Fusion (FSDAF) method from hybrid category. The relationship between the performances of the STIF methods and scene-level and local-level landscape heterogeneity index (LHI) and temporal variation index (TVI) were analyzed. Our results showed that (1) the FSDAF model was most robust regardless of variations in LHI and TVI at both scene level and local level, while it was less computationally efficient than the other models except for one-pair learning; (2) Fit-FC had the highest computing efficiency. It was accurate in predicting reflectance but less accurate than FSDAF and one-pair learning in capturing image structures; (3) One-pair learning had advantages in prediction of large-area land cover change with the capability of preserving image structures. However, it was the least computational efficient model; (4) STARFM was good at predicting phenological change, while it was not suitable for applications of land cover type change; (5) UBDF is not recommended for cases with strong temporal changes or abrupt changes. These findings could provide guidelines for users to select appropriate STIF method for their own applications.

show abstract

Section: Model Characteristics and Applicable Situationsmentioning

confidence: 83%

Comparison of Five Spatio-Temporal Satellite Image Fusion Models over Landscapes with Various Spatial Heterogeneity and Temporal Variation

Liu

Yin

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…As a result, a unique lake-vegetation system exists (Figure 1), posing a challenge for accurate lake surface mapping. [6,36,44,48,49,[55][56][57]. In contrast, vegetation generally has positive NDVI values.…”

Section: Study Areamentioning

confidence: 99%

“…First, a synergy of high-resolution images (e.g., Landsat-8 OLI and Sentinel-2 MSI as in [73,74]) provide an improved reference dataset for validation purposes. Second, high-resolution satellite images can be integrated with the MinVC NDVI to derive long-term high-resolution water surface area data (e.g., [36,37]). These data are expected to benefit monitoring small lakes with an area even in the order of several MODIS pixels [66].…”

Section: Improvements Of Current Ndvi Minvc Algorithmmentioning

confidence: 99%

Compositing the Minimum NDVI for Daily Water Surface Mapping

Fan

Liu

et al. 2020

Remote Sensing

View full text Add to dashboard Cite

Capturing high frequency water surface dynamics via optical remote sensing is important for understanding hydro-ecological processes over seasonally flooded wetlands. However, it is a difficult task due to the presence of clouds on satellite images. This study proposed the MODerate-resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (NDVI) Minimum Value Composite (MinVC) algorithm to generate daily water surface data at a 250-m resolution. The algorithm selected pixelwise minimum values from the combined daily Terra and Aqua MODIS NDVI data within a 15-day moving window. Consisting mainly of cloud and water surface information, the MinVC NDVI data were segmented for water surfaces over the Poyang Lake, China (2000–2017) by using an edge detection model. The water surface mapping result was strongly correlated with the Landsat based result (R2 = 0.914, root mean square error, RMSE = 223.7 km2), the cloud free MODIS image based result (R2 = 0.824, RMSE = 356.7 km2), the recent Landsat-MODIS image fusion based result (R2 = 0.765, RMSE = 403 km2), and the hydrodynamic modeling result (R2 = 0.799). Compared to the equivalent eight-day MOD13 NDVI based on the Constraint View-Angle Maximum Value Composite (CV-MVC) algorithm, the daily MinVC NDVI highlighted water bodies by generating spatially homogenous water surface information. Consequently, the algorithm provided spatially and temporally continuous data for calculating water submersion times and trends in water surface area, which contribute to a better understanding of hydro-ecological processes over seasonally flooded wetlands. Within the framework of sensor intercalibration, the algorithm can be extended to incorporate multiple sensor data for improved water surface mapping.

show abstract

“…Due to its relatively simple implementation, linear spatiotemporal fusion methods have been utilized in various applications, such as land-cover classification [15,16], wetland monitoring [17], land surface temperature monitoring [18,19], leaf area index monitoring [20,21], and evapotranspiration monitoring [22,23]. However, this type of method has some major limitations: (1) linear theoretical assumptions are implausible in the case of land-cover change, resulting in poor fusion performance in land-cover change prediction; and (2) the effectiveness of linear spatiotemporal fusion methods depends on the selection of the weighting function, which is empirical with limited generalization [24].…”

Section: Introductionmentioning

confidence: 99%

A Novel Deep Learning-Based Spatiotemporal Fusion Method for Combining Satellite Images with Different Resolutions Using a Two-Stream Convolutional Neural Network

et al. 2020

View full text Add to dashboard Cite

Spatiotemporal fusion is considered a feasible and cost-effective way to solve the trade-off between the spatial and temporal resolution of satellite sensors. Recently proposed learning-based spatiotemporal fusion methods can address the prediction of both phenological and land-cover change. In this paper, we propose a novel deep learning-based spatiotemporal data fusion method that uses a two-stream convolutional neural network. The method combines both forward and backward prediction to generate a target fine image, where temporal change-based and a spatial information-based mapping are simultaneously formed, addressing the prediction of both phenological and land-cover changes with better generalization ability and robustness. Comparative experimental results for the test datasets with phenological and land-cover changes verified the effectiveness of our method. Compared to existing learning-based spatiotemporal fusion methods, our method is more effective in predicting phenological change and directly reconstructing the prediction with complete spatial details without the need for auxiliary modulation.

show abstract

Dynamic monitoring of the Poyang Lake wetland by integrating Landsat and MODIS observations

Cited by 118 publications

References 53 publications

Comparison of Five Spatio-Temporal Satellite Image Fusion Models over Landscapes with Various Spatial Heterogeneity and Temporal Variation

Comparison of Five Spatio-Temporal Satellite Image Fusion Models over Landscapes with Various Spatial Heterogeneity and Temporal Variation

Compositing the Minimum NDVI for Daily Water Surface Mapping

A Novel Deep Learning-Based Spatiotemporal Fusion Method for Combining Satellite Images with Different Resolutions Using a Two-Stream Convolutional Neural Network

Contact Info

Product

Resources

About