Error caused by Timestamp values in MinTemp or MaxTemp

Hi everyone, 

I have got an error that I don't arrive to correct despite all my efforts.

Here is my code : the final aim is to compare the yield for an AgriPV and a not AgriPV field. So, first I am collecting wheater date from NASA POWER. Then, I convert all datas in the dataframe requiered by aquacrop. But finally my model do not run and I can't retrieve yield data. Can you help me please ? Thank you !

Here is my code, the error message is about run_simulation() fonction and the `success = model.run_model(till_termination=True)` command 


`import pandas as pd
import numpy as np
import requests
from math import radians, tan, sin, cos
from tqdm import tqdm
from pvlib.solarposition import get_solarposition
import matplotlib.pyplot as plt
import requests
import pandas as pd

# ==================== 1. Weather and hourly solar radiation data ====================
def get_nasa_power_data_hourly(lat, lon, start='2021-01-01', end='2021-12-31'):
    url = "https://power.larc.nasa.gov/api/temporal/hourly/point"
    params = {
        "parameters": "ALLSKY_SFC_SW_DWN,ALLSKY_SFC_SW_DIFF,ALLSKY_SFC_SW_DNI,ALLSKY_SRF_ALB",
        "community": "RE",
        "longitude": lon,
        "latitude": lat,
        "start": start.replace("-", ""),
        "end": end.replace("-", ""),
        "format": "JSON",
        "time-standard": "UTC"
    }

    print("📡 Requesting NASA POWER data...")
    response = requests.get(url, params=params)

    if response.status_code != 200:
        print("❌ NASA POWER request error:", response.status_code)
        print("↪ Raw response:", response.text)
        raise RuntimeError("NASA POWER API error")

    data = response.json()
    param = data.get("properties", {}).get("parameter", {})
    ghi = param.get("ALLSKY_SFC_SW_DWN", {})
    dhi = param.get("ALLSKY_SFC_SW_DIFF", {})
    dni = param.get("ALLSKY_SFC_SW_DNI", {})
    albedo = param.get("ALLSKY_SRF_ALB", {})

    if not (ghi and dhi and dni and albedo):
        print("⚠️ Missing data for some parameters.")
        return None

    # Extract common timestamps
    timestamps = sorted(set(ghi.keys()) & set(dhi.keys()) & set(dni.keys()) & set(albedo.keys()))

    df = pd.DataFrame({
        "date": pd.to_datetime(timestamps, format="%Y%m%d%H", utc=True),
        "GHI": [ghi[t] for t in timestamps],
        "DHI": [dhi[t] for t in timestamps],
        "DNI": [dni[t] for t in timestamps],
        "ALB": [albedo[t] for t in timestamps]
    })

    df["GHI"] = df["GHI"].astype(float)
    df["DHI"] = df["DHI"].astype(float)
    df["DNI"] = df["DNI"].astype(float)
    df["ALB"] = df["ALB"].astype(float)
    df["day_of_year"] = df["date"].dt.dayofyear

    print(f"✅ {len(df)} hourly values loaded.")
    return df

# ==================== 2. Solar geometry ====================
def compute_solar_geometry_hourly(df, lat, lon):
    times = df.index.tz_convert(None)
    solpos = get_solarposition(times, lat, lon)
    df['solar_zenith'] = solpos['apparent_zenith'].values
    df['solar_azimuth'] = solpos['azimuth'].values
    df['solar_elevation'] = 90 - df['solar_zenith']
    return df

# ==================== 3. Shadow simulation ====================
def simulate_shadow_map(df, panel_tilt, row_spacing, panel_height, leg_height, grid_size=10):
    dx = 1.0  # m/pixel
    n_pixels = grid_size
    field_size = (n_pixels, n_pixels)
    irradiance_maps = np.zeros((len(df), *field_size))
    row_interval = int(row_spacing / dx)
    panel_rows = list(range(0, n_pixels, row_interval))
    for t, (index, row) in tqdm(enumerate(df.iterrows()), total=len(df), desc="Shadow calculation"):
        elevation = row["solar_elevation"]
        azimuth = row["solar_azimuth"]
        if elevation <= 0:
            continue
        tilt_rad = radians(panel_tilt)
        h_panel = leg_height + panel_height * sin(tilt_rad)
        shadow_length = h_panel / tan(radians(elevation))
        shadow_pixels = int(round(shadow_length / dx))
        azimuth_rad = radians(azimuth)
        dx_shadow = -sin(azimuth_rad)
        dy_shadow = -cos(azimuth_rad)
        shadow_map = np.ones(field_size)

        for y_row in panel_rows:
            for x in range(n_pixels):
                for s in range(1, shadow_pixels + 1):
                    xx = int(round(x + s * dx_shadow))
                    yy = int(round(y_row + s * dy_shadow))
                    if 0 <= xx < n_pixels and 0 <= yy < n_pixels:
                        shadow_map[xx, yy] *= 0.2
        irradiance_maps[t] = shadow_map * row["GHI"]
    return irradiance_maps

# ==================== 4. Daily aggregation ====================
def aggregate_irradiance(irradiance_maps, df):
    hourly_means = np.mean(irradiance_maps.reshape(len(df), -1), axis=1)
    df = df.copy()
    df['irradiance_corrected'] = hourly_means
    df['date_only'] = df.index.date
    df_daily = df.groupby('date_only')['irradiance_corrected'].sum().reset_index()
    df_daily.columns = ['date', 'SolarRadiation']
    return df_daily

# ==================== 5. Daily weather data ====================
def get_nasa_weather_daily(lat, lon, start, end):
    url = "https://power.larc.nasa.gov/api/temporal/daily/point"
    params = {
        "parameters": "T2M_MIN,T2M_MAX,PRECTOTCORR",
        "community": "AG",
        "longitude": lon,
        "latitude": lat,
        "start": start.replace("-", ""),
        "end": end.replace("-", ""),
        "format": "JSON"
    }
    response = requests.get(url, params=params)
    data = response.json()["properties"]["parameter"]
    dates = pd.date_range(start=start, end=end)
    df = pd.DataFrame({
        "date": dates,
        "MinTemp": [data["T2M_MIN"][d.strftime("%Y%m%d")] for d in dates],
        "MaxTemp": [data["T2M_MAX"][d.strftime("%Y%m%d")] for d in dates],
        "Precipitation": [data["PRECTOTCORR"][d.strftime("%Y%m%d")] for d in dates],
    })
    return df

# ==================== 6. Merge weather and radiation data ====================
def prepare_aquacrop_input(lat, lon, start, end, shadow=True):
    df_hourly = get_nasa_power_data_hourly(lat, lon, start, end)
    df_hourly.set_index("date", inplace=True)
    df_hourly = compute_solar_geometry_hourly(df_hourly, lat, lon)

    if shadow:
        irradiance_maps = simulate_shadow_map(df_hourly, panel_tilt=25, row_spacing=5, panel_height=1, leg_height=1)
        irradiance_daily = aggregate_irradiance(irradiance_maps, df_hourly)
    else:
        df_hourly['date_only'] = df_hourly.index.date
        irradiance_daily = df_hourly.groupby('date_only')['GHI'].sum().reset_index()
        irradiance_daily.columns = ['date', 'SolarRadiation']

    meteo = get_nasa_weather_daily(lat, lon, start, end)
    meteo['date'] = pd.to_datetime(meteo['date'])
    irradiance_daily['date'] = pd.to_datetime(irradiance_daily['date'])
    meteo = meteo.merge(irradiance_daily, on='date')
    meteo['ET0'] = 0.0023 * (meteo["MaxTemp"] - meteo["MinTemp"])**0.5 * (meteo["MaxTemp"] + meteo["MinTemp"]) / 2 + 0.408 * meteo["SolarRadiation"] / 2.45

    # 🔧 Rename columns for AquaCrop
    meteo = meteo.rename(columns={
        'date': 'Date',
        'ET0': 'ReferenceET'})
    meteo = meteo[['Date', 'MinTemp', 'MaxTemp', 'Precipitation', 'ReferenceET']]
    
    meteo["MinTemp"] = pd.to_numeric(meteo["MinTemp"], errors="coerce")
    meteo["MaxTemp"] = pd.to_numeric(meteo["MaxTemp"], errors="coerce")
    meteo["Precipitation"] = pd.to_numeric(meteo["Precipitation"], errors="coerce")
    meteo["ReferenceET"] = pd.to_numeric(meteo["ReferenceET"], errors="coerce")

    print("🧪 Final data ready for AquaCrop:", meteo.columns)
    return meteo

# ==================== 7. Export to AquaCrop format ====================
def export_to_aquacrop_format(df, path):
    df_out = df[['Date', 'MinTemp', 'MaxTemp', 'Precipitation', 'ReferenceET']].copy()
    df_out.columns = ['Date', 'Tmin', 'Tmax', 'P', 'ET0']
    df_out['Date'] = df_out['Date'].dt.strftime('%Y-%m-%d')
    df_out.to_csv(path, index=False)

# ==================== 8. AquaCrop simulation using pyAquaCrop ====================
from aquacrop import AquaCropModel, Soil, Crop, InitialWaterContent

def run_simulation(meteo_df, name):
    # meteo_df must contain these columns: Date, MinTemp, MaxTemp, Precipitation, ReferenceET
    print("\n🛠️ Checking data types for MinTemp / MaxTemp:")
    print(meteo_df[["MinTemp", "MaxTemp"]].applymap(type).value_counts())

    print("\n🔎 Looking for Timestamp in MinTemp:")
    print(meteo_df[meteo_df["MinTemp"].apply(lambda x: isinstance(x, pd.Timestamp))])

    for col in ['MinTemp', 'MaxTemp']:
        wrong_types = meteo_df[~meteo_df[col].apply(lambda x: isinstance(x, (int, float)))]
        if not wrong_types.empty:
            print(f"\n❌ Problem detected in {col}:")
            print(wrong_types.head())
            raise TypeError(f"Column {col} contains non-numeric types (e.g., {type(wrong_types.iloc[0][col])})")
    
    print("\n📋 Unique types per column:")
    for col in ['MinTemp', 'MaxTemp', 'Precipitation', 'ReferenceET']:
        meteo_df[col] = pd.to_numeric(meteo_df[col], errors="coerce")
        print(f"{col}: {meteo_df[col].apply(type).value_counts()}")

    start_date = pd.to_datetime(meteo_df["Date"]).iloc[0].strftime("%Y/%m/%d")
    end_date = pd.to_datetime(meteo_df["Date"]).iloc[-1].strftime("%Y/%m/%d")
    print(f"Start: {start_date}, End: {end_date}, Type: {type(start_date)}")

    soil = Soil("SandyLoam")
    crop = Crop("Maize", planting_date="04/01")
    init_wc = InitialWaterContent("FC")

    model = AquaCropModel(
        start_date,
        end_date,
        meteo_df,
        soil,
        crop,
        init_wc
    )

    success = model.run_model(till_termination=True)
    if not success:
        print("Simulation failed")
        return None

    return model

def validate_and_clean_meteo_df(df):
    for col in ['MinTemp', 'MaxTemp', 'Precipitation', 'ReferenceET']:
        df[col] = pd.to_numeric(df[col], errors='coerce')

    for col in ['MinTemp', 'MaxTemp', 'Precipitation', 'ReferenceET']:
        non_numeric = df[~df[col].apply(lambda x: isinstance(x, (int, float)))]
        if not non_numeric.empty:
            print(f"Non-numeric values found in {col}:")
            print(non_numeric)
            raise ValueError(f"Non-numeric values found in {col}")

    df = df.dropna(subset=['MinTemp', 'MaxTemp', 'Precipitation', 'ReferenceET'])
    return df

# ==================== 9. usage ====================
lat, lon = 45.0, 3.0
start = "2021-01-01"
end = "2021-12-31"

meteo_shadow = prepare_aquacrop_input(lat, lon, start, end, shadow=True)
meteo_clear = prepare_aquacrop_input(lat, lon, start, end, shadow=False)
print(meteo_shadow.dtypes)
print(meteo_shadow.head(10))

export_to_aquacrop_format(meteo_shadow, "meteo_with_panels.csv")
export_to_aquacrop_format(meteo_clear, "meteo_without_panels.csv")
print("Types in meteo_shadow['MinTemp']:")
print(meteo_shadow["MinTemp"].apply(type).value_counts())
print("Types in meteo_shadow['MaxTemp']:")
print(meteo_shadow["MaxTemp"].apply(type).value_counts())

bad_rows = meteo_shadow[meteo_shadow["MinTemp"].apply(lambda x: isinstance(x, pd.Timestamp))]
print("Rows with Timestamp in MinTemp:")
print(bad_rows)

meteo_shadow["MinTemp"] = pd.to_numeric(meteo_shadow["MinTemp"], errors="coerce")
meteo_shadow["MaxTemp"] = pd.to_numeric(meteo_shadow["MaxTemp"], errors="coerce")
meteo_shadow = meteo_shadow.dropna(subset=["MinTemp", "MaxTemp"])

meteo_shadow = validate_and_clean_meteo_df(meteo_shadow)
meteo_clear = validate_and_clean_meteo_df(meteo_clear)

model_shadow = run_simulation(meteo_shadow, "With panels")
model_clear = run_simulation(meteo_clear, "Without panels")
`

Here is the exact error message 

`📡 Requesting NASA POWER data...
✅ 8760 hourly values loaded.
Shadow calculation: 100%|██| 8760/8760 [00:03<00:00, 2822.81it/s]
🧪 Final data ready for AquaCrop: Index(['Date', 'MinTemp', 'MaxTemp', 'Precipitation', 'ReferenceET'], dtype='object')
📡 Requesting NASA POWER data...
✅ 8760 hourly values loaded.
🧪 Final data ready for AquaCrop: Index(['Date', 'MinTemp', 'MaxTemp', 'Precipitation', 'ReferenceET'], dtype='object')
Date             datetime64[ns]
MinTemp                 float64
MaxTemp                 float64
Precipitation           float64
ReferenceET             float64
dtype: object
        Date  MinTemp  MaxTemp  Precipitation  ReferenceET
0 2021-01-01   -10.41    -1.60           1.15    67.094424        
1 2021-01-02    -7.29    -3.17           0.18    43.354624        
2 2021-01-03    -9.72    -3.03           0.44    78.721383        
3 2021-01-04    -9.56    -1.27           3.44    79.152975        
4 2021-01-05   -10.59    -1.86           0.43    50.847191        
5 2021-01-06   -10.81    -2.99           0.10    48.384140        
6 2021-01-07   -10.07    -2.62           0.22    75.721669        
7 2021-01-08    -9.55    -2.73           0.09    98.918739        
8 2021-01-09    -7.45    -3.58           0.03   103.869624        
9 2021-01-10    -9.10    -4.25           0.03    97.965490        
Types in meteo_shadow['MinTemp']:
MinTemp
<class 'float'>    365
Name: count, dtype: int64
Types in meteo_shadow['MaxTemp']:
MaxTemp
<class 'float'>    365
Name: count, dtype: int64
Rows with Timestamp in MinTemp:
Empty DataFrame
Columns: [Date, MinTemp, MaxTemp, Precipitation, ReferenceET]     
Index: []

🛠️ Checking data types for MinTemp / MaxTemp:
c:\Users\omare\OneDrive\Documents\UNSW\Research\RECHERCHE\VScode\projet_stics\mon_script_aquacrop.py:185: FutureWarning: DataFrame.applymap has been deprecated. Use DataFrame.map instead.
  print(meteo_df[["MinTemp", "MaxTemp"]].applymap(type).value_counts())
MinTemp          MaxTemp        
<class 'float'>  <class 'float'>    365
Name: count, dtype: int64

🔎 Looking for Timestamp in MinTemp:
Empty DataFrame
Columns: [Date, MinTemp, MaxTemp, Precipitation, ReferenceET]     
Index: []

📋 Unique types per column:
MinTemp: MinTemp
<class 'float'>    365
Name: count, dtype: int64
MaxTemp: MaxTemp
<class 'float'>    365
Name: count, dtype: int64
Precipitation: Precipitation
<class 'float'>    365
Name: count, dtype: int64
ReferenceET: ReferenceET
<class 'float'>    365
Name: count, dtype: int64
Start: 2021/01/01, End: 2021/12/31, Type: <class 'str'>
Traceback (most recent call last):
  File "c:\Users\omare\OneDrive\Documents\UNSW\Research\RECHERCHE\VScode\projet_stics\mon_script_aquacrop.py", line 268, in <module>
    model_shadow = run_simulation(meteo_shadow, "With panels")    
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^    
  File "c:\Users\omare\OneDrive\Documents\UNSW\Research\RECHERCHE\VScode\projet_stics\mon_script_aquacrop.py", line 219, in run_simulation
    success = model.run_model(till_termination=True)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\omare\OneDrive\Documents\UNSW\Research\RECHERCHE\VScode\.venv\Lib\site-packages\aquacrop\core.py", line 279, in run_model
    ) = self._perform_timestep()
        ^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\omare\OneDrive\Documents\UNSW\Research\RECHERCHE\VScode\.venv\Lib\site-packages\aquacrop\core.py", line 325, in _perform_timestep
    new_cond, param_struct, outputs = solution_single_time_step(  
                                      ^^^^^^^^^^^^^^^^^^^^^^^^^^  
  File "C:\Users\omare\OneDrive\Documents\UNSW\Research\RECHERCHE\VScode\.venv\Lib\site-packages\aquacrop\timestep\run_single_timestep.py", line 136, in solution_single_time_step
    gdd = growing_degree_day(
          ^^^^^^^^^^^^^^^^^^^
  File "C:\Users\omare\OneDrive\Documents\UNSW\Research\RECHERCHE\VScode\.venv\Lib\site-packages\aquacrop\solution\growing_degree_day.py", line 58, in growing_degree_day
    temp_min = min(temp_min, Tupp)
               ^^^^^^^^^^^^^^^^^^^
TypeError: '<' not supported between instances of 'float' and 'Timestamp'
`

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Error caused by Timestamp values in MinTemp or MaxTemp #196

==================== 1. Weather and hourly solar radiation data ====================

==================== 2. Solar geometry ====================

==================== 3. Shadow simulation ====================

==================== 4. Daily aggregation ====================

==================== 5. Daily weather data ====================

==================== 6. Merge weather and radiation data ====================

==================== 7. Export to AquaCrop format ====================

==================== 8. AquaCrop simulation using pyAquaCrop ====================

==================== 9. usage ====================

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Error caused by Timestamp values in MinTemp or MaxTemp #196

Description

==================== 1. Weather and hourly solar radiation data ====================

==================== 2. Solar geometry ====================

==================== 3. Shadow simulation ====================

==================== 4. Daily aggregation ====================

==================== 5. Daily weather data ====================

==================== 6. Merge weather and radiation data ====================

==================== 7. Export to AquaCrop format ====================

==================== 8. AquaCrop simulation using pyAquaCrop ====================

==================== 9. usage ====================

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions