Spatial Heatmap Comparison - DSEclimatekitaeExt

2x2 heatmaps comparing Historical vs SSP scenarios for temperature and precipitation, showing the spatial pattern across the park boundary.

Uses a Coiled cluster (4 workers) to fetch gridded snapshots in parallel.

import sys
import os
import coiled

PROJECT_ROOT = os.path.abspath(os.path.join(os.getcwd(), "..", ".."))
sys.path.insert(0, os.path.join(PROJECT_ROOT, "lib"))

from andrewAdaptLibrary import (
    ParkCatalog,
    get_spatial_snapshot,
    plot_spatial_comparison,
)

NPS_SHP = os.path.join(
    PROJECT_ROOT,
    "USA_National_Park_Service_Lands_20170930_4993375350946852027",
    "USA_Federal_Lands.shp",
)
catalog = ParkCatalog(NPS_SHP)
print("Ready")

/opt/conda/envs/py-env/lib/python3.12/site-packages/pyogrio/raw.py:200: RuntimeWarning: /workspaces/DSEBrandNew/USA_National_Park_Service_Lands_20170930_4993375350946852027/USA_Federal_Lands.shp contains polygon(s) with rings with invalid winding order. Autocorrecting them, but that shapefile should be corrected using ogr2ogr for example.
  return ogr_read(

Ready

Load Park Boundaries¶

jt = catalog.get_boundary("Joshua Tree National Park")
mj = catalog.get_boundary("Mojave National Preserve")
print(f"Joshua Tree: {jt.total_bounds}")
print(f"Mojave: {mj.total_bounds}")

Joshua Tree: [-116.45805157   33.67019667 -115.26221427   34.12935132]
Mojave: [-116.16503909   34.71693524 -114.94916936   35.59077673]

Start Coiled Cluster¶

cluster = coiled.Cluster(
    name="spatial-heatmap",
    region="us-west1",
    n_workers=4,
    worker_memory="8 GiB",
    spot_policy="spot_with_fallback",
    idle_timeout="15 minutes",
    package_sync=True,
)
client = cluster.get_client()
print(f"Workers: {len(client.scheduler_info()['workers'])}")

/tmp/ipykernel_3814/1957208973.py:1: FutureWarning: `package_sync` is a deprecated kwarg for `Cluster` and will be removed in a future release. To only sync certain packages, use `package_sync_only`, and to disable package sync, pass the `container` or `software` kwargs instead.
  cluster = coiled.Cluster(

Workers: 0

Joshua Tree: Max Temperature (2050–2069 vs Historical)¶

scenarios = ["Historical Climate", "SSP 2-4.5", "SSP 3-7.0", "SSP 5-8.5"]

jt_tmax = get_spatial_snapshot(
    variable="T_Max",
    scenarios=scenarios,
    boundary=jt,
    time_period=(2050, 2069),
    coiled_cluster=cluster,
)

for scen, da in jt_tmax.items():
    print(f"{scen}: shape={da.shape}, mean={float(da.mean()):.1f}°C")

Historical Climate: shape=(15, 39), mean=26.2°C
SSP 2-4.5: shape=(15, 39), mean=28.3°C
SSP 3-7.0: shape=(15, 39), mean=28.4°C
SSP 5-8.5: shape=(15, 39), mean=29.1°C

plot_spatial_comparison(
    jt_tmax,
    boundary=jt,
    title="Joshua Tree — Mean Monthly Max Temperature (°C)",
)

Anomalies: Change from Historical Baseline¶

Subtract the Historical grid from each SSP to see where warming is strongest.

baseline = jt_tmax["Historical Climate"]

jt_tmax_anomalies = {
    f"{scen} minus Historical": da - baseline
    for scen, da in jt_tmax.items()
    if scen != "Historical Climate"
}

# Show the range
for scen, da in jt_tmax_anomalies.items():
    vals = da.values[~__import__('numpy').isnan(da.values)]
    print(f"{scen}: {vals.min():+.2f} to {vals.max():+.2f} °C")

SSP 2-4.5 minus Historical: +1.93 to +2.33 °C
SSP 3-7.0 minus Historical: +2.01 to +2.43 °C
SSP 5-8.5 minus Historical: +2.68 to +3.14 °C

plot_spatial_comparison(
    jt_tmax_anomalies,
    boundary=jt,
    title="Joshua Tree — T_Max Anomaly vs Historical (°C)",
    cmap="RdBu_r",
    reference_panel=("Historical (1995–2014)", jt_tmax["Historical Climate"], "RdYlBu_r"),
)

Mojave: Precipitation (2050–2069 vs Historical)¶

mj_precip = get_spatial_snapshot(
    variable="Precip",
    scenarios=scenarios,
    boundary=mj,
    time_period=(2050, 2069),
    coiled_cluster=cluster,
)

for scen, da in mj_precip.items():
    print(f"{scen}: shape={da.shape}, mean={float(da.mean()):.1f} mm/month")

Historical Climate: shape=(28, 39), mean=15.7 mm/month
SSP 2-4.5: shape=(28, 39), mean=15.3 mm/month
SSP 3-7.0: shape=(28, 39), mean=14.9 mm/month
SSP 5-8.5: shape=(28, 39), mean=16.6 mm/month

plot_spatial_comparison(
    mj_precip,
    boundary=mj,
    title="Mojave — Mean Monthly Precipitation (mm/month)",
    cmap="YlGnBu",
)

Precipitation Anomalies¶

baseline = mj_precip["Historical Climate"]

mj_precip_anomalies = {
    f"{scen} minus Historical": da - baseline
    for scen, da in mj_precip.items()
    if scen != "Historical Climate"
}

for scen, da in mj_precip_anomalies.items():
    vals = da.values[~__import__('numpy').isnan(da.values)]
    print(f"{scen}: {vals.min():+.2f} to {vals.max():+.2f} mm/month")

SSP 2-4.5 minus Historical: -1.98 to +1.24 mm/month
SSP 3-7.0 minus Historical: -2.54 to +0.77 mm/month
SSP 5-8.5 minus Historical: -0.69 to +2.70 mm/month

plot_spatial_comparison(
    mj_precip_anomalies,
    boundary=mj,
    title="Mojave — Precipitation Anomaly vs Historical (mm/month)",
    cmap="BrBG",
    reference_panel=("Historical (1995–2014)", mj_precip["Historical Climate"], "YlGnBu"),
)

Shut Down¶

cluster.close()
print("Done.")

Done.