docs(about): update for HRDPS, Rust pipeline, perf work, remove ML model refs

- Add Canadian HRDPS model coverage (to 60°N)
- Document Rust pipeline (prop-grid-rs) and its performance characteristics
- Add .sgrid/.pgrid binary format details
- Note ~6s chain step and 23-band fused pass
- Remove Nx/Axon/ExLA ML model references (scaffolding was removed)
- Update roadmap: drop model training, add PSKR calibration, propagation alerts
- Fix factor count (was 9, is 10) in roadmap item

lib/microwaveprop_web/live/about_live.ex

@@ -56,10 +56,11 @@ defmodule MicrowavepropWeb.AboutLive do
           <div role="alert" class="alert alert-info">
             <p>
               <strong>Coverage:</strong>
-              right now this only covers the continental US. Every upstream source
-              we're using — NOAA HRRR, ASOS, IEMRE, the ITU terrain data — is
-              CONUS-only. Hawaii, Alaska, Canada, and Mexico will need different
-              data sources and aren't wired up yet.
+              right now this focuses on the continental US. NOAA HRRR, ASOS,
+              IEMRE, and the ITU terrain data are CONUS-only. The Canadian HRDPS
+              model extends coverage north to 60°N (southern Canada), and SRTM
+              terrain tiles cover that range. Hawaii, Alaska, northern Canada,
+              and Mexico will need different data sources and aren't wired up yet.
             </p>
           </div>
         </div>
@@ -72,13 +73,16 @@ defmodule MicrowavepropWeb.AboutLive do
           <ol class="list-decimal list-outside pl-5 space-y-2">
             <li>
               <strong>The weather.</strong>
-              Hourly 3 km HRRR forecasts give surface fields and pressure-level
-              profiles. We derive refractivity profiles, minimum refractivity
-              gradient, ducting detection, boundary-layer depth, and precipitable
-              water for every grid cell. ASOS surface obs, 12-hourly upper-air
-              soundings, and gridded IEMRE reanalysis fill in the gaps. A
-              10-factor composite score is written out for every 0.125° cell
-              across CONUS, for each hour of a 48-hour forecast.
+              Hourly 3 km NOAA HRRR forecasts give surface fields and pressure-level
+              profiles across CONUS, while the Canadian HRDPS model (0.125°
+              resolution, 4× daily) extends coverage north to 60°N. We derive
+              refractivity profiles, minimum refractivity gradient, ducting
+              detection, boundary-layer depth, and precipitable water for every
+              grid cell. ASOS surface obs, 12-hourly upper-air soundings, and
+              gridded IEMRE reanalysis fill in the gaps. A 10-factor composite
+              score is written for every 0.125° cell (~95k across CONUS) across
+              23 microwave bands (902 MHz–241 GHz), for each hour of a 48-hour
+              forecast.
             </li>
             <li>
               <strong>The contacts.</strong>
@@ -154,6 +158,16 @@ defmodule MicrowavepropWeb.AboutLive do
               (HRRR pulls, terrain, ASOS, soundings, IEMRE, solar indices,
               commercial links).
             </li>
+            <li>
+              A Rust pipeline (prop-grid-rs) handles the heavy lifting:
+              fetching GRIB2 data from NOAA S3, decoding via wgrib2, deriving
+              weather scalars, and scoring all 23 bands in a single fused pass
+              over ~95k cells. Writes dense binary artifacts (.pgrid profiles,
+              .sgrid weather scalars, and per-band .prop score files) to NFS
+              for sub-millisecond single-cell reads from Elixir. A full
+              forecast-hour chain step runs in ~6 seconds — ~3× faster than
+              the previous Elixir-only pipeline.
+            </li>
             <li>
               Leaflet + Canvas tile layers for the propagation heatmap — we
               render 0.125° cells at interactive frame rates. Tailwind v4 +
@@ -167,16 +181,12 @@ defmodule MicrowavepropWeb.AboutLive do
             </li>
             <li>
               Data sources: NOAA HRRR (AWS S3, hourly analysis + 18 h
+              forecasts), Canadian HRDPS (MSC Datamart, 4× daily 48 h
               forecasts), Iowa Environmental Mesonet (ASOS and upper-air
               soundings), IEMRE gridded reanalysis, SRTM 90 m terrain, SNMP
               polling on seven commercial microwave links near DFW at 5-minute
               intervals, and NCEP NARR for pre-2014 contact enrichment.
             </li>
-            <li>
-              There's Nx/Axon/EXLA scaffolding for a small feed-forward model
-              (13 features → 64 → 32 → 1 sigmoid). Not trained yet — waiting
-              on enough calibrated data to not overfit.
-            </li>
           </ul>
         </div>
 
@@ -191,20 +201,21 @@ defmodule MicrowavepropWeb.AboutLive do
             </li>
             <li>
               <strong>Weight calibration.</strong>
-              Fit the 9 factor weights against recorded distances / QSO counts
-              per band so the score reflects real propagation, not my
-              intuition.
-            </li>
-            <li>
-              <strong>Training the model.</strong>
-              Once there are enough clean (contact, conditions) pairs, train
-              the Axon model to augment or replace the hand-tuned scoring.
+              Re-fit the 10 factor weights against recorded distances / QSO
+              counts per band so the score reflects real propagation, not hand
+              tuning. The PSKR spot ingestion pipeline (sampling every 5 min)
+              is building the calibration dataset automatically.
             </li>
             <li>
               <strong>Better inputs.</strong> MRMS for precipitation at 24+ GHz where rain attenuation
               dominates, RTMA/URMA surface analysis blending, GOES total
               precipitable water.
             </li>
+            <li>
+              <strong>Expanded coverage.</strong> The Canadian HRDPS model gives
+              us southern Canada (to 60°N). Alaska, Hawaii, Mexico, and the
+              Caribbean need their own data sources and scoring pipelines.
+            </li>
           </ul>
         </div>
         <div role="alert" class="alert alert-info text-center">