Free Open-Access Course

SPECTRA — Personal Exposure
& Trajectory Analysis

Learn to turn raw GPS tracks from wearables, phones and low-cost sensors into personal air-pollution and heat-stress exposure estimates — no coding required. Hands-on with realistic Zimbabwe sample data, using the live SPECTRA web app on the PALs Portal. For field researchers, students, and public-health teams.

Self-paced Open enrolment 8 Modules Free
Course at a Glance
⏱
Duration
~2 hours · 11 lessons across 8 app modules
🛰
Format
Live web app · browser-based · no install
🇿🇼
Sample data
Zimbabwe GPS + PM2.5, NO₂ & heat-stress (WBGT)
🎓
Developed by
Place Alert Labs (PALS) · Midlands State University
About This Course

"Where was a person, were they indoors or outdoors, how long did they stay, and how much did they breathe in there? SPECTRA answers that — point by point."

SPECTRA — the Spatiotemporal Personal Exposure Characterization and TRAjectory Analyzer — is the trajectory-and-exposure web application on the Place Alert Labs platform, built for the CHEAQI-MNCH project (Characterizing Effects of Air Quality In Maternal, Newborn and Child Health).

Two people walking the same street can have very different exposures. SPECTRA captures this by weighting every pollutant reading by how long a person actually dwelled there — so eight hours of indoor biomass cooking counts far more than a two-minute walk past traffic. That is the time-weighted personal exposure at the heart of the tool.

Across this course you will go from a raw CSV to a finished, publishable analysis: upload GPS data, map fields, fetch building footprints, classify indoor vs outdoor, segment micro-environments, read the exposure dashboard and maps, validate against ground truth, and export results — all without writing a line of code.

Every exercise uses fabricated-but-realistic Zimbabwe datasets (Gweru, Harare, Bulawayo, Mutare) with a built-in story: indoor cooking PM2.5 spikes, roadside NO₂ on commutes, and midday outdoor heat stress (WBGT) — so the dashboards and segmentation always have something meaningful to show.

Course Modules
01
Getting Started — Account, Sign-in & the Pipeline
Register on the PALs Portal, verify your email, and learn what SPECTRA does end-to-end: upload → field mapping → dwell time → buildings → indoor/outdoor → segmentation → grid → dashboard → export.
02
Preparing & Loading Data — Input & Configuration
Build a clean CSV (or use the Zimbabwe samples), upload CSV/RDS/GeoJSON/GPX/URL, and map your columns — latitude, longitude, person ID, date/time, and one or more exposures — with SPECTRA's auto-detection.
03
Configure & Run — Buildings, Buffers & Teleports
Set grid size, dwell caps, building buffer and teleport removal; choose a building source (OSM, Microsoft, Overture or custom) for indoor/outdoor classification; then run the analysis.
04
Dashboard — Reading Time-Weighted Exposure
Interpret participant counts, indoor/outdoor share and time-weighted mean exposure per pollutant. See how indoor cooking dominates the PM2.5 dose even when outdoor raw averages look higher.
05
Geospatial & Tracks — Maps that Tell the Story
Read classified points, per-person trajectory lines and exposure-shaded grid cells on interactive Leaflet maps. Sanity-check classification and find pollution hotspots.
06
Quality & Segmentation — Defend Your Method
Score the indoor/outdoor classifier with a confusion matrix and accuracy / precision / recall / F1 against ground-truth labels, then split each track into indoor, outdoor and in-transport micro-environments.
07
Heat Stress — WBGT as an Exposure
Use the Harare heat-stress dataset to analyse Wet-Bulb Globe Temperature alongside PM2.5 — mapping WBGT to an Exposure slot and reading midday outdoor heat risk.
08
Export, Sessions & Capstone ★ Capstone
Export classified points, grid cells and daily summaries (CSV/GeoJSON/JSON), save a reusable Session RDS, and complete a CHEAQI-style mini-analysis: time-weighted exposure, a hotspot map, an indoor-vs-transport dose split and a quality figure.
Who This Course Is For

Designed for field researchers, students, public-health officers and project collaborators who have GPS + exposure data and want personal-exposure estimates without programming. No GIS or coding background required.

  • A laptop with a modern browser (Chrome or Firefox recommended)
  • A dataset with person ID, timestamp, GPS coordinates and ≥1 exposure — or just use the Zimbabwe samples
  • Curiosity about air-pollution and heat exposure — no statistics needed
  • A SPECTRA / PALs Portal account (created when you enrol below)

Everything runs in the browser on the live SPECTRA app — nothing to install. The full lesson handbook and the downloadable Zimbabwe datasets are made available to enrolled participants on the PALs Portal once your enrolment is approved.

Developed by

Developed by Place Alert Labs (PALS), Midlands State University, for the CHEAQI-MNCH project, supported by DSI Africa.

Enrol in This Course

Register once. You will be contacted once your enrolment is reviewed, and can then log in to the PALs Portal to use SPECTRA.

Request submitted!
Thank you for enrolling in SPECTRA. Your request has been received and is under review.

You will be contacted with access details once approved.

This will be your login password for the PALs Portal once your enrolment is approved.

Your details are used only to support your course access and will not be shared with third parties. Place Alert Labs privacy policy applies.

Already enrolled

Trouble signing in? Email [email protected].