June 2020 Summaries

The Spatial Data Science Conference will be held as a free virtual conference between October 19th and 23rd, featuring industry leading speakers delivering on demand content on topics such as spatial modelling in the enterprise, location analytics for city shaping, and social impact of spatial analysis. The conference aims to bring together experts from various sectors including academia, government, and private companies, with a focus on the future of Spatial Data Science in the context of the "new normal". Unlike previous conferences, SDSC20 will be an interactive and on demand event, allowing attendees to access content at their convenience.

The COVID-19 pandemic has accelerated the need for businesses, especially in the retail sector, to adapt their strategies and identify high-performing areas in a "new normal" environment. With many malls expected to close in the USA and the UK potentially being over-spaced for retail, businesses are focusing on consolidation and optimizing their locations. A key approach is using comparative market analysis real estate to determine which stores are best suited for curbside pickup and effective location strategies. This involves analyzing attributes such as demographics, socioeconomic profiles, and proximity to points of interest to find a common path that defines an optimal location, known as a "twin area". Identifying twin areas can help businesses perform site selection, relocation, and consolidation by providing a full market analysis, reviewing competitors, and loading sales data alongside multiple spatial datasets. By leveraging this approach, businesses can make informed decisions about their consolidation strategy and optimize their locations for success in the post-pandemic retail landscape.

The COVID-19 pandemic has had a significant impact on the American meat industry, with meatpacking plants becoming hotspots for infection due to low temperatures, cramped conditions, and long hours. Geospatial analysis using tools such as CARTOframes, Xarray-spatial, and Datashader has been used to assess the relationship between distance to meatpacking plants and confirmed COVID-19 cases. The analysis found a weak negative correlation between the two variables, indicating that areas farther away from meatpacking plants tend to have lower COVID-19 case rates. This research highlights the potential of geospatial analysis to inform public policy and awareness around COVID-19.

The COVID-19 pandemic has led to widespread use of public data for spatial analysis, with various providers offering datasets on human mobility metrics, case numbers, and other relevant information. To use these datasets, researchers can explore the Data Observatory, a platform providing access to publicly available datasets, and set up their preferred tool, such as CARTOframes, which offers features like data discovery, filtering, and visualization. By processing the data, aggregating it to reduce noise, and enriching it with additional information, researchers can gain insights into spatial patterns, demographics, and socioeconomic factors influencing COVID-19-related trends. The analysis can be further advanced by exploring other datasets available in the Data Observatory, watching webinars or requesting demos from experts, and continuing learning through the platform's Help Center.

SEUR, a leading parcel delivery company in Spain, collaborated with CARTO to optimize their cold transportation network. The project aimed to assess the current state of their network, quantify the impact of changes, and build an optimization model to identify optimal delivery areas for each distribution center. Spatial Data Science techniques were applied to analyze high-density areas, select spatial aggregations, and calculate the impact of opening or closing distribution centers. The results showed significant savings in distance traveled, with an average reduction of 18.23 km/order compared to the original 18.99 km/order. Future steps will focus on adding more business constraints and requirements to the analyses and models, enriching the data, and including other costs to better understand the impact of decisions.

The COVID-19 pandemic has had a significant impact on Brazil, with the country approaching one million confirmed cases and expected to surpass the US in terms of cases and deaths by the end of July. A local geospatial community in São Paulo is using data visualization to spread awareness and inform public policy, while a Brazilian company Mapfry is developing a geomarketing solution to aid in economic recovery. Researchers are also working on estimating hospital bed demand, mapping propagation risk, and protecting indigenous populations, highlighting the need for continued support and action to combat the pandemic in these vulnerable groups.

The latest research by Howsy, a UK lettings management platform, analyzed the cost of renting properties within walking distance of London's City of London office location. The study found that 20 locations across London showed an average rent of £2,001, which is 18% higher than the current London average. Mayfair emerged as the most expensive walk-to-work hotspot at £2,786, while Mile End was the most affordable option within a one-hour walk of the city, with an average rent of just £1,494. According to Howsy's founder and CEO Callum Brannan, this trend may indicate a shift in demand for rental properties within walking distance of major work hubs as Londoners opt for a more appealing commute option due to concerns about public transport during the pandemic. This trend could become a key selling point for landlords when listing their properties, potentially replacing the appeal of nearby transport hubs.

The COVID Racial Data Tracker has gathered comprehensive race and ethnicity data on COVID-19 cases and deaths in the United States, highlighting significant disparities between Black Americans and the rest of the US population. These disparities are reflected in mortality risk calculations, which show that Black Americans face a 1.5 times higher mortality rate than the general population, with varying levels of risk across different states. The data also reveals that despite being predominantly White, some counties with high COVID-19 death rates are predominantly Black, suggesting underlying factors contributing to these disparities.

WorldPop in CARTO provides global demographic insights at high granularity, enabling various geospatial analysis applications such as human exposure to hazards and climate change impact on ecosystems. The dataset is derived from census data, fine-resolution estimates, and remotely-sensed geographical covariates. Two main methods are used: dasymetric mapping and a convolutional neural network-based approach for identifying settlements using high-resolution satellite imagery. The WorldPop dataset offers coarser resolution estimates compared to the High Resolution Settlement Layer (HRSL) but provides more detailed information on variations within settled areas. It is available globally for every year between 2000 and 2020, both at a 1 km by 1 km grid and a 100 m by 100 m grid, with additional age and gender structures data.

US Protest Map & Visualizing Police Violence` The US has seen a wave of unrest and protests in cities across the country following the tragic death of African-American George Floyd in police custody, sparking a need to analyze and visualize data on police violence. Mapping Police Violence has collected comprehensive data on police killings nationwide, including 1,099 deaths in 2019, with black people accounting for 24% of those killed despite being only 13% of the population. The collaborative's data is sourced from various crowdsourced databases and original research to improve quality and completeness. An animated map visualizes this data, showing killings that took place in 2019, as well as social media response to the protests, including hashtags trending globally such as #BlackLivesMatter. Protests have spread to 140 cities across the US, with the National Guard deployed in at least 21 states, prompting efforts by organizations like CARTO to support visualizing data and promoting racial equality through grants programs.