GIS in Action 2026

GPN-PNW Emerging Professionals invites you to a panel discussion with mentors and mentees from our 2024-2025 Mentorship Program cohort. During this session you will hear from current participants about their experiences or projects they are currently working on during this year’s mentorship program. A foundational goal of the Emerging Professionals Mentorship Program is to provide real-world experience in a professional setting with a knowledgeable GIS professional. This is a great opportunity for all those interested in the program not only to learn about interesting projects, but to also hear more about the experience of being a mentor or mentee. At the end of the program outline, we will hold a Q&A session to answer any questions that arise. Please join us to learn more and become part of our network!

Last year, I closed this very conference by letting AI generate my slides dynamically. It crashed spectacularly. We laughed, I cried. Apparently, I haven't learned yet.
This year I'm back! But instead of fixing my approach, I've doubled down. I'll narrate, David Attenborough style, while Claude attempts to build a complete ArcGIS web application. I won't write a single line of code. At least, that's the plan, anyway.
AI scales effort, not judgment. With my hands off the keyboard, I'm freed up to make the real decisions: what to build, when to pivot, and whether the output is even good.
Will it work? Probably (not entirely). Things will break, and we will all learn something. That's the whole point.

We will explore how reliance on traditional blueprints and floorplans has repeatedly led to degraded and inefficient responses during critical incidents, as documented in after-action reports from national tragedies such as Uvalde, Parkland, and others. Using these real-world examples, we will illustrate the core problem facing public safety agencies: the lack of an accurate, accessible, and tactically useful common operating picture in moments where seconds matter. We will walk through the specific limitations of architectural plans during emergencies and introduce ways to provide first responders with a true common operating picture— a map designed not for construction, but for crisis. This session will move from problem to solution, highlighting technologies and strategies that are redefining how agencies prepare for and respond to incidents in schools and large infrastructure environments, and discuss how GIS fits into this picture.

As the GIS community transitions to Experience Builder, we have a unique opportunity to rethink how we present information to our users. Rather than relying on complex and static geoprocessing, one can use Esri’s Arcade language to transform data directly within the web map. This session explores how a county GIS department uses Arcade to deliver high-functioning tools while maintaining a lean, simple database.

We will walk through three levels of practical implementation, in increasing complexity. First, we will demonstrate basic data cleansing. We use Arcade scripts to standardize “messy” legacy data, such as fixing title casing and formatting soil classes for historical reporting. Second, we look at Arcade’s dynamic capabilities, with features such as buttons that dynamically generate URLs and text that displays feature attributes from multiple layers. Finally, we explore spatial relationships and calculations, making use of `FeatureSetByName` and `Intersects` to perform on-the-fly queries, and rudimentary sliver checks to catch topology errors.
Attendees will see how these techniques reduce the need for intermediate "clipped" layers and redundant data maintenance. We will conclude with a practical discussion on the trade-offs of this approach, specifically regarding client-side performance and the balance between browser-side logic and server-side preparation.

Geohazard monitoring in the real world is dominated by messy data, weather interference, occlusions, sensor drift, dropouts, and site-to-site variability. Cascade Geomatics is building a robust alerting pipeline that turns imperfect sensing into reliable, georeferenced risk intelligence for rockfall, landslide, mudslide, and avalanche hazards.
We begin by constructing a large, local database of labeled geohazard events and near-miss conditions, aligned in space and time with multi-modal sensor streams. We then train machine-learning models to sift signal from noise and detect hazard-specific precursors, deformation rates, fracture evolution, moisture/loading indicators, and thermal/seasonal dynamics, while producing calibrated confidence scores.
In parallel, we maintain a physics-informed neural network (PINN) for each hazard class, using sensor-derived inputs as boundary/forcing terms to model near-real-time dynamics and constrain predictions when data quality degrades. The system continuously compares current conditions to historical analogs, fusing outputs from the historical neural models and the PINNs to generate GIS-ready risk layers: evolving hazard polygons, runout/impact overlays, and asset exposure scores, each with uncertainty and confidence.
This talk focuses on the practical path from “bad data” to actionable alerts, its limitations, and how reliability, explainability, and geospatial delivery are engineered into the system end-to-end.

Multnomah County's Division of Assessment, Recording and Taxation is re-imagining its Residential Reappraisal Program. This program, designed to support accurate property value assessments and fair taxation, involves county appraisers inspecting properties (from the sidewalk) to verify and collect data on land and improvements.

To execute this work efficiently, the County selected ArcGIS Survey123 for field data collection. With support from GIS staff, a pilot project covering nearly 2,000 properties is approaching completion. This pilot uses a Survey123 form built on an existing feature service, which includes related child (i.e. land and buildings) and grandchild (i.e. building components and value adjustments) data with one-to-many relationships. This setup allows appraisers to view, edit, and add to the existing Computer-Assisted Mass Appraisal (CAMA) data while in the field.

In addition to Survey123 development, attendees will also hear about associated:

Quality assurance/quality control (QA/QC) processes
Workflow management and progress tracking tools developed using ArcGIS Experience Builder and Dashboards
Strategies for database integration

This project develops a time-series InSAR workflow to monitor surface stability using Sentinel-1 SAR coherence data. By compiling multi-year interferometric pairs from a consistent acquisition geometry, the approach establishes a historical baseline of expected week-to-week coherence behavior across NW Oregon and SW Washington. New observations are compared against this baseline to identify statistically anomalous changes in surface conditions. The workflow combines cloud-based interferometric processing with local geospatial analysis, including automated extraction of coherence rasters and zonal statistics across vector or raster defined regions (administrative boundaries, FAR or land classification). The result is a scalable framework for detecting unusual landscape change, with potential applications in environmental monitoring, land management, and identifying unanticipated surface disturbances to both the natural and built environment.