diff --git a/datasets/HLS_Damage_Probability_Cover_Image_FL.jpg b/datasets/HLS_Damage_Probability_Cover_Image_FL.jpg new file mode 100644 index 000000000..98f836129 Binary files /dev/null and b/datasets/HLS_Damage_Probability_Cover_Image_FL.jpg differ diff --git a/datasets/damage-probability-ian.data.mdx b/datasets/damage-probability-ian.data.mdx new file mode 100644 index 000000000..6717d40e1 --- /dev/null +++ b/datasets/damage-probability-ian.data.mdx @@ -0,0 +1,87 @@ +--- +id: damage_probability_2022-10-03 +name: "NRT HLS-Derived Damage Probability Index" +description: "Near real-time monitoring of land disturbances for CONUS based on the 30-m Harmonized Landsat Sentinel-2 (HLS) dataset." +media: + src: ::file ./HLS_Damage_Probability_Cover_Image_FL.jpg + alt: Satellite imagery over Florida showing damage probability (using Viridis color ramp, with yellow being high probability and purple being low probability) for Oct 3, 2023. + author: + name: CONUS Disturbance Watcher + url: https://gers.users.earthengine.app/view/nrt-conus +taxonomy: + - name: Topics + values: + - Disasters +layers: + - id: damage_probability_2022-10-03 + stacCol: damage_probability_2022-10-03 + name: Damage Probability + type: raster + description: "DPI values from 0 to 99. 0: no damage; 99: damage mostly likely" + zoomExtent: + - 0 + - 20 + sourceParams: + colormap_name: magma + rescale: + - 0 + - 99 + legend: + type: gradient + min: "0" + max: "99" + stops: + - '#000004' + - '#180f3d' + - '#440f76' + - '#721f81' + - '#9e2f7f' + - '#cd4071' + - '#f1605d' + - '#fd9668' + - '#fec287' + - '#f0f921' +--- + + + +The contiguous United States (CONUS) Disturbance Watcher system was built based on time-series analysis and machine learning techniques. The system consisted of two components. The first component, the retrospective disturbance analysis, extracted disturbance predictors from historical open access disturbance products and satellite datasets and built a series of machine learning models respectively for different lag stages. The second component, namely the near real-time monitoring, recursively updates per-pixel time-series models and outputted current vectors of disturbance predictor using Stochastic Continuous Change Detection (S-CCD) algorithm, and then applied the offline machine learning models from the first component to map disturbance probabilities and patches. The system could be updated regularly in an interval of one week, or updated as an user-defined interval, such as one day that was used for Hurricane Ian monitoring. The per-day updating interval provided the quickest mapping response and the most extensive damage region as some would be recovered after days, while the highest computational resources were needed. + +- **Temporal Extent:** January 1, 2022 - October 3, 2022 +- **Temporal Resolution:** Daily +- **Spatial Extent:** Continental United States +- **Spatial Resolution:** 30 m x 30 m +- **Data Units:** Disturbance probability from 0 to 1 +- **Data Type:** Research +- **Data Latency:** Daily + + +**Scientific Details:** +The Hurricane Ian disturbance probability layer shows the likelihood impacted by Hurricane Ian. The data came from the map production of [CONUS Disturbance Watcher system](https://gers.users.earthengine.app/view/nrt-conus) on Oct. 3th, 2022, when the first post-hurricane Harmonized Landsat Sentinel-2 (HLS) 2.0 dataset on Sep. 30th, 2022 became available for downloading. The [HLS dataset](https://hls.gsfc.nasa.gov/) produce harmonized surface reflectance products from four satellite sensors, i.e., Landsat-8, Landsat-9, Sentinel-2A and Sentinel-2B, providing observation once every 2-3 days. The probability layer (scaled by 100) describes the probability produced by machine learning models. The higher probability reveals the high spectral similarity to the historical disturbance. + + + + + +## Source Data Product Citation +Ye, S, Zhu, Z, 2023, Near Real-time Hurricane Ian Disturbance Probability Map: Global Environmental Remote Sensing Laboratory release + +## Disclaimer +All data provided in VEDA has been transformed from the original format (TIFF) into Cloud Optimized GeoTIFF ([COG](https://www.cogeo.org/)). Careful quality checks are used to ensure data transformation has been performed correctly. + +## Key Publications +Ye, S., Zhu, Z., & Suh, J. W. (2024). Leveraging past information and machine learning to accelerate land disturbance monitoring. Remote Sensing of Environment, 305, 114071, [https://doi.org/10.31223/X5WT2H](https://doi.org/10.31223/X5WT2H) . + +### Other Relevant Publications +Ye, S., Rogan, J., Zhu, Z., & Eastman, J. R. (2021). A near-real-time approach for monitoring forest disturbance using Landsat time series: Stochastic continuous change detection. Remote Sensing of Environment, 252, 112167, [https://doi.org/10.1016/j.rse.2020.112167](https://doi.org/10.1016/j.rse.2020.112167). + +Ye, S., Zhu, Z., & Cao, G. (2023). Object-based continuous monitoring of land disturbances from dense Landsat time series. Remote Sensing of Environment, 287, 113462, [https://doi.org/10.1016/j.rse.2023.113462](https://doi.org/10.1016/j.rse.2023.113462). + + ## Acknowledgment +This work has been supported by the USGS-NASA Landsat Science Team (LST) Program for Toward Near Real-time Monitoring and Characterization of Land Surface Change for the Conterminous US (140G0119C0008) + +## License +[Creative Commons Attribution 1.0 International](https://creativecommons.org/publicdomain/zero/1.0/legalcode) (CC BY 1.0) + + + diff --git a/datasets/entropy-difference-ian.data.mdx b/datasets/entropy-difference-ian.data.mdx new file mode 100644 index 000000000..2df23b35c --- /dev/null +++ b/datasets/entropy-difference-ian.data.mdx @@ -0,0 +1,82 @@ +--- +id: hls-entropy-difference +name: "Harmonized Landsat and Sentinel-2 Entropy Difference" +description: "Using Harmonized Landsat and Sentinel-2 to Analyze the Aftermath of Hurricane Ian in Lee and Charlotte Counties in Florida." +media: + src: ::file ../stories/ian_goes_cover.jpg + alt: Hurricane Ian as seen from space as it makes landfall with the state of Florida. NASA Earth Observatory image. + author: + name: Joshua Stevens, using GOES 16 imagery courtesy of NOAA and the National Environmental Satellite, Data, and Information Service (NESDIS) + url: https://visibleearth.nasa.gov/images/150408/hurricane-ian-reaches-florida +taxonomy: + - name: Topics + values: + - EIS +layers: + - id: hls-entropy-difference + stacCol: hls-entropy-difference + name: Entropy Difference + type: raster + description: "Bitemporal Different with higher values indicating higher likelihood of change from before to after Ian." + zoomExtent: + - 0 + - 20 + sourceParams: + colormap_name: bwr + rescale: -1,1 + legend: + type: gradient + min: "-1" + max: "1" + stops: + - '#0000ff' + - '#6666ff' + - '#ccccff' + - '#ffffff' + - '#ffcccc' + - '#ff6666' + - '#ff0000' +--- + + + +Harmonized Landsat Sentinel-2 (HLS) project from NASA is designed to integrate and harmonize data from multiple satellite sources, specifically the Operation Land Imager (OLI) on Landsat-8/9 and the Mult-Spectral Instrument (MSI) on Sentinel-2A/B satellites. This project aims to create a seamless surface reflectance record that is essential for various Earth Observation and monitoring tasks. + +- **Temporal Extent:** Landsat-9 2021-10-31; Sentinel-2B 2017-07-06 +- **Temporal Resolution:** ~3 days +- **Spatial Extent:** Global with the exception of Antarctica +- **Spatial Resolution:** 30 m x 30 m +- **Data Units:** Surface Reflectance +- **Data Type:** Research +- **Data Latency:** 2 to 3 days + + +**Scientific Details:** +HLS project incorporates several advanced scientific methodologies and technologies to harmonize data from the Landsat and Sentinel-2 satellites such as atmospheric correction, geographic co-registration and common gridding, bidirectional reflectance distribution normalization, and cloud-shadow masking. The calculation of entropy can be accomplished using the 'entropy' function from the 'scipy.stats' module, which computes the entropy of a distribution for given probability values. This function calculates the Shannon entropy. This metric quantifies the randomness or unpredictability inherent in the dataset's distribution from one time point to another. + + + + + +## Source Data Product Citation +Claverie, M., Ju, J., Masek, J. G., Dungan, J. L., Vermote, E. F., Roger, J.-C., Skakun, S. V., & Justice, C. (2018). The Harmonized Landsat and Sentinel-2 surface reflectance data set. Remote Sensing of Environment, 219, 145-161. +## Disclaimer +All data provided in VEDA has been transformed from the original format (TIFF) into Cloud Optimized GeoTIFF ([COG](https://www.cogeo.org/)). Careful quality checks are used to ensure data transformation has been performed correctly. + +## Key Publications +Su Ye, John Rogan, Zhe Zhu, J. Ronald Eastman, A near-real-time approach for monitoring forest disturbance using Landsat time series: stochastic continuous change detection, Remote Sensing of Environment, Volume 252, 2021,112167, ISSN 0034-4257, (https://doi.org/10.1016/j.rse.2020.112167)[https://doi.org/10.1016/j.rse.2020.112167]. + + +Su Ye, Zhe Zhu, Guofeng Cao, Object-based continuous monitoring of land disturbances from dense Landsat time series, Remote Sensing of Environment, Volume 287, 2023, 113462, ISSN 0034-4257, (https://doi.org/10.1016/j.rse.2023.113462)[https://doi.org/10.1016/j.rse.2023.113462]. + +### Other Relevant Publications +Ye, S., Zhu, Z., & Suh, J. W. (2024). Leveraging past information and machine learning to accelerate land disturbance monitoring. Remote Sensing of Environment, 305, 114071. + + ## Acknowledgment +This work has been supported by the USGS-NASA Landsat Science Team (LST) Program for Toward Near Real-time Monitoring and Characterization of Land Surface Change for the Conterminous US (140G0119C0008) + +## License +[Creative Commons Attribution 1.0 International](https://creativecommons.org/publicdomain/zero/1.0/legalcode) (CC BY 1.0) + + + \ No newline at end of file diff --git a/datasets/hls-ndvi-ian.data.mdx b/datasets/hls-ndvi-ian.data.mdx new file mode 100644 index 000000000..be50b3cb7 --- /dev/null +++ b/datasets/hls-ndvi-ian.data.mdx @@ -0,0 +1,108 @@ +--- +id: hls-ndvi-ian +name: "Harmonized Landsat and Sentinel-2 Normalized Difference Vegetation Index" +description: "Using Harmonized Landsat and Sentinel-2 to Analyze the Aftermath of Hurricane Ian in Lee and Charlotte Counties in Florida." +media: + src: ::file ../stories/ian_goes_cover.jpg + alt: Hurricane Ian as seen from space as it makes landfall with the state of Florida. NASA Earth Observatory image. + author: + name: Joshua Stevens, using GOES 16 imagery courtesy of NOAA and the National Environmental Satellite, Data, and Information Service (NESDIS) + url: https://visibleearth.nasa.gov/images/150408/hurricane-ian-reaches-florida +taxonomy: + - name: Topics + values: + - Hurricane + - UAH +layers: + - id: hls-ndvi + stacCol: hls-ndvi + name: NDVI + type: raster + description: "NDVI: 0 to 1; 0 = little to no vegetation; 1 = heavy vegetation" + zoomExtent: + - 0 + - 20 + sourceParams: + colormap_name: rdylgn + rescale: -1,1 + legend: + type: gradient + min: "-1" + max: "1" + stops: + - "#a50026" + - "#f46d43" + - "#fee08b" + - "#d9ef8b" + - "#66bd63" + - "#006837" + compare: + datasetId: hls-ndvi-ian + layerId: hls-ndvi + - id: ndvi_difference + stacCol: hls-ndvi-difference + name: NDVI Difference + type: raster + description: "NDVI Difference: -1 to 1; -1 = decrease in vegetation; 1 = increase in vegetation" + initialDatetime: newest + zoomExtent: + - 0 + - 20 + sourceParams: + colormap_name: rdbu + rescale: -1,1 + legend: + type: gradient + min: "-1" + max: "1" + stops: + - "#67001f" + - "#d6604d" + - "#fddbc7" + - "#d1e5f0" + - "#4393c3" + - "#053061" +--- + + + +Harmonized Landsat Sentinel-2 (HLS) project from NASA is designed to integrate and harmonize data from multiple satellite sources, specifically the Operation Land Imager (OLI) on Landsat-8/9 and the Mult-Spectral Instrument (MSI) on Sentinel-2A/B satellites. This project aims to create a seamless surface reflectance record that is essential for various Earth Observation and monitoring tasks. + +- **Temporal Extent:** Landsat-9 2021-10-31; Sentinel-2B 2017-07-06 +- **Temporal Resolution:** ~3 days +- **Spatial Extent:** Global with the exception of Antarctica +- **Spatial Resolution:** 30 m x 30 m +- **Data Units:** Surface Reflectance +- **Data Type:** Research +- **Data Latency:** 2 to 3 days + + +**Scientific Details:** +HLS project incorporates several advanced scientific methodologies and technologies to harmonize data from the Landsat and Sentinel-2 satellites such as atmospheric correction, geographic co-registration and common gridding, bidirectional reflectance distribution normalization, and cloud-shadow masking. To calculate the Normalized Difference Vegetation Index (NDVI) from the HLS-2 Dataset, we utilize the red and near-infrared bands to assess vegetation health by applyin the formula NDVI = NIR - Red / NIR + Red, where 'NIR' refers to the near-infrared surface reflectance, and 'Red' denotes the red light surface reflectance, both harmonized from the Landsat and Sentinel satellites. + + + + + +## Source Data Product Citation +Claverie, M., Ju, J., Masek, J. G., Dungan, J. L., Vermote, E. F., Roger, J.-C., Skakun, S. V., & Justice, C. (2018). The Harmonized Landsat and Sentinel-2 surface reflectance data set. Remote Sensing of Environment, 219, 145-161. +## Disclaimer +All data provided in VEDA has been transformed from the original format (TIFF) into Cloud Optimized GeoTIFF ([COG](https://www.cogeo.org/)). Careful quality checks are used to ensure data transformation has been performed correctly. + +## Key Publications +Su Ye, John Rogan, Zhe Zhu, J. Ronald Eastman, A near-real-time approach for monitoring forest disturbance using Landsat time series: stochastic continuous change detection, Remote Sensing of Environment, Volume 252, 2021,112167, ISSN 0034-4257, (https://doi.org/10.1016/j.rse.2020.112167)[https://doi.org/10.1016/j.rse.2020.112167]. + + +Su Ye, Zhe Zhu, Guofeng Cao, Object-based continuous monitoring of land disturbances from dense Landsat time series, Remote Sensing of Environment, Volume 287, 2023, 113462, ISSN 0034-4257, (https://doi.org/10.1016/j.rse.2023.113462)[https://doi.org/10.1016/j.rse.2023.113462]. + +### Other Relevant Publications +Ye, S., Zhu, Z., & Suh, J. W. (2024). Leveraging past information and machine learning to accelerate land disturbance monitoring. Remote Sensing of Environment, 305, 114071. + + ## Acknowledgment +This work has been supported by the USGS-NASA Landsat Science Team (LST) Program for Toward Near Real-time Monitoring and Characterization of Land Surface Change for the Conterminous US (140G0119C0008) + +## License +[Creative Commons Attribution 1.0 International](https://creativecommons.org/publicdomain/zero/1.0/legalcode) (CC BY 1.0) + + + \ No newline at end of file diff --git a/stories/comp_damage_sanibel.jpg b/stories/comp_damage_sanibel.jpg new file mode 100644 index 000000000..cea3061a4 Binary files /dev/null and b/stories/comp_damage_sanibel.jpg differ diff --git a/stories/comp_fort_meyers_dmg.jpg b/stories/comp_fort_meyers_dmg.jpg new file mode 100644 index 000000000..6e6932d05 Binary files /dev/null and b/stories/comp_fort_meyers_dmg.jpg differ diff --git a/stories/comp_sanibel_causeway.jpg b/stories/comp_sanibel_causeway.jpg new file mode 100644 index 000000000..549ddeea2 Binary files /dev/null and b/stories/comp_sanibel_causeway.jpg differ diff --git a/stories/comp_surge_track_figure.jpg b/stories/comp_surge_track_figure.jpg new file mode 100644 index 000000000..0fd7b1fce Binary files /dev/null and b/stories/comp_surge_track_figure.jpg differ diff --git a/stories/dpi.jpg b/stories/dpi.jpg new file mode 100644 index 000000000..4510f51ac Binary files /dev/null and b/stories/dpi.jpg differ diff --git a/stories/entropy_difference.jpg b/stories/entropy_difference.jpg new file mode 100644 index 000000000..041c4c665 Binary files /dev/null and b/stories/entropy_difference.jpg differ diff --git a/stories/fort-myers.jpeg b/stories/fort-myers.jpeg new file mode 100644 index 000000000..60605a794 Binary files /dev/null and b/stories/fort-myers.jpeg differ diff --git a/stories/hurricane-ian.stories.mdx b/stories/hurricane-ian.stories.mdx new file mode 100644 index 000000000..cb0eba5bb --- /dev/null +++ b/stories/hurricane-ian.stories.mdx @@ -0,0 +1,247 @@ +--- +id: hurricane-ian +name: Hurricane Ian and Impacts on Vegetation +description: "A in-depth look on how hurricane Ian impacted vegetation on Sanibel Island near Fort Myers." +media: + src: ::file ./ian_goes_cover.jpg + alt: Hurricane Ian as seen from space as it makes landfall with the state of Florida. NASA Earth Observatory image. + author: + name: Joshua Stevens, using GOES 16 imagery courtesy of NOAA and the National Environmental Satellite, Data, and Information Service (NESDIS) + url: https://visibleearth.nasa.gov/images/150408/hurricane-ian-reaches-florida +pubDate: 2024-07-17 +taxonomy: + - name: Topics + values: + - UAH + - Hurricanes +--- + + + + ## Introduction + Trent Cowan1, Andrew Blackford1, Udaysankar Nair1, and Ashley Riddle1 + + 1: University of Alabama in Huntsville + + Disclaimer: This research is ongoing and is not yet peer-reviewed. + + Atlantic hurricane season stretches from June 1 to November 30, and generally reaches a peak in August and September. On September 28, 2022, Hurricane Ian made landfall in southwest Florida as a Category 4 hurricane (based on the Saffir-Simpson Wind Scale), with estimated wind speeds of 150 MPH. The toll was high, leaving 158 people dead and producing $110 billion in damage in Florida alone. The hurricane also left a devastating impact on the landscape in portions of southwest Florida near Fort Myers. A storm surge reached unprecedented levels of 12 to 18 feet around the regions of Cape Coral and Fort Myers. A storm surge is the sudden rise in ocean water above normal tide levels during a storm such as a tropical cyclone or other strong low pressure system as wind from the storm pushes water ashore. An analysis in Figure 1 from the U.S. National Hurricane Center’s review of Hurricane Ian shows the storm surge’s heavy impact on the southwest Florida coast. The storm surge is attributed as the primary cause of fatalities along the coast of southwestern Florida. + + + + + + + This work aims to provide a comprehensive analysis of Hurricane Ian’s devastating environmental impact on southwestern Florida and explore the critical role of remote sensing in monitoring and mitigating the effects of natural disasters using advanced remote sensing techniques. Parameters such as Normalized Difference Vegetation Index (NDVI) and entropy changes offer insights into Ian’s effects on vegetation and surface textures. The HLS (Harmonized Landsat and Sentinel-2) project serves as a great tool in this analysis by providing essential data for response and recovery; moreover, this article introduces the UCONN Geographic Environmental Remote Sensing (GERS) lab’s CONUS Disturbance Water to highlight the importance of near real-time land disturbance detection. + + ## Inland Effects of Ian + On September 24, Hurricane Ian became a named storm southeast of Jamaica. Ian quickly intensified into a hurricane and made landfall in western Cuba near Havana as a Category 3 storm, where it caused catastrophic damage and widespread power outages. After exiting Cuba, the storm quickly intensified into a massive Category 5 storm near the southwestern coast of Florida with sustained wind speeds in excess of 158 mph. As Ian continued moving north and west, the storm made landfall on Florida’s Cayo Costa Island as a strong Category 4 hurricane, with maximum sustained winds of 149 mph. Ian weakened to a tropical storm as it moved further inland of the Florida Peninsula before exiting into the Atlantic. + + + + + + The damage from Hurricane Ian was particularly catastrophic on Pine Island and Sanibel Island just off the coast of Cape Coral. The scrollytelling cards shows images from the Associated Press encompassing Fort Myers Beach and Sanibel Island and illustrating the destruction Hurricane Ian left in its wake. Most homes along this portion of Fort Myers beach were moved due to the storm surge and were either destroyed or severely damaged. A section of the Sanibel Causeway (the causeway that connects Sanibel Island to the mainland of Florida) was taken out by Hurricane Ian leaving residents stranded on Sanibel Island. The damage to houses, roads, and other structures was catastrophic, and the hurricane also inflicted severe harm on the area’s vegetation. + + +
+ + + Figure 1: National Hurricane Center analysis of the storm surge from Hurricane Ian. Black line indicates Ian’s path. Red/magenta colors indicate areas of highest storm surge. Credit: National Hurricane Center. + +
+ + + +
+ + + Figure 2 and 3: HLS NDVI pre/post-Ian over Sanible Island, Cape Coral, and portions of Fort Myers. + +
+ + ## NDVI & Entropy Analysis + + Normalized Difference Vegetation Index (NDVI) serves as a fundamental metric for understanding the vitality and abundance of vegetation within a specific geographical region. In this data story, we compare NDVI data of the area derived from HLS data for two different days pre- and post-hurricane. The primary vegetation in this portion of Florida includes grasslands and forested regions. NDVI is a key metric used to assess the health and density of vegetation in a specific area based on satellite imagery. Figure 2 (left portion of the NDVI slider) depicts the NDVI around Cape Coral a month prior to Hurricane Ian, while Figure 3 (right portion of the NDVI slider) shows the same landscape the week following the storm. In Figure 2, darker green colors indicate a relatively high abundance of healthy vegetation at a particular location. The light green, yellow, and red colors indicate areas with little to no healthy vegetation. + + Pine and Sanibel Islands, east of Cape Coral, are particularly notable because the hurricane inflicted so much damage. The post-Ian image (Figure 3) shows a significant change in the vegetation landscape. Storm surge inundated much of the access points to the islands and washed away or damaged much of the vegetation in these locations. In addition to statistical analyses, the notable transformation of vegetation in these dispersed areas holds significant scientific importance, as it provides compelling evidence of the formidable impacts of this catastrophic storm. Hurricanes can have a massive and varied impact on nature, demonstrating why it is crucial to understand their wider environmental effects. + + + + + +
+ + + Figure 4: HLS NDVI Difference for Sanibel Island, Cape Coral, and portions of Fort Myers. Red colors indicate an increase in vegetation and blue indicates a decrease in vegetation. + +
+ + Figure 4 shows a significant change in the post-hurricane vegetation landscape by tracking the difference between pre- and post-Ian NDVI. The blue colors show a negative change in NDVI in the difference image, suggesting there was a substantial loss in vegetation after Ian. Ian caused widespread devastation in coastal areas, with many places experiencing severe damage or complete flooding. The storm surge washed away much of the natural vegetation, severely impacting local ecosystems and communities. This highlights the broader issue of the destructive power of hurricanes on natural environments near and along the coast, underscoring the need for preparedness and resilience strategies. + + + + + + + Entropy, in the context of remote sensing and image analysis, is a measure of the randomness or disorder in an image. More specifically, we can see how surface textures such as vegetation and buildings changed after the hurricane. A high entropy value suggests a higher level of complexity and texture, while a lower value indicates uniformity and smoothness. By analyzing the entropy change associated with Hurricane Ian using the same HLS data as was used in the NDVI examination above, we analyze alterations in surface patterns and characteristics. + + Depicted in Figure 5, the observed increase in entropy in certain coastal regions post-hurricane suggests that the once smooth and uniform beach areas became more complex in texture. This complexity could be attributed to the catastrophic storm surge that pushed coastal sand further inland. This not only altered the coastal terrain but also created a lot of land surface changes further inland, where homes and businesses that were once present are now either destroyed or heavily damaged. The amount of sand and debris scattered throughout the area of interest can be seen within our entropy change image from Hurricane Ian. These differences are highlighted in Figure 8, where areas in light to dark red represent locations where the highest change occurred. + + +
+ + + Figure 5. HLS entropy difference for Sanibel Island, Cape Coral, and portions of Fort Myers. The red indicates areas with a significant increase in complexity and texture following the hurricane. + +
+ + + + + ## Damage Probability Predictions + + The UCONN GERS lab’s CONUS Disturbance Watcher offers a pioneering approach to near real-time land disturbance detection utilizing the HLS dataset. By harnessing the capabilities of the Stochastic Continuous Change Detection (S-CCD), the dataset establishes a reference image from sustained observation of the land surface and is triggered after a natural disaster to identify land surface disturbance features in post-event satellite imagery. S-CCD is a remote sensing technique used to monitor and detect ongoing changes in land surface characteristics. It highlights areas of disturbances or alteration by comparing new images against a reference baseline. Once these features are defined, the system predicts damage probability by comparing a post-disaster image against a reference image for a given location. In the context of Hurricane Ian, this damage probability dataset serves as a valuable complement to evaluating vegetation and coastline changes associated with wind and storm surge. + + While our primary assessments, including entropy changes, provide direct insights into Ian’s impact, the CONUS Disturbance Watcher augments our understanding by delivering probabilistic insights into areas most likely affected by disturbances. The damage probability index primarily highlights changes in forested regions. To provide a more comprehensive analysis, we have combined vegetation and entropy indices in this study to encompass both the impact on vegetation and changes to human-influenced land cover. By integrating these data and combining direct observation with predictive analytics to determine areas of potential vulnerability and change, we gain a more comprehensive view of Ian’s aftermath. + + + + + +
+ + + Figure 6: Damage Probability from UCONN GERS lab for Lee and Charlotte counties in Florida. This product is on a scale of 0 to 99 with 0 indicating 0 probability of damage and 99 indicating the highest probability of damage. + +
+ + + + + ## Damage on the Sanibel Causeway + + The devastating impact of Hurricane Ian resulted in the collapse of the Sanibel Causeway, a critical link connecting Sanibel Island to Florida's mainland. This destruction stranded numerous residents, necessitating boat-based evacuations and delivery of vital supplies and underscoring the hurricane's profound effects on offshore communities. The area around Punta Rassa, where the causeway commences, suffered extensive damage, further illustrating the severe consequences of Hurricane Ian on these coastal regions. + +
+ + + Damage to the Sanibel Causeway. Credit to the Wilfredo Lee - Associated Press. + +
+ + + ## Fort Myers Beach + + On the mainland Florida coast, high storm surge led to major inundation of Fort Myers Beach. Most of this area is developed; however, wetlands several miles inland experienced major vegetation damage from the water. Homes and businesses were mostly damaged or destroyed in this region. + +
+ + + Damage to a portion of Fort Myers Beach. Credit to Kevin Fogarty - REUTERS. + +
+ + + ## Sanibel Island + + Overall, the most significant change in vegetation as a result of Hurricane Ian occurred on Sanibel Island. The highest winds and energy occurred in this area, resulting in some of the highest storm surge and inundation from the other locations mentioned. + +
+ + + Damage to a portion of Sanibel Island. Credit to Wilfredo Lee - Associated Press. + +
+ + + + + + ## Conclusion + + Hurricane Ian left a trail of devastation across southwestern Florida, highlighting the urgent need for enhanced disaster preparedness and resilience strategies. The catastrophic impact on life, property, and the environment, evidenced through detailed NDVI and entropy analyses using HLS data, underscores the challenges communities face in recovery. The use of advanced remote sensing tools like the CONUS Disturbance Watcher emphasizes the importance of real-time data in disaster response. + + + + + + + ## Supplemental Resources + + [National Hurricane Center (NHC)](https://www.nhc.noaa.gov/) + + [Hurricane Ian Overview](https://www.nhc.noaa.gov/data/tcr/AL092022_Ian.pdf) + + [University of Connecticut - Global Environmental Remote Sensing Laboratory (GERs Lab)](https://gerslab.uconn.edu/) + + Author of Cover Photo - [Joshua Stevens](https://earthobservatory.nasa.gov/about/joshua-stevens) + + NESDIS - [National Environmental Satellite, Data, and Information Service](https://www.nesdis.noaa.gov/) (NESDIS) + + + + diff --git a/stories/ian_goes_cover.jpg b/stories/ian_goes_cover.jpg new file mode 100644 index 000000000..c486149b6 Binary files /dev/null and b/stories/ian_goes_cover.jpg differ diff --git a/stories/ndvi_after.jpg b/stories/ndvi_after.jpg new file mode 100644 index 000000000..468ffab35 Binary files /dev/null and b/stories/ndvi_after.jpg differ diff --git a/stories/ndvi_before.jpg b/stories/ndvi_before.jpg new file mode 100644 index 000000000..0cdd667f3 Binary files /dev/null and b/stories/ndvi_before.jpg differ diff --git a/stories/ndvi_difference.jpg b/stories/ndvi_difference.jpg new file mode 100644 index 000000000..bb81fc720 Binary files /dev/null and b/stories/ndvi_difference.jpg differ