Browsing by Subject "Google Earth Engine"

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Automating Offshore Infrastructure & Vessel Identifications Using Synthetic Aperture Radar & Distributive Geoprocessing
    (2018-04-27) Wong, Brian
    Global Fishing Watch (GFW) recently published the first worldwide industrial fishing effort data set learned from processing 22 billion Automatic Identification System (AIS) observations. Despite quantifying 40 million hours of fishing activity that extended to over 55% of the ocean’s surface area in 2016, GFW now aims to quantify fishing effort not captured by current analyses through multimodal remotely-sensed imagery. Such imagery-based vessel identifications are commonly confounded with offshore infrastructure, though, so a global offshore infrastructure data set is first required to disentangle the two. This study first establishes robust and scalable methods for automating offshore infrastructure identifications using synthetic aperture radar in the Gulf of Mexico, and then evaluates the feasibility to adopt these methods for vessel identifications. Results indicate our model identifies offshore infrastructure with a probability of detection of 96.3%, an overall accuracy of 91.9%, a commission error rate of 4.7%, and an omission error rate 3.7%. Additionally, a cloud-native geoprocessing framework using the Google Earth Engine Python API was implemented to automate vessel identifications globally. Over 45,000 SAR images or approximately 100TB of data were processed to build a new database overlaying both SAR-derived and AIS-derived vessel locations.
  • Thumbnail Image
    ItemOpen Access
    Quantifying Land Cover Change to Inform Carbon Offset Projects in Madagascar
    (2022-04-19) Golden, Israel
    Duke University has committed to becoming carbon neutral by the year 2024. This commitment will be met through a combination of local emissions reductions and global carbon offset projects. In support of this effort, the Duke Carbon Offsets Initiative (DCOI) and the Duke Lemur Center (DLC) have teamed up to identify potential carbon offset project sites in the SAVA region of Madagascar. The SAVA region is home to globally significant biodiversity, including twelve species of lemur and many other rare, endemic species. Unfortunately, many of these species are threatened with extirpation from habitat loss. Intensified shifting agriculture and unsustainable forestry practices have reduced primary humid forest habitat by at least 48% in the SAVA region since 2002. DCOI and DLC have identified four potential project sites on degraded former agricultural land for Afforestation, Reforestation, and Revegetation (ARR) carbon offset projects. ARR offset projects sequester atmospheric carbon by restoring ecosystem function through forestland restoration. Successful restoration of these project sites could assist Duke University in its climate goals, restore habitat for Madagascar’s unique biota, and protect ecosystem services for local Malagasy communities. Carbon offset projects must satisfy the requirement of sequestering additional carbon to receive verified carbon credits. The additionality of a project is assessed through evidence- based, counterfactual logic outlined in Verified Carbon Standard (VCS) methodology. In this Masters Project, we investigate one aspect of additionality for the four proposed carbon offset project sites with an analysis of past land cover trends and estimated aboveground carbon flux at each project site. Land cover trends were assessed with a classification and regression tree (CART) model trained on Landsat 8 imagery and validated with ground control points collected by collaborators in the SAVA region in 2021. This model was then used to classify representative images for each year to reveal changes in forest, grassland, marshland, water, and built up land cover extent since 2013. Aboveground carbon flux was then estimated based on carbon-by-area coefficients for each land cover derived from Alcorn et al. (2021). Based on this analysis, forested land cover has either remained stable or slightly declined over the study period at each of the proposed project sites. As expected, estimated carbon fluxes mirror land cover trends on each site. Additionally, there has been little to no natural forest regeneration on any of the proposed project sites since 2013. This outcome suggests that funding an ARR offset project would likely support the sequestration of additional atmospheric carbon. However, the remaining social and economic analyses required by VCS must be completed before moving forward with the proposed offset project. Finally, we recommend field-based biomass surveys of each project site to produce fine-scale estimates of aboveground carbon for accurate carbon accounting.