foo

clouds.py

@@ -3,12 +3,12 @@ from pathlib import Path
 from datetime import datetime
 
 
-def detect_clouds(year, site_name):
+def detect_clouds(season, site_name):
-    output_file = Path(f"data/{site_name}/{year}/clouds.json")
+    output_file = Path(f"data/{site_name}/{season}/clouds.json")
     clouds = {"s2": [], "s3": []}
 
     for source in ["s2", "s3"]:
-        timeseries_file = Path(f"data/{site_name}/{year}/ndvi/{source}/timeseries.json")
+        timeseries_file = Path(f"data/{site_name}/{season}/raw/ndvi/{source}/timeseries.json")
         if not timeseries_file.exists():
             print(f"[CLOUDS-{source.upper()}] No timeseries.json found")
             continue

download_s2.py

@@ -7,12 +7,12 @@ from rasterio.windows import from_bounds, transform as window_transform
 from pystac_client import Client
 
 
-def download_s2(year, site_position, site_name, date_range=None):
+def download_s2(season, site_position, site_name, date_range=None):
     lat, lon = site_position
-    datetime_range = date_range or f"{year}-01-01/{year}-12-31"
+    datetime_range = date_range or f"{season}-01-01/{season}-12-31"
-    output_dir = f"data/{site_name}/{year}/s2/"
+    output_dir = f"data/{site_name}/{season}/raw/s2/"
 
-    print(f"[S2] Starting download: {site_name} ({lat:.6f}, {lon:.6f}), {year}")
+    print(f"[S2] Starting download: {site_name} ({lat:.6f}, {lon:.6f}), {season}")
 
     bbox_size = 0.011
     bbox = [

download_s3.py

@@ -12,12 +12,12 @@ from rasterio.transform import from_bounds
 load_dotenv()
 
 
-def download_s3(year, site_position, site_name, date_range=None):
+def download_s3(season, site_position, site_name, date_range=None):
     lat, lon = site_position
-    datetime_range = date_range or f"{year}-01-01/{year}-12-31"
+    datetime_range = date_range or f"{season}-01-01/{season}-12-31"
-    output_dir = Path(f"data/{site_name}/{year}/s3/")
+    output_dir = Path(f"data/{site_name}/{season}/raw/s3/")
 
-    print(f"[S3] Starting download: {site_name} ({lat:.6f}, {lon:.6f}), {year}")
+    print(f"[S3] Starting download: {site_name} ({lat:.6f}, {lon:.6f}), {season}")
 
     bbox_size = 0.011
     bbox = [

efast.py

@@ -43,11 +43,11 @@ else:
         )
 
 
-def prepare_s2(year, site_position, site_name, date_range=None):
+def prepare_s2(season, site_position, site_name, date_range=None):
-    s2_dir = Path(f"data/{site_name}/{year}/s2/")
+    s2_dir = Path(f"data/{site_name}/{season}/raw/s2/")
-    s3_dir = Path(f"data/{site_name}/{year}/s3/")
+    s3_dir = Path(f"data/{site_name}/{season}/raw/s3/")
-    s2_output_dir = Path(f"data/{site_name}/{year}/efast/s2/")
+    s2_output_dir = Path(f"data/{site_name}/{season}/prepared/s2/")
-    clouds_file = Path(f"data/{site_name}/{year}/clouds.json")
+    clouds_file = Path(f"data/{site_name}/{season}/clouds.json")
 
     clouds = {"s2": set(), "s3": set()}
     if clouds_file.exists():
@@ -159,10 +159,10 @@ def prepare_s2(year, site_position, site_name, date_range=None):
                     dst.write(distance_to_cloud_lr, 1)
 
 
-def prepare_s3(year, site_position, site_name, date_range=None):
+def prepare_s3(season, site_position, site_name, date_range=None):
-    s3_dir = Path(f"data/{site_name}/{year}/s3/")
+    s3_dir = Path(f"data/{site_name}/{season}/raw/s3/")
-    s3_preprocessed_dir = Path(f"data/{site_name}/{year}/efast/s3/")
+    s3_preprocessed_dir = Path(f"data/{site_name}/{season}/prepared/s3/")
-    clouds_file = Path(f"data/{site_name}/{year}/clouds.json")
+    clouds_file = Path(f"data/{site_name}/{season}/clouds.json")
 
     clouds = {"s3": set()}
     if clouds_file.exists():
@@ -180,18 +180,18 @@ def prepare_s3(year, site_position, site_name, date_range=None):
         shutil.copy2(s3_file, output_path)
 
 
-def run_efast(year, site_position, site_name, date_range=None):
+def run_efast(season, site_position, site_name, date_range=None):
     lat, lon = site_position
-    datetime_range = date_range or f"{year}-01-01/{year}-12-31"
+    datetime_range = date_range or f"{season}-01-01/{season}-12-31"
 
-    efast_base_dir = Path(f"data/{site_name}/{year}/efast/")
+    efast_base_dir = Path(f"data/{site_name}/{season}/prepared/")
     s2_output_dir = efast_base_dir / "s2"
     s3_output_dir = efast_base_dir / "s3"
     fusion_output_dir = efast_base_dir / "fusion"
 
     fusion_output_dir.mkdir(parents=True, exist_ok=True)
 
-    print(f"[EFAST] Starting fusion: {site_name} ({lat:.6f}, {lon:.6f}), {year}")
+    print(f"[EFAST] Starting fusion: {site_name} ({lat:.6f}, {lon:.6f}), {season}")
 
     start_date = datetime.strptime(datetime_range.split("/")[0], "%Y-%m-%d")
     end_date = datetime.strptime(datetime_range.split("/")[1], "%Y-%m-%d")

ndvi.py

@@ -51,7 +51,7 @@ def _create_timeseries_for_dir(output_dir, site_position, source_name):
                 lon, lat = site_position[1], site_position[0]
                 x, y = transform_coords("EPSG:4326", src.crs, [lon], [lat])
                 samples = list(src.sample([(x[0], y[0])]))
-                if samples and len(samples) > 0:
+                if samples:
                     value = float(samples[0][0])
                     if value != 0 and not np.isnan(value):
                         ndvi_value = value
@@ -68,10 +68,10 @@ def _create_timeseries_for_dir(output_dir, site_position, source_name):
     print(f"[NDVI-{source_name}] Saved: {timeseries_file} ({len(timeseries)} entries)")
 
 
-def generate_ndvi(year, site_position, site_name):
+def generate_ndvi_raw(season, site_position, site_name):
     for source in ["s2", "s3"]:
-        input_dir = Path(f"data/{site_name}/{year}/{source}/")
+        input_dir = Path(f"data/{site_name}/{season}/raw/{source}/")
-        output_dir = Path(f"data/{site_name}/{year}/ndvi/{source}/")
+        output_dir = Path(f"data/{site_name}/{season}/raw/ndvi/{source}/")
         output_dir.mkdir(parents=True, exist_ok=True)
 
         print(f"[NDVI-{source.upper()}] Processing {input_dir}...")
@@ -94,15 +94,46 @@ def generate_ndvi(year, site_position, site_name):
         print(f"[NDVI-{source.upper()}] Completed")
 
 
-def create_ndvi_timeseries(year, site_position, site_name):
+def create_ndvi_timeseries_raw(season, site_position, site_name):
     for source in ["s2", "s3"]:
-        output_dir = Path(f"data/{site_name}/{year}/ndvi/{source}/")
+        output_dir = Path(f"data/{site_name}/{season}/raw/ndvi/{source}/")
         _create_timeseries_for_dir(output_dir, site_position, source.upper())
 
 
-def generate_ndvi_fusion(year, site_position, site_name):
+def generate_ndvi_prepared(season, site_position, site_name):
-    input_dir = Path(f"data/{site_name}/{year}/efast/fusion/")
+    for source in ["s2", "s3"]:
-    output_dir = Path(f"data/{site_name}/{year}/ndvi/fusion/")
+        input_dir = Path(f"data/{site_name}/{season}/prepared/{source}/")
+        output_dir = Path(f"data/{site_name}/{season}/prepared/ndvi/{source}/")
+        output_dir.mkdir(parents=True, exist_ok=True)
+
+        print(f"[NDVI-PREPARED-{source.upper()}] Processing {input_dir}...")
+
+        geotiff_files = sorted(input_dir.glob("*.geotiff")) + sorted(input_dir.glob("*.tif"))
+        if not geotiff_files:
+            print(f"[NDVI-PREPARED-{source.upper()}] No files found")
+            continue
+
+        for geotiff_file in geotiff_files:
+            if geotiff_file.suffix == ".tif":
+                if "REFL" in geotiff_file.stem:
+                    date_str = geotiff_file.stem.split("_")[1]
+                    output_file = output_dir / f"{date_str}_ndvi.geotiff"
+                else:
+                    output_file = output_dir / geotiff_file.name.replace(".tif", ".geotiff")
+            else:
+                output_file = output_dir / geotiff_file.name
+
+            if output_file.exists():
+                print(f"[NDVI-PREPARED-{source.upper()}] Skipping {geotiff_file.name} (exists)")
+                continue
+
+            _calculate_and_write_ndvi(geotiff_file, output_file)
+            print(f"[NDVI-PREPARED-{source.upper()}] Saved: {output_file}")
+
+        print(f"[NDVI-PREPARED-{source.upper()}] Completed")
+
+    input_dir = Path(f"data/{site_name}/{season}/prepared/fusion/")
+    output_dir = Path(f"data/{site_name}/{season}/prepared/ndvi/fusion/")
     output_dir.mkdir(parents=True, exist_ok=True)
 
     print(f"[NDVI-FUSION] Processing {input_dir}...")
@@ -126,6 +157,10 @@ def generate_ndvi_fusion(year, site_position, site_name):
     print(f"[NDVI-FUSION] Completed")
 
 
-def create_ndvi_timeseries_fusion(year, site_position, site_name):
+def create_ndvi_timeseries_prepared(season, site_position, site_name):
-    output_dir = Path(f"data/{site_name}/{year}/ndvi/fusion/")
+    for source in ["s2", "s3"]:
+        output_dir = Path(f"data/{site_name}/{season}/prepared/ndvi/{source}/")
+        _create_timeseries_for_dir(output_dir, site_position, f"PREPARED-{source.upper()}")
+
+    output_dir = Path(f"data/{site_name}/{season}/prepared/ndvi/fusion/")
     _create_timeseries_for_dir(output_dir, site_position, "FUSION")

run.py

@@ -1,34 +1,42 @@
 from efast import run_efast, prepare_s2, prepare_s3
-from ndvi import generate_ndvi_fusion, create_ndvi_timeseries_fusion
+from ndvi import (
+    generate_ndvi_raw,
+    create_ndvi_timeseries_raw,
+    generate_ndvi_prepared,
+    create_ndvi_timeseries_prepared,
+)
+from download_s2 import download_s2
+from download_s3 import download_s3
+from clouds import detect_clouds
 
-year = 2024
-site_position = (47.116171, 11.320308)
-site_name = "innsbruck"
 
-# print(f"Downloading data for {site_name}, {year}")
+def run_pipeline(season, site_position, site_name):
-# download_s2(year, site_position, site_name)
+    try:
-# download_s3(year, site_position, site_name)
+        print(f"Downloading data for {site_name}, {season}")
-# print("All downloads completed")
+        download_s2(season, site_position, site_name)
+        download_s3(season, site_position, site_name)
 
-# print(f"Generating NDVI for {site_name}, {year}")
+        print(f"Generating NDVI for raw data: {site_name}, {season}")
-# generate_ndvi(year, site_position, site_name)
+        generate_ndvi_raw(season, site_position, site_name)
-# create_ndvi_timeseries(year, site_position, site_name)
+        create_ndvi_timeseries_raw(season, site_position, site_name)
-# print("All NDVI generation completed")
 
-# print(f"Detecting clouds for {site_name}, {year}")
+        print(f"Detecting clouds for {site_name}, {season}")
-# detect_clouds(year, site_name)
+        detect_clouds(season, site_name)
-# print("Cloud detection completed")
 
-# print(f"Preparing data for EFAST fusion for {site_name}, {year}")
+        print(f"Preparing data for EFAST fusion for {site_name}, {season}")
-# prepare_s2(year, site_position, site_name)
+        prepare_s2(season, site_position, site_name)
-# prepare_s3(year, site_position, site_name)
+        prepare_s3(season, site_position, site_name)
-# print("Data preparation completed")
 
-# print(f"Running EFAST fusion for {site_name}, {year}")
+        print(f"Running EFAST fusion for {site_name}, {season}")
-# run_efast(year, site_position, site_name)
+        run_efast(season, site_position, site_name)
-# print("EFAST fusion completed")
 
-print(f"Generating NDVI for fusion outputs: {site_name}, {year}")
+        print(f"Generating NDVI for prepared outputs: {site_name}, {season}")
-generate_ndvi_fusion(year, site_position, site_name)
+        generate_ndvi_prepared(season, site_position, site_name)
-create_ndvi_timeseries_fusion(year, site_position, site_name)
+        create_ndvi_timeseries_prepared(season, site_position, site_name)
-print("Fusion NDVI generation completed")
+    except Exception as e:
+        print(f"Error: {e}")
+        raise
+
+
+if __name__ == "__main__":
+    run_pipeline(2024, (47.116171, 11.320308), "innsbruck")