scripts to import wetbulb from earthengine

scripts/wet_bulb/generate_wet_bulb_events.py

@@ -0,0 +1,89 @@
+# Copyright 2023 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the 'License');
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#         https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an 'AS IS' BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Script to generate WetBulb events using the events pipeline.
+
+The events pipeline is setup with the following stages:
+  1. earthengine:
+     It uses the NASA/GSFC/MERRA/slv/2 data set from EarthEngine to get regions
+     with high wetbulb temperature as a geoTif per day.
+
+  2. raster_csv
+     The geoTif is converted into a CSV with each pixel with regions with high
+     wet bulb temperature converted into an S2 cell of level 10
+     using the raster_to_csv.py utilities.
+
+  3. events
+     The resulting CSV with s2 cells is aggregated into events using
+     the process_events.py. Neighbouring s2 cells with high wet-bulb temperatures
+     are collated into a single event.
+     s2 cells with web bulb temperature over successive days are also
+     added into the same event.
+
+     Once the regions with wet bulb temperature collated into events,
+     the process_events.py utilities also generate a CSV with
+     StatVar Observations for Area_WetBulbTemperatureEvent and
+     Count_WetBulbTemperatureEvent for each events s2 cells at level 10, and places it is
+     contained in, such as AdministrativeArea, Country, Continent and Earth.
+
+  The generated files including the geoTif, CSVs with S2 cells and events are saved
+  in GCS.
+
+  The script runs once a month to update wetbulb events for the current year.
+"""
+
+import os
+import re
+import sys
+import time
+
+from absl import app
+from absl import flags
+from absl import logging
+
+_SCRIPTS_DIR = os.path.dirname(__file__)
+sys.path.append(_SCRIPTS_DIR)
+sys.path.append(os.path.dirname(_SCRIPTS_DIR))
+sys.path.append(os.path.dirname(os.path.dirname(_SCRIPTS_DIR)))
+sys.path.append(os.path.join(os.path.dirname(_SCRIPTS_DIR), 'earthengine'))
+sys.path.append(
+    os.path.join(os.path.dirname(os.path.dirname(_SCRIPTS_DIR)), 'util'))
+
+flags.DEFINE_string(
+    'wet_bulb_pipeline_config',
+    os.path.join(_SCRIPTS_DIR, 'wet_bulb_events_pipeline_config.py'),
+    'Config for the pipeline as a py dictionary of json')
+
+flags.DEFINE_list(
+    'wet_bulb_pipeline_stages', [],
+    'List of stages in the wetbulb events pipeline config to be run.')
+
+_FLAGS = flags.FLAGS
+
+import file_util
+
+from config_map import ConfigMap
+from events_pipeline import EventPipeline
+
+
+def main(_):
+    config=ConfigMap(
+        filename=_FLAGS.wet_bulb_pipeline_config)
+    if _FLAGS.start_date:
+      config.get('defaults', {})['start_date'] = _FLAGS.start_date
+    pipeline = EventPipeline(config=config)
+    pipeline.run(run_stages=_FLAGS.wet_bulb_pipeline_stages)
+
+
+if __name__ == '__main__':
+    app.run(main)

scripts/wet_bulb/wet_bulb_events.tmcf

@@ -0,0 +1,13 @@
+Node: E:Events->E0
+dcid: C:Events->dcid
+typeOf: C:Events->typeOf
+name: C:Events->name
+startDate: C:Events->startDate
+endDate: C:Events->endDate
+observationPeriod: C:Events->observationPeriod
+startLocation: C:Events->startLocation
+affectedPlace: C:Events->affectedPlace
+area: C:Events->area
+observationDate: C:Events->observationDate
+geoJsonCoordinates: C:Events->geoJsonCoordinates
+wetBulbTemperature: C:Events->wetBulbTemperature

scripts/wet_bulb/wet_bulb_events_pipeline_config.py

@@ -0,0 +1,187 @@
+# Config to generate WetBulbTemperatureEvent through the script: events_pipeline.py
+{
+    'defaults': {
+        'import_name': 'NASAWetBulb',
+        # Set start_date to start of year to be processed.
+        # Defaults to Jan 1 of current year if left empty.
+        'start_date': '',
+        'end_date': '',
+        'time_period': 'P1D',
+
+        # GCS settings
+        'gcs_project': 'datcom',
+        'gcs_bucket': 'datcom-prod-imports',
+        'gcs_folder': 'scripts/wet_bulb',
+    },
+    # State of previous run of the pipeline with input/output for each stage.
+    'pipeline_state_file':
+        'gs://datcom-prod-imports/scripts/wet_bulb/wet_bulb_event_pipeline_state_{year}.py',
+
+    # Pipeline stages to generate wet_bulb events.
+    'stages': [
+        # Generate geoTiff from EarthEngine Dynamic World data set.
+        {
+            'stage': 'earthengine',
+
+            # EE dataset from NASA/GSFC/MERRA
+            # https://developers.google.com/earth-engine/datasets/catalog/NASA_GSFC_MERRA_slv_2#description
+            'ee_image_collection': 'NASA/GSFC/MERRA/slv/2',
+            # Image processing settings.
+            'ee_reducer': 'max',
+
+            # Filter by min web bulb temperature of 32 deg C
+            'band': 'T2MWET',
+            'band_min': 303,  # 273.15(K) + 30,
+            # preserve the original temp after filtering by min threshold.
+            'ee_band_bool': False,
+            'ee_mask': 'land',
+
+            # Output image settings
+            'ee_output_data_type': 'float',
+            'scale': 10000,
+            'gcs_folder': 'scripts/wet_bulb/{stage}/{year}',
+            'ee_export_image': True,
+            # Generate daily images for a year at a time.
+            # Events are processed annually from Jan-Dec.
+            'ee_image_count': 365,
+            # 'ee_image_count': 31,
+            'skip_existing_output': True,
+        },
+
+        # Convert geoTiff to CSV with S2 cells.
+        {
+            'stage':
+                'raster_csv',
+            # debug
+            #'debug': True,
+            #'limit_points': 10,
+            's2_level':
+                10,
+            'aggregate':
+                'max',
+            'rename_columns': {
+              'band:1': 'T2MWET',
+            },
+            'input_data_filter': {
+                # Convert WetBulbTemperature to Celsius
+                'area': {
+                    # pick max area for s2 cell.
+                    'aggregate': 'max'
+                },
+                'T2MWET': {
+                    # convert value from Kelvin to Celsius
+                    'eval': '{T2MWET}-273.15',
+                    # Pick s2Cells with a min wetBulbTemperature
+                    'min': 28,
+                    'aggregate': 'max',
+                }
+            },
+
+            # use output from download stage as input
+            'input_files':
+                'gs://{gcs_bucket}/{gcs_folder}/earthengine/*{year}*.tif',
+            'output_dir':
+                'gs://{gcs_bucket}/{gcs_folder}/{stage}/{year}',
+            'skip_existing_output':
+                True,
+        },
+
+        # Generate events from the CSV with wet_bulbed S2 cells per month.
+        {
+            'stage':
+                'events',
+
+            # Process all data files for the whole year.
+            'input_files': 'tmp/fixed-temp/*.csv',
+            #'input_files':
+            #    'gs://{gcs_bucket}/{gcs_folder}/raster_csv/{year}/*{year}*raster_csv.csv',
+            'output_dir':
+                'gs://{gcs_bucket}/{gcs_folder}/{stage}/{import_name}-{stage}-{year}-',
+            'event_type':
+                'WetBulbTemperatureEvent',
+
+            # Input settings.
+            # Columms of input_csv that are added as event properties
+            'data_columns': ['area', 'T2MWET'],
+            'input_rename_columns': {
+                'date': 'observationDate',
+                'T2MWET': 'wetBulbTemperature',
+            },
+            # Input column for date.
+            'date_column':
+                'observationDate',
+            # Columns of input_csv that contains the s2 cell id.
+            'place_column':
+                's2CellId',
+            'input_filter_config': {
+                'wetBulbTemperature': {
+                    'min': 30,
+                },
+            },
+
+            # Processing settings
+            # Maximum distance within which 2 events are merged.
+            'max_overlap_distance_km':
+                0,
+            # Maximum number of cells of same level in between 2 events to be merged.
+            'max_overlap_place_hop':
+                2,
+            # S2 level to which data is aggregated.
+            's2_level':
+                10,  # Events are at resolution of level-10 S2 cells.
+            'aggregate':
+                'max',  # default aggregation for all properties
+            # Per property settings
+            'property_config': {
+                'area': {
+                    'aggregate': 'sum',
+                    'unit': 'SquareKilometer',
+                },
+                'wetBulbTemperature': {
+                    'aggregate': 'max',
+                    'unit': 'Celsius',
+                },
+                'affectedPlace': {
+                    'aggregate': 'list',
+                },
+            },
+            # Treat events at the same location more than 7 days apart as separate events.
+            'max_event_interval_days':
+                7,
+
+            # Enable DC API lookup for place properties
+            'dc_api_enabled':
+                False,
+            'dc_api_batch_size':
+                200,
+            # Cache file for place properties like name, location, typeOf
+            # Cache is updated with new places looked up.
+            'place_property_cache_file':
+                'gs://datcom-prod-imports/place_cache/place_properties_cache_with_s2_10.pkl',
+
+            # Output settings.
+            #'output_delimiter': ';',
+            'output_delimiter':
+                ',',
+            'output_svobs':
+                True,
+            'output_affected_place_polygon':
+                'geoJsonCoordinates',
+            'polygon_simplification_factor':
+                None,
+            'output_geojon_string':
+                True,
+
+            # Output svobs per place
+            'output_place_svobs':
+                True,
+            'output_place_svobs_properties': ['area', 'count'],
+            'output_place_svobs_dates': ['YYYY-MM-DD', 'YYYY-MM', 'YYYY'],
+            # Generate stats for all containedInPlaces for the event.
+            # Uses the containedInPlace property in the
+            # place_property_cache_file.
+            'aggregate_by_contained_in_place':
+                True,
+        },
+    ],
+}

scripts/wet_bulb/wet_bulb_place_svobs.tmcf

@@ -0,0 +1,16 @@
+Node: E:EventPlaces->E0
+typeOf: dcs:StatVarObservation
+variableMeasured: dcs:Area_WetBulbTemperatureEvent
+observationAbout: C:EventPlaces->observationAbout
+observationDate: C:EventPlaces->observationDate
+observationPeriod: C:EventPlaces->observationPeriod
+value: C:EventPlaces->area
+unit: SquareKilometer
+
+Node: E:EventPlaces->E1
+typeOf: dcs:StatVarObservation
+variableMeasured: dcs:Count_WetBulbTemperatureEvent
+observationAbout: C:EventPlaces->observationAbout
+observationDate: C:EventPlaces->observationDate
+observationPeriod: C:EventPlaces->observationPeriod
+value: C:EventPlaces->count