Nicolas Saravia

Education: Master of Geography (University of Bristol)  

The following profile was compiled by Brendan Vander Weil (Texas State University) for the Encoding Geography initiative. To learn more, visit: https://www.ncrge.org/encoding-geography/

 


Please describe your job, employer, and the primary tasks you perform in your position.  

I currently analyze transportation businesses and their innovations. On the side, I help micro-mobility start-ups by advising their leadership on how to improve their businesses in areas such as capital raising, operations, and data science. Most of my career has been at the intersection of operations, data analysis and business intelligence. I’ve also written tech patents to solve physical infrastructure issues with IoT and machine learning. 

How has your education/background in geography prepared you for this position? 

In my opinion, the ability to think spatially is a geographer’s greatest strength — the world is full of challenges that need 3D thinking to solve them efficiently. Through the course of my career, my strategy has shifted from viewing geocomputation tools as means on their own, to a more auxiliary, albeit important role. In many real-life business scenarios, one can solve spatial problems without geographical methods; however, in my case, geocomputation tools such as GIS, spatial statistics, and web mapping have certainly enabled me to find the needle in the haystack faster than otherwise and in a way that is visually compelling and factual.  

What is an example of applying geography concepts and skills in order to analyze and solve problems in your work? 

In businesses where you have physical assets, there is a real need to analyze the human and physical factors that affect the management of these across time and space. Human geography variables like population density, traffic patterns, and infrastructure may affect demand depending on the type of business. In my career, I’ve also focused on measuring the impact of physical geography and meteorological variables, everything from elevation to distinct weather variables and natural disasters.  

What types of geographic questions did you ask and think about in your project? 

The main questions I asked seek to answer how the different variables affect demand and supply in a business’s geography, which ultimately may impact the bottom line. I also look at how to optimize operations based on the analysis of the human and physical factors that affect the area. Additionally, I ask which methods should we use to predict external factors, and how do we balance speed and quality in results; how do we automate certain repetitive tasks without impacting costs; and what software/tools do we use to handle latency based on the amount of data we are processing?  

What types of data did you acquire to support your project?  

Being able to find reliable and open data is 90% of the battle in many startup jobs. In my case, I have heavily relied on free NOAA data for different weather variables and OpenStreetMap for infrastructure data (QGIS has a great plug in for that!). In larger companies and in consulting, there is sometimes the option to purchase big data, such as traffic flows. Familiarizing oneself with the nuances and quality issues of a dataset and being able to process that data with automation to remove any noise, will generally set up things for better decision making.  

What types of content knowledge and skills (both geographic and more general) did you use to evaluate, process, and analyze the data you gathered for your project? 

The foundational skill that is most practical in operations and data analysis is SQL.  Start-ups generally rely on open-source software and tools to get the job done without impacting the team’s budget allocation. QGIS has been especially useful throughout my career, both for visualizing data and in running algorithms like k-nearest neighbor or performing spatial randomness experiments. Having had a prior understanding of the statistical methods that the tools run helps me understand what they are visualizing. This is significantly more important than knowing where the tools are located (which is more readily searchable).  I generally have used the R language for statistical analysis of geographic data, and Python to automate repetitive tasks. Some knowledge of JavaScript has been useful, especially when visualizing results on a map platform like Leaflet. General business intelligence programs like Looker, PowerBI, or Tableau (this one has Leaflet plugins) are also good to have in the toolbox, especially when delivering results to executives. Having these listed on a resumé can open doors.    

How did you communicate the results of your project (e.g., writing technical reports, making maps and geo-visualizations, creating graphics, data tables, etc.)? Do you have a recent product or publication to share with us as an example?  

Throughout my career I’ve had to write standard operating procedures, technical whitepapers, and websites; however, much more frequently I’ve had to summarize large amounts of information in concise emails and bullet points, with a quick chart or map. I’ve found that results are most effectively communicated when they are direct, with clean and clear visualizations.   

What are the criteria that you use to assess the quality of your results?  

At a dataset level, keeping in mind sample size and implementing proper data cleaning and further investigating any observation errors. A/B testing is a great way to evaluate insights and decisions. It is always important to review results once they are in and do a proper post-mortem digging of what has changed in the data and measure the adjustments. 


This material is based upon work supported by the National Science Foundation under Grants No. 2031418, 2031407, and 2031380 (Collaborative Research: Encoding Geography – Scaling up an RPP to achieve inclusive geocomputational education). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation 

    Share

Tracy Whelen

Education: M.S. in Geography (University of South Carolina), B.A. in Geography (Mount Holyoke College)  

The following profile was compiled by Brendan Vander Weil (Texas State University) for the Encoding Geography initiative. To learn more, visit: https://www.ncrge.org/encoding-geography/ 

 


Please describe your job, employer, and the primary tasks you perform in your position. 

I am a Geospatial Consultant and participant in the Business Insights & Analytics Leadership Development Program at Travelers Insurance. Travelers is a leading property and casualty insurance company, offering a wide range of personal and business insurance products primarily in the United States and Canada. 

I recently completed an enterprise rotation in Enterprise Data & Analytics, working on data management and quality assessment of enterprise geospatial datasets and ad-hoc geospatial business consulting requests.   

I am currently in a rotation for Claim Business Intelligence & Analytics. My work includes geospatial information delivery and analysis for Claim senior leadership and field offices. Part catastrophe response, part improving everyday claim handling processes.  

Prior to joining Travelers, I received bachelor’s and master’s degrees in geography (Mount Holyoke College and University of South Carolina, respectively). In between my degrees I worked as a GIS Specialist in a remote sensing lab at University of Massachusetts, Amherst.  

How has your education/background in geography prepared you for this position? 

All the data I work with has a spatial component to it, and we often work with thousands (sometimes millions) of records at a time, necessitating strong geography and computer science skills to efficiently store, process, and analyze data, and to deliver actionable outputs.  

Relevant courses from my education that I use today in my job include: 

Geography 

  • GIS/spatial analysis (intro and advanced)  
  • Remote sensing 
  • Spatial modeling  
  • Web GIS  
  • Basic human and physical geography 
  • Electives: Meteorology, hazards geography, business geography 

Computer Science 

  • Introductory scripting (if statements, loops, functions, etc.) 
  • Python 
  • SQL 
  • Data structures 
  • UI/UX design 
  • Javascript (web app development)  

Math 

  • Discrete math (basic logic and set theory)  
  • Statistics (non-spatial and spatial) 

What geographic skills and information do you use most often in your work?  

Geographic concepts that I use in my daily work are important for things such as asking what business problems have a spatial component to them or analyzing the spatial relationship between two or more datasets (e.g. spatial joins and other geospatial analysis). I also need to understand a wide variety of spatial data formats, how to convert between them, and what formats are most appropriate for a given use case (e.g. basic raster and vector formats, enterprise SQL databases, APIs, published feature services, etc.). Finally, I need to know when to use geographic coordinates versus a projection (and what an appropriate projection might be).  

What is an example of applying geography concepts and skills in order to analyze and solve problems in your work? 

One of the many risks Travelers seeks to mitigate are natural hazard events, such as wildfires and hurricanes that climate change may make more extreme. Sustainability at Travelers means performing today, transforming for tomorrow and fulfilling our promise to our customers, communities and employees. Where these two come together is how our Claims department responds to natural hazard events, especially large wildfires or damaging wind events. The following videos capture the spirit of what we do, and the geospatial component of Claims catastrophe response. 

 

Note that Travelers is organized along an Agile structure, with cross-functional teams continuously delivering improvements. While there are always new products and applications being developed, there are also lots of long-term operational systems being continuously used and improved upon. Often employees build on past work and may not see a large project or system from beginning to end. My team’s catastrophe response work is an example of this type of long-term system, and my answers are on behalf of the team.  

What types of geographic questions did you ask and think about in your project? 

The broad business question underlying this issue is, “How can we optimally respond to catastrophe events, meeting customer needs with the most efficient use of business resources?” Underlying questions include: 

  • What location has been/will be impacted? 
  • What is our exposure in the area? (i.e. number of policies, associated financial exposure) 
  • Where have claims already been reported? 
  • How many claims might we expect? 
  • What types of claims do we expect to see from this event? (e.g., wind, water, fire, etc.) 
  • What types of damage occurred, and how severe is the damage? 
    • Will this impact our ability to respond, either because an area is inaccessible or because local offices or employee homes have been damaged? 
  • Where can we acquire the necessary data from? 
  • Can we develop models to more efficiently review post-event imagery as part of the catastrophe response process? 
    • If so, what features are we trying to spot in the imagery?  
    • How does this vary by event type? 
    • What might be appropriate modeling algorithms to use? 
    • What are some of the challenges the model might encounter? 

My team does not directly answer all these questions, but we need to be able to provide appropriate data to the senior leadership and other decision makers or support staff who can build a final answer.  

What types of data did you acquire to support your project? 

  • Business data (e.g. claims, policies) 
  • Event data – wildfire boundaries, hurricane wind footprints, precipitation measurements, tornado damage reports, etc. 
  • Aerial imagery and derived model output 
  • Property geometry data (e.g. building footprints, parcel boundaries)

What types of content knowledge and skills (both geographic and more general) did you use to evaluate, process, and analyze the data you gathered for your project? 

In the moment skills that we use on this project for responding to a single catastrophe event include: 

  • Querying databases (spatial and nonspatial joins, filters) 
  • Combining and reformatting a variety of data formats  
  • Running models in python scripts 
  • Common sense/data quality checks 

For long-term projects, the output of which gets used in catastrophe response (multiple team effort), the skills we use are: 

  • Internal model development in partnership with data scientists  
    • Curate input data (image locations, image clipping geometry, training data, etc.) using SQL and python 
    • Evaluating model results against other sources of truth 
  • Evaluating new 3rd party datasets (accuracy, timeliness, availability, cost, other potential sources for the same information)

How did you communicate the results of your project (e.g., writing technical reports, making maps and geo-visualizations, creating graphics, data tables, etc.)? Do you have a recent product or publication to share with us as an example? 

We publish web GIS content as both data layers and maps, creating different versions for different user groups in order to control access to sensitive information. We also use frequent email communication, whether it is one on one, small group with specific questions and answers, or larger list-serv communications (with standardized templates) at key time points during catastrophe response (e.g. web map published, imagery collected, etc.). Additionally, we communicate results through spatial SQL data pulls (tabular format)  

See the below list for examples of broader enterprise or external communications about projects and programs mentioned in this interview. 

  • Travelers 2021 Q3 Earnings call. Note CEO Alan Schnitzer’s introductory remarks including, “location intelligence at the parcel level” and our “AI Assisted Claim Damage Detection Model was a key part of our Ida claim response” 
  • Interview of Adam Sobek (Travelers AVP of Geospatial) at NearMap Navig8 Conference 2020 (Travelers’ use of imagery, including for catastrophe events) 

What are the criteria that you use to assess the quality of your results?   

Most important criterium: Has the business need been met?  

Other important criteria: 

  • Validate data quality 
  • Spatial scale and level of accuracy  
  • Minimizing false negatives, minimizing false positives.  
  • Minimizing process (time, number of steps)

The business need at hand dictates which criteria are important, which varies from question to question. Examples include: 

  • Level of address accuracy needed to plot individual policies versus summarize at a zip code level 
  • Some analytics results are only valuable if they can be completed faster than more manual processes out in the field. 

This material is based upon work supported by the National Science Foundation under Grants No. 2031418, 2031407, and 2031380 (Collaborative Research: Encoding Geography – Scaling up an RPP to achieve inclusive geocomputational education). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation 

    Share

Managing Information Workflows for Writing, Research, and Data

Geospatial Data Science for Social Media Analytics

Bayesian Spatial and Spatiotemporal Modeling Using R

Novel Topics in Geoinformatics: Drones, EO, Space-Time, Deep Learning

Developing Open Source SpatiaLite Databases in QGIS

Communicating Geographic Research Through Interactive Web-based Dashboards and Data Storytelling

Integrative Geospatial Institute Secures $15 million NSF Grant

Overview of I-GUIDE’s six interrelated areas of focus.

The National Science Foundation has established the new Institute for Geospatial Understanding through an Integrative Discovery Environment (I-GUIDE, one of five centers to meet the goal of harnessing the data revolution. AAG is among the partners supporting the new effort.

Led by University of Illinois Urbana-Champaign with an NSF award of $15 million, I-GUIDE will “create an integrative geospatial discovery environment that harnesses geospatial data to understand interconnected interactions across diverse socioeconomic-environmental systems — with a goal of enhancing community resilience and environmental sustainability.” The institute will generate a new set of analytic tools that carefully address data interdependencies to help better estimate and predict risk and anticipate impacts from disasters or climate change.

“This is unprecedented level of support from the National Science Foundation, a victory lap for geographers,” said Shaowen Wang, head of the Department of Geography and Geographic Information Science at U of I and founding director of the CyberGIS Center for Advanced Digital and Spatial Studies (CyberGIS Center), which will lead the institute and with which AAG is also a partner. “The establishment of this institute is an important recognition of geographers as a leading force in the data revolution.”

Geography and geographic data are critical to understanding and visualizing the many interacting causes and impacts of climate change,” said Gary Langham, Executive Director of the American Association of Geographers. “I-GUIDE provides a new, vital framework which supports research partnerships and accelerates data sharing and analyses. 

I-GUIDE, under Wang’s leadership, will receive the funding over five years to work with a broad range of partners and communities, creating a shared geospatial tool for better understanding the risks and impacts of climate change and disasters.

“The goal is to revolutionize theories, concepts, methods, and tools focused on data-intensive geospatial understanding for driving innovative cyberGIS and cyberinfrastructure capabilities to address the most pressing resilience and sustainability challenges of our world such as biodiversity, food security, and water security,” said Wang.

I-GUIDE will bring together about 40 researchers initially, along with a variety of partners, to transform data practice across many fields from computer, data, and information sciences to atmospheric sciences, ecology, economics, environmental science and engineering, human-environment and geographical sciences, hydrology and water sciences, industrial engineering, sociology, and statistics.

“I-GUIDE nurtures a diverse and inclusive geospatial discovery community across many disciplines by bridging disciplinary digital data divides with broader impacts amplified through a well-trained and diverse workforce and proactive engagement of minority and underrepresented groups,“ said Wang of the Institute’s approach and vision.

Because the challenging issues of sustainability and resilience call for interdisciplinary expertise, I-GUIDE is intended to play a central role in bridging disciplines, creating partnerships and frameworks for collaboration across domains. I-GUIDE will also work with community consortia such as Consortium of Universities for the Advancement of Hydrologic Science (CUAHSI), Open Geospatial Consortium (OGC), and the University Consortium for GIScience (UCGIS) to build consensus and establish priorities for developing a community-oriented and integrative I-GUIDE platform. The institute will also foster open collaboration among diverse communities, bridging the digital divide that hinders participation from underrepresented communities. I-GUIDE will also offer education and training programs, as well as access to cutting-edge geospatial data capabilities (e.g., CyberGIS, GeoEDF, and HydroShare).

I-GUIDE is organized across six interrelated focus areas: (A) Convergence Science Catalysts; (B) Geospatial Artificial Intelligence (AI) & Data Science; (C) Core CI Capabilities & Services; (D) Education & Workforce Development; (E) Engagement & Partnerships; and (F) Evaluation & Knowledge Transfer. I-GUIDE brings together leaders in scientific research, technical development, education, knowledge transfer, and engagement through focus area teams. This strategy ensures that research, education, broadening participation, and knowledge transfer activities are deeply integrated across the institute from individual teams to the leadership.

For more information, visit the I-GUIDE website.

    Share

Stephen Ladochy

Education: Ph.D. in Geography (University of Manitoba), M.S. in Atmospheric Sciences (Colorado State University), B.A. in Meteorology (University of California, Los Angeles)

Describe your job. What are some of the most important tasks or duties for which you are responsible?
Besides teaching classes in physical geography, meteorology and climatology, I mentor graduate students in their research and Masters’ theses.  I also write several letters of recommendation for students seeking jobs, graduate schools and research opportunities.  I continue collaborating with other scientists on climate research and occasionally answer requests from media on environmental stories.

What attracted you to this career path?
I always liked math, and found that it could be applied in meteorology.  At UCLA I interned at the National Weather Service as well as at air pollution consultants. While working at the L.A. County Air Pollution Control District, someone showed me information on “Jobs in Geography”, where you could teach weather courses at universities.  I was hired by the University of Winnipeg in the Great White North teaching weather and later climatology and environmental courses.  I enjoyed teaching, so went on to a Ph.D. in geography/climatology.

How has your education/background in geography prepared you for this position? Most of my education was in Atmospheric Sciences, so I had a lot of prep work to teach geography courses.  I found my niche and passion in meteorology and climatology and have been studying them since.

What geographic skills and information do you use most often in your work? What general skills and information do you use most often?
I like to show satellite images in my classes and the latest climatic data, such as from NASA. So using remote sensing, weather maps and oceanic conditions (being on the coast), I use statistics and recent environmental data in my classes and research.  Mostly, I’m looking at ENSO-Pacific Ocean Indices, weather maps and satellites and climate data to follow climate change.  We also have field instruments so my classes can measure surface weather data in different land uses in urban settings.

Are there any skills or information you need for your work that you did not obtain through your academic training? If so, how/where did you obtain them?
I was fortunate to have summer employment at the Jet Propulsion Laboratory, NASA. There I collaborated with oceanography scientists and student interns on several climate-related projects.  My boss there was a wonderful science communicator, so I learned a lot from him that carried into classrooms and media interviews.

Do you participate in hiring, screening, or training of new employees? If so, what qualities and/or skills do you look for?
I was just on two search committees for new hires. We looked for someone who would be a good instructor with our students, many of whom had English as a second language and were working while in school. We also looked for good mentors for these students, who could relate and encourage high achievement.  Scholarships were also important where they could lead student research

What advice would you give to someone interested in a job like yours?
You need to have a passion for your work and for helping students. That makes the hard work actually fun and something you look forward to doing.  We have special students that work hard and often reach their goals. You need to be a good mentor and inspiration to your students.  Your enthusiasm for your subjects will rub off.

What is the occupational outlook for career opportunities in your field/organization, esp. for geographers?
Our graduates have been fairly successful in finding employment in geography and related environmental fields.  Having skills in computer programming, GIS and remote sensing training or certificates, statistics and the sciences are all helpful.  Internships or summer help in companies or government agencies can often lead to more permanent employment.

    Share