Progress 07/01/19 to 02/29/20

Outputs
Target Audience:Azavea's research centers on developing new data-driven tools that will support greater resilience to climate variability in rural, tribal, and agricultural communities. Our initial target audience is the many rural and tribal communities in the State of California.By building on Azavea's previous climate research outcomes under a grant from the U.S. Department of Energy and our partnership with ICLEI-USA, a standard-bearer in the field of community sustainability, this new research will result in capabilities that will enable local decision-makers to combine key datasets about their community's facets with fine-resolution climate projection data in simple yet powerful ways that will support informed climate resilience planning going forward. Changes/Problems:Since beginning our project, we have had some minortechnical issues scaling the initial prototype of the custom area of interest feature.Building on our research and prototyping efforts completed prior to this Phase I project, we believed that the most effective path forward to developing this feature was to rewrite the backend data processing components to take advantage of GeoTrellis Layers + HTTP ranged reads on S3. During development, we discovered that the GeoTrellis Layer data structure would enable us rapidly read and compute data from spatial dimensions but would not enable us to efficiently read large chunks in a temporal dimension, a fundamental requirement for creating a performant and responsive web application. In addition, the ingest process for the down-sampled LOCA dataset was too slow due to the formatting of the source NetCDF climate data. We attempted two optimizations of to the ingest process: (1) attempt to read the NetCDF climate data with the GDAL library and (2) mount the NetCDF climate data as a FUSE filesystem. Neither provided the order of magnitude performance gain necessary to proceed within the Phase I schedule. Even with these issues, we were able to ingest the LOCA data for a temporal subset for all models, atmospheric carbon scenarios, and variables using the initially planned approach. This has enabled the development of a working Area Indicators API for the time period ingested and validation of its results against the current Climate API for a number of indicators.After some testing using the data subset in the Area Indicators API to query for indicators in several remote areas of California and comparing that to the results currently returned by the Climate API, we determined that while the Area Indicators API results are somewhat more accurate, the point-based Climate API results are sufficiently similar that the existing data computation infrastructure can be used to complete the intended work. What opportunities for training and professional development has the project provided?Due to staff changes at ICLEI-USA Local Governments for Sustainability, Jesse Carpentier, Community Resilience Program Officer, has acted as our advisory group member.We brought Ms. Carpentier up to speed on both Temperate in general and the Rural Temperate project in particular. Given Ms. Carpentier's expertise in adaptation planning, the transition has been seamless and beneficial. How have the results been disseminated to communities of interest?User testing thus far has been limited to the Advisory Board, which has been convened twice thus far--once for a kick-off meeting and again to review the wireframes for the new maps recently added. Azavea staff are incorporating the Advisory Board's feedback presently, and plan to start conducting very limited user testing with other potential users during the next reporting period. What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period, we will continue to process and integrate new datasets into Temperate, complete the custom geography tool, improve the top indicators list, update the user interface, and conduct iterative user testing.

Impacts
What was accomplished under these goals? To date, Azavea has made considerable strides in understanding which datasets are most critical to rural and tribal communities in California. This effort has been significantly supported by our subconsultant, Prosper Sustainably, a California-based nonprofit. Their research has outlined more than forty datasets, representing both climate hazards and potential impacts, that would enhance Temperate's value to future users.Several of those high value datasets are best suited to map visualizations, and to date, we have created five such map layers.

Publications

Progress 07/01/19 to 02/29/20

Outputs
Target Audience:Azavea's research has centered on developing new data-driven tools that will support greater resilience to climate variability in rural, tribal, and agricultural communities, with initial focus on those in the State of California.The Phase I prototype provides rural decision-makers with straightforward, step-by-step guidance on identifying the hazards, vulnerabilities, and risks specific to their communities and ultimately crafting and implementing appropriate action strategies. In so doing, it addresses a critical need for readily accessible modeling tools that can help rural decision-makers with limited resources and institutional capacity to proactively manage disaster risk, ensure access to safe water and food, and strengthen preparedness and response strategies that will benefit everyone. Changes/Problems:Azavea engaged partners to provide subject matter expertise regarding which additional datasets would provide the most value in our pilot location of California when coupled with the rigorous temperature and precipitation data already in Temperate. While we had initially intended to incorporate physical asset information into the map interface, including the locations of hospitals, nursing homes, and other medical facilities, Prosper Sustainably recommended that we defer that task in favor of adding more climate exposure and impact data sources, which they felt would be the most useful to rural and tribal communities struggling with adaptation planning. However, we soon learned thatthere are anumber of climate exposure datasets that are either only provided in non-machine readable formats, such as PDF, or have licenses that prohibit them from being used in a commercial product. Due to these limitations, and in order to make significant Phase I progress, we focused instead on adding the following datasets to Temperate that we felt would provide the greatest value to future users: Centers for Disease Control and Prevention's National Environmental Public Health Tracking California Office of Environmental Health Hazard Assessment's CalEnviroScreen United States Environmental Protection Agency's Climate Change and Risk Analysis (2.0) United States Forest Service's Wildfire Hazard Potential During the Phase I project, we were also able to leverage additional information from these datasets, including agricultural yields for nine crop types, the potential economic damage of inland flooding, and the number of people within a FEMA-designated flood hazard area. Finally, we ingested the Robert Wood Johnson Health Foundations' County Health Rankings into the Temperate application for testing. In Phases II, we plan to integrate additional datasets that will enable us to scale the prototype from California to nationwide coverage. We currently anticipate that the available data may vary across geographies, but we are confident that we can provide rural and tribal communities with informative data that will assist them with the difficult local decisions brought about by climate volatility. What opportunities for training and professional development has the project provided?Through our California-based environmental subconsultant, Prosper Sustainably, Azavea wasable to better understand and assess the suitability of incorporatingdatasetsrepresenting both climate hazards and potential impacts into the Temperate solution: first, for the State of California, and in Phase II, for the rest of the United States. We also gained a greater understanding of themany available resources documenting the impacts of specific climate risks on different aspects of daily life. An important component of the Phase Iresearch aimed to identify those broadly applicable impacts most relevant to rural areas and incorporate them into the vulnerability assessment component of Temperate. We believe the knowledge we have gained through the course of Phase I has enabled us to build a prototype software solution that will help California decision-makers to better prioritize potential impacts most important for their rural and tribal communities, while providing us with a strong foundation to scale the application for nationwide use in Phase II. How have the results been disseminated to communities of interest?Throughout the project, we met with our Advisory Board and individuals who work to improve climate resilience in rural and hard-to-reach communities. We regularly reviewed wireframes and planned features with a variety of stakeholders and conducted iterative user testing as we worked to confirm that the planned designs were effective for various user types. User testing was also conducted with three San Diego area tribal communities, in addition to three individuals from non-urban, East Coast areas(Centre County, PA and New Paltz, NY).Overall, our user testing revealed that the inclusion of visual, interactive map layers dramatically increased the accessibility of the data, since people working in this space are very comfortable looking at maps and drawing conclusions from them. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Azavea achieved substantial success toward achieving our goals. Working with our California-based nonprofit subconsultant, Prosper Sustainably, we were able to identify several California departments and organizations produce and publish a great deal of data that is useful to tribal and rural locations. In addition, we were able to successfully implement a custom geographic search feature that enables users to work with very large tribal and rural areas of interest (up to the size of a small state). Using these new data sources and computation capabilities, we were also able to develop a revised user experience and test this rural- and tribal-adapted system with users from these communities. Many of our conversations with users and potential users have reaffirmed the need for a decision support tool like Temperate as a key means of aggregating information from key datasets that might be consulted when conducting a vulnerability assessment without the assistance of a paid consultant. These individuals have indicated that theywant more data, preferably on a map or in other graphical form, and that they need it to be presented in a clear and flexible manner, whether that means downloadable for inclusion in other documents or filterable by other data parameters.Thanks to the dataset research completed by our subconsultant, and the relationships we have built with both service providers and community users in Phase I, we are confident that we can build out our prototype in Phase II into a useful commercial product that serves rural and tribal communities across the United States.

Publications