In [1]:
import pandas as pd
import csv
import matplotlib.pyplot as plt
import numpy as np
import json
import os
from datetime import datetime
In [3]:
#copied files for 2025 from
#/uufs/chpc.utah.edu/common/home/horel-group/archive/uunet_loggernet
#where are 2024 files now??
# Read the CSV files and extract column names and units
def read_weather_csv(file_path):
    # Read the first row separately to extract column names
    with open(file_path, 'r') as file:
        lines = file.readlines()
    
    # Extract column names and units, removing unnecessary quotes
    column_names = [col.strip('"') for col in lines[1].strip().split(',')]
    units = {col.strip('"'): unit.strip('"') for col, unit in zip(column_names, lines[2].strip().split(','))}
    
    # Read the data starting from the 4th row
    df = pd.read_csv(file_path, skiprows=4, names=column_names)
    
    return df, dict(zip(column_names, units))

# Read the 1 min CSV files
df1, units1 = read_weather_csv('./uupya_2024.dat')
df2, units2 = read_weather_csv('./uupya_2025.dat')

# Read the 15 min soilVUE min CSV files
df1_soil, units1_soil = read_weather_csv('./uupya_soil_2024.dat')
df2_soil, units2_soil = read_weather_csv('./uupya_soil_2025.dat')

# Read the 30 min flux CSV files
df1_flux, units1_flux = read_weather_csv('./uupya_flux_2024.dat')
df2_flux, units2_flux = read_weather_csv('./uupya_flux_2025.dat')

# Merge the dataframes
df = pd.concat([df1, df2], ignore_index=True)
df_soil = pd.concat([df1_soil, df2_soil], ignore_index=True)
df_flux = pd.concat([df1_flux, df2_flux], ignore_index=True)

# Replace 'NaN' string values with np.nan in all columns
df.replace('NaN', np.nan, inplace=True)
df_soil.replace('NaN', np.nan, inplace=True)
df_flux.replace('NaN', np.nan, inplace=True)

# Convert TIMESTAMP to datetime and localize to Denver timezone
df['TIMESTAMP'] = pd.to_datetime(df['TIMESTAMP'], errors='coerce')
df = df.dropna(subset=['TIMESTAMP'])  # Drop NaT values
df['TIMESTAMP'] = df['TIMESTAMP'].dt.tz_localize('UTC').dt.tz_convert('America/Denver')

df_soil['TIMESTAMP'] = pd.to_datetime(df_soil['TIMESTAMP'], errors='coerce')
df_soil = df_soil.dropna(subset=['TIMESTAMP'])  # Drop NaT values
df_soil['TIMESTAMP'] = df_soil['TIMESTAMP'].dt.tz_localize('UTC').dt.tz_convert('America/Denver')

df_flux['TIMESTAMP'] = pd.to_datetime(df_flux['TIMESTAMP'], errors='coerce')
df_flux = df_flux.dropna(subset=['TIMESTAMP'])  # Drop NaT values
df_flux['TIMESTAMP'] = df_flux['TIMESTAMP'].dt.tz_localize('UTC').dt.tz_convert('America/Denver')

df = df.set_index('TIMESTAMP')
df_soil = df_soil.set_index('TIMESTAMP')
df_flux = df_flux.set_index('TIMESTAMP')
df = df.astype(float)
df_soil = df_soil.astype(float)
df_flux = df_flux.astype(float)

# Save the merged DataFrame to a new CSV file
df.to_csv('uupya_2024_2025.csv', index=False)
df_soil.to_csv('uupya_soil_2024_2025.csv', index=False)
df_flux.to_csv('uupya_flux_2024_2025.csv', index=False)

# Print the units dictionary
print("Units:", units1)
print("Units:", units1_soil)
print("Units:", units1_flux)

/tmp/ipykernel_613800/1588017085.py:15: DtypeWarning: Columns (26,28,35,36,37,38,39,40,41,42) have mixed types. Specify dtype option on import or set low_memory=False.
  df = pd.read_csv(file_path, skiprows=4, names=column_names)
/tmp/ipykernel_613800/1588017085.py:15: DtypeWarning: Columns (27,29) have mixed types. Specify dtype option on import or set low_memory=False.
  df = pd.read_csv(file_path, skiprows=4, names=column_names)

Units: {'TIMESTAMP': 'TIMESTAMP', 'RECORD': 'RECORD', 'BattV_Avg': 'BattV_Avg', 'PTemp_C_Avg': 'PTemp_C_Avg', 'SWin_Avg': 'SWin_Avg', 'SWout_Avg': 'SWout_Avg', 'LWin_Avg': 'LWin_Avg', 'LWout_Avg': 'LWout_Avg', 'UVin_Avg': 'UVin_Avg', 'UVout_Avg': 'UVout_Avg', 'WS_ms_1_Avg': 'WS_ms_1_Avg', 'WS_ms_2_Avg': 'WS_ms_2_Avg', 'WS_ms_3_Avg': 'WS_ms_3_Avg', 'WS_ms_4_Avg': 'WS_ms_4_Avg', 'WS_ms_5_Avg': 'WS_ms_5_Avg', 'WS_ms_6_Avg': 'WS_ms_6_Avg', 'WindDir': 'WindDir', 'AirTC_2m_Avg': 'AirTC_2m_Avg', 'RH_2m': 'RH_2m', 'AirTC_5m_Avg': 'AirTC_5m_Avg', 'RH_5m': 'RH_5m', 'AirTC_10m_Avg': 'AirTC_10m_Avg', 'RH_10m': 'RH_10m', 'AirTC_10m_TC_Avg': 'AirTC_10m_TC_Avg', 'SoilTempAv_Avg(1)': 'SoilTempAv_Avg(1)', 'SoilTempAv_Avg(2)': 'SoilTempAv_Avg(2)', 'shf_Avg(1)': 'shf_Avg(1)', 'shf_Avg(2)': 'shf_Avg(2)', 'shf_mV_Avg(1)': 'shf_mV_Avg(1)', 'shf_mV_Avg(2)': 'shf_mV_Avg(2)', 'Sensit_Tot': 'Sensit_Tot', 'SenSec': 'SenSec', 'Rain_mm_Tot': 'Rain_mm_Tot', 'LWmV_Avg': 'LWmV_Avg', 'LWMWet': 'LWMWet', 'PM_1_Avg': 'PM_1_Avg', 'PM_25_Avg': 'PM_25_Avg', 'PM_4_Avg': 'PM_4_Avg', 'PM_10_Avg': 'PM_10_Avg', 'PM_Flow_Avg': 'PM_Flow_Avg', 'PM_Pressure_Avg': 'PM_Pressure_Avg', 'PM_Temp_Avg': 'PM_Temp_Avg', 'PM_RH_Avg': 'PM_RH_Avg'}
Units: {'TIMESTAMP': 'TIMESTAMP', 'RECORD': 'RECORD', 'VWC_05cm': 'VWC_05cm', 'Ka_05cm': 'Ka_05cm', 'T_05cm': 'T_05cm', 'BulkEC_05cm': 'BulkEC_05cm', 'VWC_10cm': 'VWC_10cm', 'Ka_10cm': 'Ka_10cm', 'T_10cm': 'T_10cm', 'BulkEC_10cm': 'BulkEC_10cm', 'VWC_20cm': 'VWC_20cm', 'Ka_20cm': 'Ka_20cm', 'T_20cm': 'T_20cm', 'BulkEC_20cm': 'BulkEC_20cm', 'VWC_30cm': 'VWC_30cm', 'Ka_30cm': 'Ka_30cm', 'T_30cm': 'T_30cm', 'BulkEC_30cm': 'BulkEC_30cm', 'VWC_40cm': 'VWC_40cm', 'Ka_40cm': 'Ka_40cm', 'T_40cm': 'T_40cm', 'BulkEC_40cm': 'BulkEC_40cm', 'VWC_50cm': 'VWC_50cm', 'Ka_50cm': 'Ka_50cm', 'T_50cm': 'T_50cm', 'BulkEC_50cm': 'BulkEC_50cm'}
Units: {'TIMESTAMP': 'TIMESTAMP', 'RECORD': 'RECORD', 'shf_Avg(1)': 'shf_Avg(1)', 'shf_Avg(2)': 'shf_Avg(2)', 'shf_cal(1)': 'shf_cal(1)', 'shf_cal(2)': 'shf_cal(2)', 'shf_mV_run_Avg(1)': 'shf_mV_run_Avg(1)', 'shf_mV_run_Avg(2)': 'shf_mV_run_Avg(2)', 'shf_mV_0_Avg(1)': 'shf_mV_0_Avg(1)', 'shf_mV_0_Avg(2)': 'shf_mV_0_Avg(2)', 'shf_mV_180_Avg(1)': 'shf_mV_180_Avg(1)', 'shf_mV_180_Avg(2)': 'shf_mV_180_Avg(2)', 'shf_mV_end_Avg(1)': 'shf_mV_end_Avg(1)', 'shf_mV_end_Avg(2)': 'shf_mV_end_Avg(2)', 'V_Rf_Avg(1)': 'V_Rf_Avg(1)', 'V_Rf_Avg(2)': 'V_Rf_Avg(2)', 'V_Rf_run_Avg(1)': 'V_Rf_run_Avg(1)', 'V_Rf_run_Avg(2)': 'V_Rf_run_Avg(2)', 'V_Rf_180_Avg(1)': 'V_Rf_180_Avg(1)', 'V_Rf_180_Avg(2)': 'V_Rf_180_Avg(2)'}
In [4]:
#Plot time series for 'shf_Avg(2)'
plt.figure(figsize=(10, 5))
plt.plot(df_flux['shf_Avg(2)'])
#plt.xlabel('Timestamp (Denver Time)')
#plt.ylabel(units1_flux.get('shf_Avg(2)', 'Unknown'))
#plt.title('Time Series of shf_Avg(2)')
#plt.legend()
#plt.grid()
#plt.xticks(rotation=45)
plt.show()
In [5]:
# Plot time series for 'shf_Avg(1)'
plt.figure(figsize=(10, 5))
plt.plot( df_flux['shf_Avg(1)'], label='shf_Avg(1)', color='b')
plt.xlabel('Time)')
#plt.ylabel(units1_flux.get('shf_Avg(2)', 'Unknown'))
#plt.title('Time Series of shf_Avg(2)')
#plt.legend()
#plt.grid()
plt.show()
In [6]:
RN = df['SWin_Avg']+df['LWin_Avg']-(df['SWout_Avg']+df['LWout_Avg'])
# Plot time series for 'shf_Avg(1)'
plt.figure(figsize=(10, 5))
plt.plot(RN, color='b')
plt.xlabel('Time)')
#plt.ylabel(units1_flux.get('shf_Avg(2)', 'Unknown'))
#plt.title('Time Series of shf_Avg(2)')
#plt.legend()
#plt.grid()
plt.xticks(rotation=45)
plt.show()
In [7]:
# Compute 30-minute means for each column
df_30min = df.resample('30T').mean().reset_index()
df_30min = df_30min.set_index('TIMESTAMP')
df_30min['RN'] = df_30min['SWin_Avg']+df_30min['LWin_Avg']-(df_30min['SWout_Avg']+df_30min['LWout_Avg'])
# Plot time series for 'shf_Avg(1)'
plt.figure(figsize=(10, 5))
plt.plot(df_30min['RN'], color='b')
plt.xlabel('Time)')
#plt.ylabel(units1_flux.get('shf_Avg(2)', 'Unknown'))
#plt.title('Time Series of shf_Avg(2)')
#plt.legend()
plt.grid()
plt.show()
In [8]:
# Compute 30-minute means for each column
df_soil_30min = df_soil.resample('30T').mean().reset_index()
df_soil_30min = df_soil_30min.set_index('TIMESTAMP')
# Plot time series for 'shf_Avg(1)'
plt.figure(figsize=(10, 5))
plt.plot(df_soil_30min['T_05cm'], color='b')
plt.xlabel('Time)')
#plt.ylabel(units1_flux.get('shf_Avg(2)', 'Unknown'))
#plt.title('Time Series of shf_Avg(2)')
#plt.legend()
plt.grid()
plt.show()
In [9]:
df_30min['RNG'] = df_30min['RN']-df['shf_Avg(2)']
plt.figure(figsize=(10, 5))
plt.plot(df_30min['RNG'], color='b')
plt.xlabel('Time)')
#plt.ylabel(units1_flux.get('shf_Avg(2)', 'Unknown'))
#plt.title('Time Series of shf_Avg(2)')
#plt.legend()
plt.grid()
plt.show()
In [10]:
df_30min['Tsoil'] = df_30min['SoilTempAv_Avg(2)']
plt.figure(figsize=(10, 5))
plt.plot(df_30min['Tsoil']-df_soil_30min['T_05cm'], color='b')
plt.xlabel('Time)')
#plt.ylabel(units1_flux.get('shf_Avg(2)', 'Unknown'))
#plt.title('Time Series of shf_Avg(2)')
#plt.legend()
plt.grid()
plt.show()
In [11]:
dfc = df_30min[df_30min['RN'] > 20]
dfcd = dfc.resample('D').mean().reset_index()
dfcd = dfcd.set_index('TIMESTAMP')
dfcd.index = dfcd.index.tz_localize(None)  # Ensure index is timezone naive
plt.figure(figsize=(10, 5))
plt.plot(dfcd['RNG'], color='b')
plt.xlabel('Time)')
#plt.ylabel(units1_flux.get('shf_Avg(2)', 'Unknown'))
#plt.title('Time Series of shf_Avg(2)')
#plt.legend()
plt.grid()
plt.show()
In [ ]:
In [12]:
# Directory containing JSON files
directory = "/uufs/chpc.utah.edu/common/home/horel-group9/flux/level1"

#current version is v25
# List JSON files matching the pattern
json_files = [f for f in os.listdir(directory) if f.endswith("_v25.json") and "uupyf_combo" in f]

# Initialize empty lists to store data
data_ec150 = []
data_li710 = []
table_name_ec150 = "table_ec150"  # Adjust if the table name is different
table_name_li710 = "table_li710"  # Adjust if the table name is different

# Process each JSON file
for file in json_files:
    file_path = os.path.join(directory, file)
    with open(file_path, 'r') as f:
        json_data = json.load(f)
        
        # Extract records from the specified tables
        if table_name_ec150 in json_data['data_tables']:
            records_ec150 = json_data['data_tables'][table_name_ec150]['records']
            data_ec150.extend(records_ec150)
        
        if table_name_li710 in json_data['data_tables']:
            records_li710 = json_data['data_tables'][table_name_li710]['records']
            data_li710.extend(records_li710)

# Convert lists to DataFrames
df_ec150 = pd.DataFrame(data_ec150)
df_li710 = pd.DataFrame(data_li710)

Extract the table names¶

tablenames = list(data['data_tables'].keys()) print("data_tables:")

Traverse through each table and extract the nested structure¶

for tablename in tablenames: print(f" {tablename}:")

# Access the table
table = data['data_tables'][tablename]

# Extract metadata
metadata = table.get('metadata', {})
print("    metadata:")
for key, value in metadata.items():
    print(f"      {key}: {value}")

# Extract variables
variables = table.get('variables', {})
print("    variables:")
for var_name, var_details in variables.items():
    print(f"      {var_name}:")
    for detail_key, detail_value in var_details.items():
        print(f"        {detail_key}: {detail_value}")

# Extract records
records = table.get('records', [])
print("    records:")
if isinstance(records, list) and len(records) > 0:
    print(f"      Total Records: {len(records)}")
    print(f"      Example Record:")
    for key, value in records[0].items():
        print(f"        {key}: {value}")
else:
    print("      No records found.")
In [13]:
if 'TIMESTAMP' in df_ec150.columns and 'LE' in df_ec150.columns:
    # Convert the TIMESTAMP to a pandas datetime object
    df_ec150['TIMESTAMP'] = pd.to_datetime(df_ec150['TIMESTAMP'], errors='coerce')
    df_li710['TIMESTAMP'] = pd.to_datetime(df_li710['TIMESTAMP'], errors='coerce')
    df_ec150['LE'] = pd.to_numeric(df_ec150['LE'], errors='coerce')
    df_li710['LE'] = pd.to_numeric(df_li710['LE'], errors='coerce')
    df_ec150 = df_ec150.dropna(subset=['TIMESTAMP'])  # Drop NaT values
    df_ec150['TIMESTAMP'] = df_ec150['TIMESTAMP'].dt.tz_localize('UTC').dt.tz_convert('America/Denver')
    df_li710 = df_li710.dropna(subset=['TIMESTAMP'])  # Drop NaT values
    df_li710['TIMESTAMP'] = df_li710['TIMESTAMP'].dt.tz_localize('UTC').dt.tz_convert('America/Denver')

    # Sort by TIMESTAMP
    df_ec150 = df_ec150.sort_values('TIMESTAMP')
    df_li710 = df_li710.sort_values('TIMESTAMP')
    
    # Plot the ET variable as a time series
    plt.figure(figsize=(12, 6))
    plt.plot(df_ec150['TIMESTAMP'], df_ec150['LE'].astype('float64'), label='LE', marker='o', linestyle='-')
    #plt.plot(df_li710['TIMESTAMP'], df_li710['LE'].astype('float64'), label='LE', marker='+', linestyle='--')
    plt.xlabel('Time')
    plt.ylim(-100.,400)
    plt.ylabel('LE')
    plt.title('LE Time Series')
    plt.legend()
    plt.grid()
    plt.show()
else:
    print("The required 'TIMESTAMP' or 'LE' variable is not present in the data.")
In [14]:
    
# Filter for March 24, 2025
start_date = pd.Timestamp('2025-03-25', tz='America/Denver') 
end_date = pd.Timestamp('2025-03-26', tz='America/Denver')
df_ec150_day = df_ec150[(df_ec150['TIMESTAMP'] >= start_date) & (df_ec150['TIMESTAMP'] < end_date)]
df_li710_day = df_li710[(df_li710['TIMESTAMP'] >= start_date) & (df_li710['TIMESTAMP'] < end_date)]

# Sort by TIMESTAMP
df_ec150_day = df_ec150_day.sort_values('TIMESTAMP')
df_li710_day = df_li710_day.sort_values('TIMESTAMP')

# Plot the ET variable as a time series
plt.figure(figsize=(12, 6))
plt.plot(df_ec150_day['TIMESTAMP'], df_ec150_day['LE'].astype('float64'), label='EC150 LE', marker='o', linestyle='-',color='grey')
plt.plot(df_li710_day['TIMESTAMP'], df_li710_day['LE'].astype('float64'), label='LI-710 LE', marker='+', linestyle='--',color='grey')
plt.plot(df_ec150_day['TIMESTAMP'], df_ec150_day['H'].astype('float64'), label='EC150 H', marker='o', linestyle='-',color='orange')
plt.plot(df_li710_day['TIMESTAMP'], df_li710_day['H'].astype('float64'), label='LI-710 H', marker='+', linestyle='--',color='orange')
plt.xlabel('Time')

plt.ylabel('Flux (W/m$^2$)')
plt.title('Time Series for 2025-03-25')
plt.legend()
plt.grid()
plt.show()
In [15]:
# Convert numeric columns to proper data types
def convert_numeric(df):
    for col in df.columns:
        df[col] = pd.to_numeric(df[col], errors='coerce')
    return df

df_ec150 = convert_numeric(df_ec150)
df_li710 = convert_numeric(df_li710)

# Rename columns to distinguish sources
df_ec150 = df_ec150.add_prefix("ec_")
df_li710 = df_li710.add_prefix("li_")

# Convert timestamps to datetime
df_ec150['ec_TIMESTAMP'] = pd.to_datetime(df_ec150['ec_TIMESTAMP'], errors='coerce')
df_li710['li_TIMESTAMP'] = pd.to_datetime(df_li710['li_TIMESTAMP'], errors='coerce')

# Set ec_TIMESTAMP as index
df_ec150.set_index('ec_TIMESTAMP', inplace=True)

# Merge the two DataFrames using concat to handle different index types

df_merged = pd.concat([df_ec150, df_li710.set_index('li_TIMESTAMP')], axis=1, join='inner')
    
# Apply filters to consider only when all conditions are met simultaneously
df_merged = df_merged[(df_merged['ec_LE_QC'] <= 6) & (df_merged['ec_H_QC'] <= 6) & (df_merged['li_licor_diag'] < 128)]
In [16]:
df_merged['ec_LE_H'] = df_merged['ec_LE']+df_merged['ec_H']
df_merged['li_LE_H'] = df_merged['li_LE']+df_merged['li_H']

#remove really bad values. handle better later
df_merged = df_merged[(df_merged['li_LE_H'] <= 400)]
   
plt.figure(figsize=(12, 6))
plt.plot(df_merged['ec_LE_H'],label='EC', marker='o', linestyle='-')
plt.plot(df_merged['li_LE_H'], label='LI', marker='_', linestyle='--')

plt.xlabel('Time')
plt.ylim(-100.,400)
plt.ylabel('LE')
plt.title('LE Time Series')
plt.legend()
plt.grid()
plt.show()
In [ ]:
In [17]:
dfmd = df_merged.resample('D').mean()
dfmd.index = pd.to_datetime(dfmd.index)  # Ensure index remains as datetime
dfmd.index = dfmd.index.tz_localize(None)  # Ensure index is timezone naive
plt.figure(figsize=(15, 5))
plt.plot(dfmd.index,dfmd['ec_LE_H'],label='EC', marker='o', linestyle='-')
plt.plot(dfmd.index,dfmd['li_LE_H'], label='LI', marker='_', linestyle='--')
plt.xlabel('Time')
plt.legend()
plt.grid()
plt.show()
In [18]:
plt.figure(figsize=(15, 5))
plt.plot(dfmd.index,dfmd['ec_ET'],label='EC', marker='o', linestyle='-')
plt.plot(dfmd.index,dfmd['li_ET'], label='LI', marker='_', linestyle='--')
plt.xlabel('Time')
plt.legend()
plt.grid()
plt.show()
In [19]:
plt.figure(figsize=(15, 5))
plt.plot(dfmd.index,dfmd['ec_ET'],label='EC', marker='o', linestyle='-')
plt.plot(dfmd.index,dfmd['li_ET'], label='LI', marker='_', linestyle='--')
plt.xlabel('Time')
plt.legend()
plt.grid()
plt.show()
In [20]:
plt.figure(figsize=(15, 5))
dfmd['et_per']=100*(dfmd['li_ET']-dfmd['ec_ET'])/dfmd['ec_ET']
plt.plot(dfmd.index,dfmd['et_per'], marker='o', linestyle='-')
#plt.plot(dfmd.index,dfmd['li_ET'], label='LI', marker='_', linestyle='--')
plt.xlabel('Time')
#plt.legend()

plt.ylim(-150.,0)
plt.grid()
plt.show()
In [21]:
plt.figure(figsize=(15, 5))
dfmd['LE_per']=100*(dfmd['li_LE']-dfmd['ec_LE'])/dfmd['ec_LE']
plt.plot(dfmd.index,dfmd['LE_per'], marker='o', linestyle='-')
plt.xlabel('Time')

plt.ylim(-150.,50)
plt.grid()
plt.show()
In [22]:
plt.figure(figsize=(15, 5))
dfmd['H_per']=100*(dfmd['li_H']-dfmd['ec_H'])/dfmd['ec_H']
plt.plot(dfmd.index,dfmd['H_per'], marker='o', linestyle='-')
plt.xlabel('Time')

plt.ylim(-150.,50)
plt.grid()
plt.show()
In [23]:
dfmd = df_merged.resample('D').mean()
dfmd.index = pd.to_datetime(dfmd.index)  # Ensure index remains as datetime
dfmd.index = dfmd.index.tz_localize(None)  # Ensure index is timezone naive
plt.figure(figsize=(15, 5))
plt.plot(dfmd.index,dfmd['ec_LE'],label='EC', marker='o', linestyle='-')
plt.plot(dfmd.index,dfmd['li_LE'], label='LI', marker='_', linestyle='--')
plt.xlabel('Time')
plt.legend()
plt.grid()
plt.show()
In [24]:
dfmd = df_merged.resample('D').mean()
dfmd.index = pd.to_datetime(dfmd.index)  # Ensure index remains as datetime
dfmd.index = dfmd.index.tz_localize(None)  # Ensure index is timezone naive
plt.figure(figsize=(15, 5))
plt.plot(dfmd.index,dfmd['ec_H'],label='EC', marker='o', linestyle='-')
plt.plot(dfmd.index,dfmd['li_H'], label='LI', marker='_', linestyle='--')
plt.xlabel('Time')
plt.legend()
plt.grid()
plt.show()
In [25]:
#Merge dfmd with dfcd on matching index times
df_pya_pyf = dfmd.merge(dfcd, left_index=True, right_index=True, how='inner')
df_pya_pyf['li_C']=df_pya_pyf['li_LE_H']/df_pya_pyf['RNG']
df_pya_pyf['ec_C']=df_pya_pyf['ec_LE_H']/df_pya_pyf['RNG']
plt.figure(figsize=(10, 5))

plt.plot(df_pya_pyf.index,df_pya_pyf['ec_C'], marker='o', linestyle='-')
plt.plot(df_pya_pyf.index,df_pya_pyf['li_C'], marker='+', linestyle='--')
plt.xlabel('Time)')

plt.ylim(-0.5,1.0)
#plt.legend()
plt.grid()
plt.show()
In [26]:
plt.figure(figsize=(10, 5))
plt.plot(dfmd.index,dfmd['ec_LE_H']/df_pya_pyf['ec_C'],label='EC', marker='o', linestyle='-')
plt.plot(dfmd.index,dfmd['li_LE_H']/df_pya_pyf['li_C'], label='LI', marker='_', linestyle='--')
plt.xlabel('Time)')
#plt.legend()
plt.grid()
plt.show()
In [27]:
plt.figure(figsize=(10, 5))
plt.plot(dfmd.index,dfmd['ec_ET'],label='EC', marker='o', linestyle='-')
plt.plot(dfmd.index,dfmd['li_ET']/df_pya_pyf['li_C'], label='LI', marker='_', linestyle='--')
plt.xlabel('Time)')
plt.ylim(-.05,0.25)
#plt.legend()
plt.grid()
plt.show()
In [28]:
# Convert latent heat energy (LE in W/m^2) to evaporation (mm/30min)
L = 2.45e6  # Latent heat of vaporization (J/kg)
rho_w = 1000  # Density of water (kg/m^3)
# convert from m to mm by 1000.
seconds_per_30min = 30*60  

df_merged['li_EC'] = 1000.*df_merged['li_LE'] * seconds_per_30min / (L * rho_w)
df_merged['ec_EC'] = 1000.*df_merged['ec_LE'] * seconds_per_30min / (L * rho_w)

df_pya_pyf['date'] = df_pya_pyf.index.date  # Extract date from daily dataframe
df_merged['date'] = df_merged.index.date  # Extract date from timestamp
In [29]:
# scale every 30 min by the daily correction factor
#'dattim' is the 30 min times
df_merged['dattim'] = df_merged.index
df_merged = df_merged.merge(df_pya_pyf[['ec_C','li_C','date']], on='date', how='left')
df_merged['ec_EC_cor'] = df_merged['ec_EC'] / df_merged['ec_C']
df_merged['li_EC_cor'] = df_merged['li_EC'] / df_merged['li_C']
df_merged = df_merged.set_index('dattim')
df_merged.drop(columns=['date'], inplace=True)  # Remove temporary column
#df_merged.drop(columns=['dattim'], inplace=True)  # Remove temporary column
In [30]:
plt.figure(figsize=(10, 5))
df_merged.index = pd.to_datetime(df_merged.index)  # Ensure index remains as datetime
df_merged.index = df_merged.index.tz_localize(None)  # Ensure index is timezone naive
plt.plot(df_merged.index,df_merged['ec_EC_cor'],label='ec_EC_cor', marker='o', linestyle='-')
plt.plot(df_merged.index,df_merged['li_EC_cor'], label='li_EC_cor', marker='_', linestyle='--')
plt.plot(df_merged.index,df_merged['ec_ET'],label='ec_ET', marker='o', color='cyan',linestyle='-')
plt.plot(df_merged.index,df_merged['li_ET'], label='li_ET', marker='_', color='red',linestyle='--')
plt.xlabel('Time)')
plt.ylim(-1,2)
plt.legend()
plt.grid()
plt.show()
In [31]:
plt.figure(figsize=(10, 5))

#plt.plot(df_merged.index,df_merged['ec_EC_cor'],label='ec_EC_cor',  linestyle='-')
plt.plot(df_merged.index,df_merged['li_EC_cor'], label='li_EC_cor', linestyle='--')
plt.plot(df_merged.index,df_merged['ec_ET'],label='ec_ET',  color='cyan',linestyle='-')
#plt.plot(df_merged.index,df_merged['li_ET'], label='li_ET', marker='_', color='red',linestyle='--')
plt.xlabel('Time)')
plt.ylim(-1,2)
plt.legend()
plt.grid()
plt.show()
In [32]:
plt.figure(figsize=(10, 5))
#df_merged.index = pd.to_datetime(df_merged.index)  # Ensure index remains as datetime
#df_merged.index = df_merged.index.tz_localize(None)  # Ensure index is timezone naive

# Filter for February and March
df_filtered = df_merged.loc[df_merged.index.month.isin([2, 3])]

# df_filtered = df_merged.loc[(df_merged.index.month == 2) & (df_merged.index.day == 26)]
plt.scatter(df_filtered['ec_ET'],df_filtered['li_EC_cor'], marker='o', s=1)
plt.scatter(df_filtered['ec_ET'],df_filtered['li_EC'], color='red', s=1)
plt.ylim(-.1,0.5)
plt.grid()
plt.show()
In [33]:
plt.figure(figsize=(10, 5))
#df_merged.index = pd.to_datetime(df_merged.index)  # Ensure index remains as datetime
#df_merged.index = df_merged.index.tz_localize(None)  # Ensure index is timezone naive
plt.scatter(df_merged['ec_ET'],df_merged['li_ET'], marker='o',s=1)
plt.ylim(-.1,0.5)
plt.grid()
plt.show()
In [34]:
# Convert wind direction to radians for polar plot
wind_dir_rad = np.radians(df_pya_pyf['WindDir'])

# Create polar plot
fig, ax = plt.subplots(figsize=(8, 8), subplot_kw={'projection': 'polar'})

# Plot footprints at different percentiles
ax.scatter(wind_dir_rad, df_pya_pyf['ec_FETCH_40'], s=5,  color='b',alpha=0.5)

ax.scatter(wind_dir_rad, df_pya_pyf['ec_FETCH_55'], s=5,  color='cyan',alpha=0.5)


ax.scatter(wind_dir_rad, df_pya_pyf['ec_FETCH_90'], s=5,  color='orange',alpha=0.5)


ax.scatter(wind_dir_rad, df_pya_pyf['ec_FETCH_MAX'], s=5, color='red', alpha=0.5)



# Labels and legend
ax.set_theta_zero_location('N')  # Set 0 degrees at the top (North)
ax.set_theta_direction(-1)  # Set direction to clockwise
ax.set_xlabel("Wind Direction (degrees)")
ax.set_ylabel("Footprint (meters)")
ax.legend(loc="upper right", bbox_to_anchor=(1.1, 1.1))

plt.title("Wind Direction v Fetch 55")
plt.show()

No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.