Is the river through my town running high, and is a flood building upstream?

Until recently, the only way to know how high a river was sitting was a gauge bolted to a bridge — and most rivers on Earth have none. NASA and CNES’s SWOT satellite changed that: it measures the water-surface elevation of rivers wider than about 100 m everywhere it flies over, breaking each river into short “reaches” and reporting a height and a width for each one. Watch the same reach across several overpasses and you can see the water climbing — and watch the reaches above your town to catch a flood wave while it is still upstream.

Verified locally. For the Lower Mississippi near Baton Rouge, the SWOT RiverSP reach product read a real water-surface elevation of 1.88 m (with a channel width of about 999 m) on a reach in the AOI from the 1 Jan 2024 overpass — one of 11 valid reaches the satellite measured here that pass. Heights are reported relative to a geoid datum, so the single number matters less than how it changes from one overpass to the next: a steady rise across passes is the signature of a flood building.

What you can answer

• How high a river reach is sitting on a given overpass — the wse field (water-surface elevation, in metres) for every reach SWOT measured over your town
• How wide the channel is, and whether it is widening — the width field, a direct sign of water spilling toward the banks
• Whether the river is rising — line up wse for the same reach_id across successive overpasses and read the trend
• Whether a flood is building upstream — check the reaches above your town first; a rise there precedes a rise at home by hours to days
• How far above normal today is — compare this overpass to the spread of past overpasses for the same reach
• Where water historically sits — overlay JRC Global Surface Water to see the floodplain and permanent-water baseline your reading should be judged against

What you can NOT answer with these datasets alone

• The actual discharge (flow in m³/s) — SWOT gives height and width, not volume; turning that into flow needs a rating relationship or a hydraulic model, not the product alone
• Exact heights against your local benchmark — wse is referenced to a satellite geoid datum, not the gauge datum your town’s flood stage is defined on; reconcile datums before comparing to a flood threshold
• Small streams — SWOT only resolves rivers roughly wider than 100 m; a creek through your neighbourhood is below its reach
• Continuous, real-time tracking — each reach is seen only on overpasses (every few days), so a fast flash flood can crest and fall between visits
• Why the river is rising — pair with GPM IMERG rainfall upstream to attribute a rise to a storm rather than a dam release or snowmelt
• Whether your specific street will flood — reach-scale elevation is not a street-level inundation map; that needs a terrain model and local hydraulics

Code template (Python, cloud-direct)

Verified locally. SWOT_L2_HR_RiverSP_reach_D is delivered as a zipped shapefile, not NetCDF. Download the granule, then open it with geopandas using a zip:// path. Each row is a river reach with reach_id, wse (metres), and width (metres); the fill value is -999999999999, so mask it out before computing anything.

import os, re, warnings, earthaccess, geopandas as gpd
warnings.filterwarnings("ignore")

# load Earthdata creds from .env without `source` (passwords can break the shell)
for line in open(".env"):
    m = re.match(r'\s*(?:export\s+)?([A-Z0-9_]+)\s*=\s*(.*)\s*$', line)
    if m: os.environ.setdefault(m.group(1), m.group(2).strip().strip('"').strip("'"))
earthaccess.login(strategy="environment")   # free Earthdata Login

W, S, E, N = -91.3, 30.2, -90.9, 30.6        # Lower Mississippi near Baton Rouge
g = earthaccess.search_data(short_name="SWOT_L2_HR_RiverSP_reach_D",
                            temporal=("2024-01-01", "2024-06-30"),
                            bounding_box=(W, S, E, N))

# the product is a ZIPPED SHAPEFILE — download, then read with a zip:// path
files = earthaccess.download(g[:1], local_path="swot")
gdf = gpd.read_file(f"zip://{files[0]}")

reaches = gdf.cx[W:E, S:N]                    # clip to your town
reaches = reaches[reaches["wse"] != -999999999999]   # drop the fill value
for _, r in reaches.iterrows():
    print(f"reach {r['reach_id']}: wse={r['wse']:.2f} m  width={r['width']:.0f} m")

# flood build-up = the SAME reach_id rising across successive overpasses.
# Loop several granules in date order, keep one reach_id, and plot wse over time;
# check the reaches UPSTREAM first to see a flood wave before it reaches you.