loone-data-prep 1.1.1__tar.gz → 1.2.0__tar.gz

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: loone_data_prep
-Version: 1.1.1
+Version: 1.2.0
 Summary: Prepare data to run the LOONE model.
 Author-email: Osama Tarabih <osamatarabih@usf.edu>
 Maintainer-email: Michael Souffront <msouffront@aquaveo.com>, James Dolinar <jdolinar@aquaveo.com>
@@ -24,7 +24,7 @@ Requires-Dist: numpy<2
 Requires-Dist: pandas
 Requires-Dist: scipy
 Requires-Dist: geoglows>=2.0.0
-Requires-Dist: herbie
+Requires-Dist: herbie-data[extras]
 Requires-Dist: openmeteo_requests
 Requires-Dist: requests_cache
 Requires-Dist: retry-requests

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/loone_data_prep/GEOGLOWS_LOONE_DATA_PREP.py

@@ -44,7 +44,7 @@ def main(input_dir: str, output_dir: str, ensemble_number: str) -> None:  # , hi
     # Read LO Average Stage (ft)
     LO_Stage = pd.read_csv(f"{input_dir}/LO_Stage.csv")
     # Create Column (EOD Stg(ft, NGVD)) in File (SFWMM_Daily_Outputs)
-    LO_Stage = DF_Date_Range(LO_Stage, M3_Yr, M3_M, M3_D, En_Yr, En_M, En_D)
+    # LO_Stage = DF_Date_Range(LO_Stage, M3_Yr, M3_M, M3_D, En_Yr, En_M, En_D)
     LO_Stage.index = LO_Stage["date"]
     # Calculate average
     if "Average_Stage" not in LO_Stage.columns:
@@ -457,6 +457,7 @@ def main(input_dir: str, output_dir: str, ensemble_number: str) -> None:  # , hi
     LOWS["LZ40WS"] = LZ40WS["LZ40_WNDS_MPH"]
     LOWS = LOWS.set_index("date")
     LOWS["LO_Avg_WS_MPH"] = LOWS.mean(axis=1)
+    LOWS = LOWS.resample("D").mean()
     LOWS.to_csv(f"{output_dir}/LOWS_predicted.csv")
 
     # # RFVol acft
@@ -592,7 +593,7 @@ def main(input_dir: str, output_dir: str, ensemble_number: str) -> None:  # , hi
 
     # Write Data into csv files
     # write Avg Stage (ft, m) Storage (acft, m3) SA (acres) to csv
-    LO_Stg_Sto_SA_df.to_csv(f"{output_dir}/Average_LO_Storage_3MLag_{ensemble_number}.csv", index=False)
+    # LO_Stg_Sto_SA_df.to_csv(f"{output_dir}/Average_LO_Storage_3MLag_{ensemble_number}.csv", index=False)
     # Write S65 TP concentrations (mg/L)
     S65_total_TP.to_csv(f"{output_dir}/S65_TP_3MLag_{ensemble_number}.csv", index=False)
     # TP External Loads 3 Months Lag (mg)

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/loone_data_prep/LOONE_DATA_PREP.py

@@ -34,7 +34,7 @@ def main(input_dir: str, output_dir: str) -> None:
     # Read LO Average Stage (ft)
     LO_Stage = pd.read_csv(f'{input_dir}/LO_Stage.csv')
     # Create Column (EOD Stg(ft, NGVD)) in File (SFWMM_Daily_Outputs)
-    LO_Stage = DF_Date_Range(LO_Stage, M3_Yr, M3_M, M3_D, En_Yr, En_M, En_D)
+    # LO_Stage = DF_Date_Range(LO_Stage, M3_Yr, M3_M, M3_D, En_Yr, En_M, En_D)
     # Calculate average
     if "Average_Stage" not in LO_Stage.columns:
         LO_Stage = LO_Stage.loc[:, ~LO_Stage.columns.str.contains('^Unnamed')]

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/loone_data_prep/flow_data/get_forecast_flows.py

@@ -392,10 +392,10 @@ def main(
 
     # Get the flow data for each station
     stations_inflow_by_comid = {
-        750072741: "S65E_S",   # TODO: Should this be S65E_total or S65E_S? - this is a station we definitely want
-        750069782: "S84_S",        # 
-        # 750053211: "S129_C",       # TODO: Should this be S129_C or S129_PMP_P? - Also right now it is all 0s
-        # 750035446: "S133_P",       # TODO: Should this be S133_P or S133_C? - Also right now it is all 0s
+        750072741: "S65E_S", 
+        750069782: "S84_S",       
+        # 750053211: "S129_C",
+        # 750035446: "S133_P",
         750064453: "S154_C",       # This is primarily 0s
     }
 
@@ -444,8 +444,7 @@ def main(
     for reach_id in MATCHED_IDS:
         stations_matched_by_comid = {
             750068601: "S71_S",
-            750052624: "S135_C",       # TODO: Should this be S135_C or S135_P?
-           #  750052624: "S308",       # NOTE: Same COMID as S135 — only one key allowed!
+            750052624: "S135_C",
             750053213: "FISHP",
             750038416: "S77_S",
             750050259: "S79_TOT",
@@ -454,7 +453,7 @@ def main(
             750051428: "S49_S",
             # 750038427: "S40",
             750057357: "S191_S",
-            750028935: "S127_C", #TODO: Should this be S127_C or S127_P?
+            750028935: "S127_C",
         }
 
         station_ensembles = get_flow_forecast_ensembles(

loone_data_prep-1.2.0/loone_data_prep/forecast_scripts/Chla_merged.py

@@ -0,0 +1,27 @@
+import pandas as pd
+def loads_predicted(input_dir, output_dir):
+    """
+    Calculate Chlorophyll-a loads based on inflows and chlorophyll-a data.
+    
+    input_dir: Directory where the input files are located.
+    output_dir: Directory where the output files will be saved.
+    St_Yr, St_M, St_D: Start date (year, month, day).
+    En_Yr, En_M, En_D: End date (year, month, day).
+    """
+    
+    # Read forecast inflow file
+    # TODO: Should this be an average/median of all of the ensembles? worst case?
+    Flow_df = pd.read_csv(f"{input_dir}/geoglows_flow_df_ens_01_predicted.csv")
+    Flow_df['date'] = pd.to_datetime(Flow_df['date'])
+
+    
+    # Read S65E Chlorophyll-a data
+    S65E_Chla = pd.read_csv(f'{output_dir}/S65E_Chla_Merged_forecast.csv')
+    S65E_Chla['date'] = pd.to_datetime(S65E_Chla['date'])  # Ensure date column is datetime
+    # Merge on date
+    merged = pd.merge(Flow_df[['date', 'Inflows']], S65E_Chla[['date', 'Data']], on='date', how='inner')
+    # Calculate Chlorophyll-a loads
+    merged['Chla_Loads'] = merged['Inflows'] * merged['Data']
+    # Save results
+    Chla_Loads_In = merged[['date', 'Chla_Loads']]
+    Chla_Loads_In.to_csv(f'{output_dir}/Chla_Loads_In_forecast.csv', index=False)

loone_data_prep-1.2.0/loone_data_prep/forecast_scripts/get_Chla_predicted.py

@@ -0,0 +1,109 @@
+import os
+import pandas as pd
+import datetime
+from loone_data_prep.utils import get_synthetic_data
+
+def get_Chla_predicted(input_dir, output_dir):
+    """
+    input_dir: Directory where the input files are located.
+    output_dir: Directory where the output files will be saved.
+    """
+    # Read forecast inflow file and get overall date range
+    # TODO: Should this be an average/median of all of the ensembles? worst case?
+    Q_in = pd.read_csv(os.path.join(input_dir, 'LO_Inflows_BK_forecast_01.csv'))
+    Q_in['date'] = pd.to_datetime(Q_in['date'])
+    date_start = Q_in['date'].min()
+    date_end = Q_in['date'].max()
+
+    # Define stations
+    stations = {
+        "L001": True,
+        "L004": True,
+        "L005": True,
+        "L006": True,
+        "L007": True,
+        "L008": True,
+        "LZ40": True
+    }
+
+    def load_and_check_forecast(station, suffix, start_date, end_date, forecast_suffix="_forecast"):
+        fname = f"water_quality_{station}_CHLOROPHYLL-A{suffix}.csv"
+        fpath = os.path.join(input_dir, fname)
+        df_full = pd.read_csv(fpath).drop(columns=["days"], errors="ignore")
+        df_full['date'] = pd.to_datetime(df_full['date'])
+        # Rename the specific column if it exists
+        possible_cols = [
+            f"{station}_CHLOROPHYLL-A, CORRECTED_ug/L",
+            f"{station}_CHLOROPHYLL-A(LC)_ug/L"
+        ]
+
+        original_col_name = None
+        for col in possible_cols:
+            if col in df_full.columns:
+                df_full.rename(columns={col: "Data"}, inplace=True)
+                original_col_name = col
+                break
+
+        # Filter df to only rows between start_date and end_date
+        df_filtered = df_full[(df_full['date'] >= start_date) & (df_full['date'] <= end_date)]
+
+        # Check if full date range is covered; if not, fill with synthetic data
+        missing_dates = pd.date_range(start_date, end_date).difference(df_filtered['date'])
+        if len(missing_dates) > 0:
+            # Pass the original full historical df_full to get_synthetic_data, along with the forecast start_date
+            synthetic_df = get_synthetic_data(start_date, df_full)
+
+            # Rename "Data" back to original column name before saving
+            if original_col_name is not None:
+                synthetic_df.rename(columns={"Data": original_col_name}, inplace=True)
+
+            # Save synthetic forecast file
+            forecast_fname = f"water_quality_{station}_CHLOROPHYLL-A{suffix}{forecast_suffix}.csv"
+            synthetic_df.to_csv(os.path.join(input_dir, forecast_fname), index=False)
+
+            return synthetic_df
+
+        return df_filtered
+
+    # Load data for all stations and both suffix types
+    chla_data = {}
+    chla_data_lc = {}
+
+    for station in stations:
+        chla_data[station] = load_and_check_forecast(station, ", CORRECTED", date_start, date_end)
+        chla_data_lc[station] = load_and_check_forecast(station, "(LC)", date_start, date_end)
+
+    # Merge function
+    def merge_chla_sources(chla_dict):
+        merged = None
+        for df in chla_dict.values():
+            if merged is None:
+                merged = df
+            else:
+                merged = pd.merge(merged, df, on="date", how="left")
+            merged = merged.loc[:, ~merged.columns.str.startswith("Unnamed")]
+        return merged
+
+    # Calculate aggregates
+    def calculate_chla_aggregates(df, suffix=""):
+        df = df.set_index("date")
+        df["Mean_Chla"] = df.mean(axis=1)
+        df["Chla_North"] = df[[col for col in df.columns if any(site in col for site in ["L001", "L005", "L008"])]].mean(axis=1)
+        df["Chla_South"] = df[[col for col in df.columns if any(site in col for site in ["L004", "L006", "L007", "L008", "LZ40"])]].mean(axis=1)
+        df = df.reset_index()
+        return df[["date", "Mean_Chla", "Chla_North", "Chla_South"]].rename(
+            columns={"Mean_Chla": f"Chla{suffix}", "Chla_North": f"Chla_N{suffix}", "Chla_South": f"Chla_S{suffix}"}
+        )
+
+    # Process and merge
+    LO_Chla = calculate_chla_aggregates(merge_chla_sources(chla_data))
+    LO_Chla_LC = calculate_chla_aggregates(merge_chla_sources(chla_data_lc))
+
+    # Merge the two dataframes (no date slicing here since all are limited by Q_in dates)
+    LO_Chla_Merge = pd.concat([LO_Chla, LO_Chla_LC]).reset_index(drop=True)
+
+    # Export
+    LO_Chla_Merge.to_csv(os.path.join(output_dir, "LO_Chla_Obs_predicted.csv"), index=False)
+    LO_Chla_Merge[["date", "Chla_N"]].rename(columns={"Chla_N": "Chla"}).to_csv(os.path.join(output_dir, "N_Merged_Chla_predicted.csv"), index=False)
+    LO_Chla_Merge[["date", "Chla_S"]].rename(columns={"Chla_S": "Chla"}).to_csv(os.path.join(output_dir, "S_Merged_Chla_predicted.csv"), index=False)
+    return

loone_data_prep-1.2.0/loone_data_prep/forecast_scripts/get_NO_Loads_predicted.py

@@ -0,0 +1,97 @@
+import os
+import pandas as pd
+import datetime
+from loone_data_prep.utils import photo_period, get_synthetic_data
+
+def get_NO_Loads_predicted(input_dir, output_dir):
+    """
+    input_dir: Directory where the input files are located.
+    output_dir: Directory where the output files will be saved.
+    This function reads the forecast inflow file, retrieves nitrate data for specified stations,
+    """
+    # TODO: Should this be an average/median of all of the ensembles? worst case?
+    Q_in = pd.read_csv(os.path.join(input_dir, 'LO_Inflows_BK_forecast_01.csv'))
+
+    datetime_str = Q_in['date'].iloc[0]
+    date_start = datetime.datetime.strptime(datetime_str, '%Y-%m-%d')
+    stations = [
+        "S65E", "S71", "S72", "S84", "S127", "S133", 
+        "S154", "S191", "S308C", "FECSR78", "CULV10A", "S4"
+    ]
+
+    station_alias_map = {
+        "S65E": "S65_NO",
+        "S71": "S71_NO",
+        "S72": "S72_NO",
+        "S84": "S84_NO",
+        "S127": "S127_NO",
+        "S133": "S133_NO",
+        "S154": "S154_NO",
+        "S191": "S191_NO",
+        "S308C": "S308_NO",
+        "FECSR78": "FISHP_NO",
+        "CULV10A": "L8_NO",
+        "S4": "S4_NO"
+    }
+
+    NO_list = {}
+    NO_names = []
+
+    for station, alias in station_alias_map.items():
+        filename = f'water_quality_{station}_NITRATE+NITRITE-N_Interpolated.csv'
+        file_path = os.path.join(input_dir, filename)
+        
+        try:
+            df = pd.read_csv(file_path)
+        except FileNotFoundError:
+            print(f"{filename} not found.")
+            continue
+
+        # Forecast if needed
+        if datetime_str not in df['date'].values:
+            df = get_synthetic_data(date_start, df)
+            df.to_csv(os.path.join(input_dir, f'water_quality_{station}_NITRATE+NITRITE-N_Interpolated_forecast.csv'), index=False)
+
+        NO_list[alias] = df
+        NO_names.append(alias)
+
+    # date_NO = pd.date_range(start='1/1/2008', end='3/31/2023', freq='D')
+    # Because of the flow df, I think this will be generated for every single ensemble member
+    for ensemble in range(1, 52):
+        Flow_df =pd.read_csv(f"{input_dir}/geoglows_flow_df_ens_{ensemble:02d}_predicted.csv")
+        Flow_df['date'] = pd.to_datetime(Flow_df['date'])
+
+        # Use Flow_df as the base for merging nitrate data
+        NO_df = Flow_df[['date']].copy()
+
+        for name in NO_names:
+            y = NO_list[name]
+            y.rename(columns={y.columns[-1]: name}, inplace=True)
+            NO_df = pd.merge(NO_df, y[['date', name]], on='date', how='left')
+
+        # Flow_df = DF_Date_Range(Flow_df, St_Yr, St_M, St_D, En_Yr, En_M, En_D)
+
+        NO_df['date'] = pd.to_datetime(NO_df['date'])
+
+        merged = pd.merge(NO_df, Flow_df, on='date', how='inner')
+
+        NO_Loads_In = merged[['date']].copy()
+
+        # Compute individual loads (edit flow variable names if needed)
+        NO_Loads_In['S65_NO_Ld'] = merged['S65_Q'] * merged['S65_NO'] * 1000
+        NO_Loads_In['S71_NO_Ld'] = merged['S71_Q'] * merged['S71_NO'] * 1000
+        # NO_Loads_In['S72_NO_Ld'] = merged['S72_Q'] * merged['S72_NO'] * 1000 # No RFS forecast data
+        NO_Loads_In['S84_NO_Ld'] = merged['S84_Q'] * merged['S84_NO'] * 1000
+        # NO_Loads_In['S127_NO_Ld'] = merged['S127_In'] * merged['S127_NO'] * 1000 # This should be in here, figure out where it went
+        NO_Loads_In['S133_NO_Ld'] = merged['S133_P_Q'] * merged['S133_NO'] * 1000
+        NO_Loads_In['S154_NO_Ld'] = merged['S154_Q'] * merged['S154_NO'] * 1000
+        # NO_Loads_In['S191_NO_Ld'] = merged['S191_Q'] * merged['S191_NO'] * 1000 #This should be in here, figure out where it went
+        NO_Loads_In['S308_NO_Ld'] = merged['S308_In'] * merged['S308_NO'] * 1000
+        NO_Loads_In['FISHP_NO_Ld'] = merged['FISHP_Q'] * merged['FISHP_NO'] * 1000
+        # NO_Loads_In['L8_NO_Ld'] = merged['L8_In'] * merged['L8_NO'] * 1000 # No RFS forecast data
+        # NO_Loads_In['S4_NO_Ld'] = merged['S4_P_Q'] * merged['S4_NO'] * 1000 # No RFS Forecast data
+
+        NO_Loads_In['External_NO_Ld_mg'] = NO_Loads_In.sum(axis=1, numeric_only=True)
+
+        NO_Loads_In.to_csv(f'{output_dir}/LO_External_Loadings_NO_ens_{ensemble:02d}_predicted.csv', index=False)
+    return

loone_data_prep-1.2.0/loone_data_prep/forecast_scripts/loone_q_predict.py

@@ -0,0 +1,159 @@
+import os
+import pandas as pd
+from datetime import datetime, timedelta
+
+def generate_historical_predictions(workspace, forecast_days=16):
+    """
+    Generate predictions for the next `forecast_days` days using historical daily averages
+    from the same calendar dates across previous years. Includes all the files for loone_q.
+    
+    Args:
+    workspace : str
+        Path to the folder containing the CSV files.
+
+    forecast_days : int
+        Number of future days to predict (default = 16).
+
+    """
+
+    file_list = [
+        "Estuary_needs_water_Input.csv",
+        "Multi_Seasonal_LONINO.csv", 
+        "Seasonal_LONINO.csv",
+        "SFWMM_Daily_Outputs.csv",
+        "Water_dmd.csv",
+        "EAA_MIA_RUNOFF_Inputs.csv",
+    ]
+
+    possible_date_cols = ['date', 'Date']
+    today = datetime.today()
+    current_year = today.year
+
+    for filename in file_list:
+        path = os.path.join(workspace, filename)
+        
+        try:
+            df = pd.read_csv(path)
+        except Exception as e:
+            print(f"Could not read {filename}. Error: {e}")
+            continue
+        
+        if filename in ["Multi_Seasonal_LONINO.csv", "Seasonal_LONINO.csv"]:
+            if "Year" not in df.columns:
+                print(f"No 'Year' column in {filename}. Skipping.")
+                continue
+
+            # ➤ Skip if current year already exists
+            if current_year in df["Year"].values:
+                print(f"{current_year} already present in {filename}. No changes made.")
+                continue
+
+            # Otherwise calculate averages and append
+            month_cols = [col for col in df.columns if col != "Year"]
+            monthly_means = df[month_cols].mean()
+
+            new_row = {"Year": current_year}
+            new_row.update(monthly_means.to_dict())
+
+            updated_df = pd.concat(
+                [df, pd.DataFrame([new_row])],
+                ignore_index=True
+            )
+            output_name = filename.replace(".csv", f"_forecast.csv")
+            output_path = os.path.join(workspace, output_name)
+            updated_df.to_csv(output_path, index=False)
+            print(f"Appended {current_year} row and saved to {output_path}")
+            continue
+
+        # Identify date column
+        date_col = None
+        for col in df.columns:
+            if col in possible_date_cols:
+                date_col = col
+                break
+
+        if date_col is None:
+            print(f"Could not detect date column in {filename}. Skipping.")
+            continue
+
+        # Parse dates
+        if filename in ["SFWMM_Daily_Outputs.csv", "Water_dmd.csv"]:
+            df[date_col] = pd.to_datetime(
+                df[date_col],
+                format="%d-%b-%y",
+                errors="coerce"
+            )
+        else:
+            df[date_col] = pd.to_datetime(df[date_col], errors='coerce')
+        df = df.dropna(subset=[date_col])
+        df["month_day"] = df[date_col].dt.strftime("%m-%d")
+
+        predictions_list = []
+
+        # Check if special handling is needed for the boolean file
+        if filename == "Estuary_needs_water_Input.csv":
+            bool_col = "Estuary Needs Water?"
+
+            if bool_col not in df.columns:
+                print(f"Column '{bool_col}' not found in {filename}. Skipping.")
+                continue
+
+            # Convert string "True"/"False" to boolean if necessary
+            if df[bool_col].dtype == object:
+                df[bool_col] = df[bool_col].map({"True": True, "False": False}).fillna(df[bool_col])
+
+            # Compute mode (most frequent value) for each day for each boolean column
+            mode_series = df.groupby("month_day")[bool_col].agg(
+                lambda x: x.mode().iloc[0] if not x.mode().empty else None
+            )
+            for i in range(1, forecast_days + 1):
+                future_date = today + timedelta(days=i)
+                mmdd = future_date.strftime("%m-%d")
+
+                if mmdd in mode_series.index:
+                    pred_value = mode_series.loc[mmdd]
+                else:
+                    print(f"No historical data for {mmdd} in {filename}. Skipping that day.")
+                    pred_value = None
+
+                predictions_list.append({
+                    date_col: future_date,
+                    bool_col: pred_value
+                })
+
+            pred_df = pd.DataFrame(predictions_list)
+            pred_df = pred_df[[date_col, bool_col]]
+
+        else:
+            # Numeric file handling
+            numeric_cols = df.select_dtypes(include='number').columns.tolist()
+            if not numeric_cols:
+                print(f"No numeric columns in {filename}. Skipping.")
+                continue
+
+            historical_means = df.groupby("month_day")[numeric_cols].mean()
+
+            for i in range(0, forecast_days + 1):
+                future_date = (today + timedelta(days=i)).date()
+                mmdd = future_date.strftime("%m-%d")
+                if mmdd in historical_means.index:
+                    row = historical_means.loc[mmdd].copy()
+                    row[date_col] = future_date
+                    predictions_list.append(row)
+                else:
+                    print(f"No historical data for {mmdd} in {filename}. Skipping that day.")
+
+            if predictions_list:
+                pred_df = pd.DataFrame(predictions_list)
+                pred_df = pred_df[[date_col] + [col for col in pred_df.columns if col != date_col]]
+            else:
+                print(f"No predictions generated for {filename}.")
+                continue
+
+        # Save predictions
+        output_name = filename.replace(".csv", f"_forecast.csv")
+        output_path = os.path.join(workspace, output_name)
+        pred_df.to_csv(output_path, index=False)
+        print(f"Predictions saved to {output_path}")
+
+    return

loone_data_prep-1.2.0/loone_data_prep/forecast_scripts/loone_wq_predict.py

@@ -0,0 +1,71 @@
+import pandas as pd
+import os
+
+def create_forecasts(workspace):
+    """
+    Reads the four specified CSV files from `workspace`,
+    creates forecast versions using historical daily averages,
+    and writes new CSV files into the same folder.
+    
+    The forecast always starts today and goes 16 days forward.
+    """
+
+    # List of filenames
+    files = [
+        'N_OP.csv',
+        'S_OP.csv',
+        'N_DIN.csv',
+        'S_DIN.csv',
+        'LO_DO_Clean_daily.csv'
+    ]
+
+    def forecast_df(df, date_column='date'):
+        # Parse dates
+        df[date_column] = pd.to_datetime(df[date_column])
+
+        # Add month and day columns
+        df['month'] = df[date_column].dt.month
+        df['day'] = df[date_column].dt.day
+
+        # Identify numeric columns to forecast
+        value_columns = df.columns.difference([date_column, 'month', 'day'])
+
+        # Compute historical averages
+        avg = df.groupby(['month', 'day'])[value_columns].mean().reset_index()
+
+        # Create forecast dates: today + next 15 days
+        forecast_dates = pd.date_range(
+            start=pd.Timestamp.today().normalize(),
+            periods=16,
+            freq='D'
+        )
+
+        forecast_df = pd.DataFrame({date_column: forecast_dates})
+        forecast_df['month'] = forecast_df[date_column].dt.month
+        forecast_df['day'] = forecast_df[date_column].dt.day
+
+        # Merge with historical averages
+        forecast_df = forecast_df.merge(avg, on=['month', 'day'], how='left')
+
+        # Drop helper columns
+        forecast_df.drop(columns=['month', 'day'], inplace=True)
+
+        return forecast_df
+
+    # Process each file
+    for filename in files:
+        # Read file
+        file_path = os.path.join(workspace, filename)
+        df = pd.read_csv(file_path)
+
+        # Build forecast
+        forecast = forecast_df(df, date_column='date')
+
+        # Save new file
+        forecast_filename = filename.replace('.csv', '_forecast.csv')
+        forecast_path = os.path.join(workspace, forecast_filename)
+        forecast.to_csv(forecast_path, index=False)
+
+    print("Forecast files created successfully.")
+
+

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/loone_data_prep/forecast_scripts/trib_cond.py

@@ -50,7 +50,15 @@ def create_trib_cond (weather_data, net_inflows, main_tributary, PI, output, ens
 
     # Calculate NetRF and NetInf
     Trib_Cond_Wkly['NetRF'] = Net_RF_Weekly['tp_corrected'].values - Net_RF_Weekly['evapotranspiration'].values
-    Trib_Cond_Wkly['NetInf'] = Net_Inflow_Weekly['Net_Inflows'].values
+    # First, reset index so that 'date' becomes a column in Net_Inflow_Weekly
+    Net_Inflow_Weekly_reset = Net_Inflow_Weekly.reset_index()
+
+    # Merge the dataframes on 'date'
+    Trib_Cond_Wkly = Trib_Cond_Wkly.merge(Net_Inflow_Weekly_reset[['date', 'Net_Inflows']], on='date', how='left')
+
+    # Now Trib_Cond_Wkly will have a new 'Net_Inflows' column aligned by date
+    Trib_Cond_Wkly.rename(columns={'Net_Inflows': 'NetInf'}, inplace=True)
+
 
     # Select only the desired ensemble column and rename it
     S65E_selected = S65E_Weekly[[ensemble_col]].rename(columns={ensemble_col: "S65E"})

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/loone_data_prep/utils.py

@@ -466,12 +466,15 @@ def wind_induced_waves(
     lo_stage_in: str = "LO_Stg_Sto_SA_2008-2023.csv",
     wind_shear_stress_out: str = "WindShearStress.csv",
     current_shear_stress_out: str = "Current_ShearStress.csv",
+    forecast: bool = False,
 ):
     # Read Mean Wind Speed in LO
     LO_WS = pd.read_csv(os.path.join(f"{input_dir}/", wind_speed_in))
     LO_WS["WS_mps"] = LO_WS["LO_Avg_WS_MPH"] * 0.44704  # MPH to m/s
     # Read LO Stage to consider water depth changes
     LO_Stage = pd.read_csv(os.path.join(f"{input_dir}/", lo_stage_in))
+    if forecast:
+        LO_Stage["Stage_ft"] = LO_Stage["Stage"].astype(float)
     LO_Stage["Stage_m"] = LO_Stage["Stage_ft"] * 0.3048
     Bottom_Elev = 0.5  # m (Karl E. Havens • Alan D. Steinman 2013)
     LO_Wd = LO_Stage["Stage_m"] - Bottom_Elev
@@ -998,7 +1001,15 @@ def get_synthetic_data(date_start: str, df: pd.DataFrame):
     
     # Group by the month and day, then calculate the average for each group
     average_values = filtered_data.groupby('month_day')['Data'].mean()
-    
+    # Interpolate in case there are missing values:
+    start_date = pd.to_datetime('2001-' + start_month_day)
+    end_date = pd.to_datetime('2001-' + end_month_day)
+
+    full_dates = pd.date_range(start=start_date, end=end_date)
+    full_index = full_dates.strftime('%m-%d')
+
+    average_values = average_values.reindex(full_index)
+    average_values = average_values.interpolate(method='linear')
     average_values_df = pd.DataFrame({
         'date': pd.date_range(start=date_start, end=date_end),
         'Data': average_values.values

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/loone_data_prep.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: loone_data_prep
-Version: 1.1.1
+Version: 1.2.0
 Summary: Prepare data to run the LOONE model.
 Author-email: Osama Tarabih <osamatarabih@usf.edu>
 Maintainer-email: Michael Souffront <msouffront@aquaveo.com>, James Dolinar <jdolinar@aquaveo.com>
@@ -24,7 +24,7 @@ Requires-Dist: numpy<2
 Requires-Dist: pandas
 Requires-Dist: scipy
 Requires-Dist: geoglows>=2.0.0
-Requires-Dist: herbie
+Requires-Dist: herbie-data[extras]
 Requires-Dist: openmeteo_requests
 Requires-Dist: requests_cache
 Requires-Dist: retry-requests

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/loone_data_prep.egg-info/SOURCES.txt

@@ -18,8 +18,13 @@ loone_data_prep/flow_data/get_forecast_flows.py
 loone_data_prep/flow_data/get_inflows.py
 loone_data_prep/flow_data/get_outflows.py
 loone_data_prep/flow_data/hydro.py
+loone_data_prep/forecast_scripts/Chla_merged.py
 loone_data_prep/forecast_scripts/create_forecast_LOWs.py
 loone_data_prep/forecast_scripts/forecast_stages.py
+loone_data_prep/forecast_scripts/get_Chla_predicted.py
+loone_data_prep/forecast_scripts/get_NO_Loads_predicted.py
+loone_data_prep/forecast_scripts/loone_q_predict.py
+loone_data_prep/forecast_scripts/loone_wq_predict.py
 loone_data_prep/forecast_scripts/predict_PI.py
 loone_data_prep/forecast_scripts/trib_cond.py
 loone_data_prep/forecast_scripts/weather_forecast.py

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/loone_data_prep.egg-info/requires.txt

@@ -4,7 +4,7 @@ numpy<2
 pandas
 scipy
 geoglows>=2.0.0
-herbie
+herbie-data[extras]
 openmeteo_requests
 requests_cache
 retry-requests

{loone_data_prep-1.1.1 → loone_data_prep-1.2.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "loone_data_prep"
-version = "1.1.1"
+version = "1.2.0"
 description = "Prepare data to run the LOONE model."
 readme = "README.md"
 license = { file = "LICENSE" }
@@ -20,7 +20,7 @@ dependencies = [
     "pandas",
     "scipy",
     "geoglows>=2.0.0",
-    "herbie",
+    "herbie-data[extras]",
     "openmeteo_requests",
     "requests_cache",
     "retry-requests"