cavapy 0.1.1__py3-none-any.whl → 0.1.3__py3-none-any.whl

{cavapy-0.1.1.dist-info → cavapy-0.1.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: cavapy
-Version: 0.1.1
+Version: 0.1.3
 Summary: CAVA Python package. Retrive and analyze climate data.
 Home-page: https://github.com/Risk-Team/cavapy
 License: MIT
@@ -23,30 +23,48 @@ Requires-Dist: xclim (>=0.53.2,<0.54.0)
 Project-URL: Repository, https://github.com/Risk-Team/cavapy
 Description-Content-Type: text/markdown
 
-# Introduction
-cavapy is a Python library that allows retrieval of climate data hosted in THREDDS servers at the University of Cantabria thanks to the [CAVA project](https://risk-team.github.io/CAVAanalytics/articles/CAVA.html). 
+# cavapy: CORDEX-CORE Climate Data Access Simplified
 
-## Data source
-The climate data is available at the THREDDS data server of the University of Cantabria as part of the CAVA (Climate and Agriculture Risk Visualization and Assessment) product developed by FAO, the University of Cantabria, the University of Cape Town and Predictia. 
-CAVA has available CORDEX-CORE climate models, the high resolution (25 Km) dynamically-downscaled climate models used in the IPCC report AR5. Additionally, CAVA  offers access to state-of-the-art reanalyses datasets, such as ERA5 and the observational dataset W5E5 v2.
+## Introduction
 
-The currently available data is:
+`cavapy` is a Python library designed to streamline the retrieval of CORDEX-CORE climate models hosted on THREDDS servers at the University of Cantabria. Using the Open-source Project for a Network Data Access Protocol (**OPeNDAP**), users can directly access and subset datasets without the need to download large NetCDF files. This capability is part of the Climate and Agriculture Risk Visualization and Assessment (CAVA) [project](https://risk-team.github.io/CAVAanalytics/articles/CAVA.html), which focuses on providing high-resolution climate data for scientific, environmental, and agricultural applications.
 
-- CORDEX-CORE simulations (3 GCMs donwscaled with 2 RCMs for two RCPs)
-- W5E5 and ERA5 datasets
-  
-Available variables:
+With `cavapy`, users can efficiently integrate CORDEX-CORE data into their workflows, making it an ideal resource for hydrological and crop modeling, among other climate-sensitive analyses. Additionally, `cavapy` enables bias correction, potentially enhancing the precision and usability of the data for a wide range of applications.
 
-- Daily maximum temperature (tasmax) (°C)
-- Daily minimum temperature (tasmin) (°C)
-- Daily precipitation (pr) (mm)
-- Daily relative humidity (hurs) (%)
-- Daily wind speed (sfcWind) (2 m level m/s)
-- Daily solar radiation (rsds) (W/m^2)
+---
 
+## Data Source
+
+The climate data provided by `cavapy` is hosted on the THREDDS data server of the University of Cantabria as part of the CAVA project. CAVA is a collaborative effort by FAO, the University of Cantabria, the University of Cape Town, and Predictia, aimed at democratising accessibility and usability of climate information.
+
+### Key Datasets:
+- **CORDEX-CORE Simulations**: Dynamically downscaled high-resolution (25 km) climate models, used in the IPCC AR5 report, featuring simulations from:
+  - 3 Global Climate Models (GCMs)
+  - 2 Regional Climate Models (RCMs)
+  - Two Representative Concentration Pathways (RCPs: RCP4.5 and RCP8.5)
+- **Reanalyses and Observational Datasets**:
+  - ERA5
+  - W5E5 v2
+
+These datasets provide robust inputs for climate and environmental modeling, supporting scientific and policy-driven decision-making.
+
+---
+
+## Available Variables
+
+`cavapy` grants access to critical climate variables, enabling integration into diverse modeling frameworks. The variables currently available include:
+
+- **Daily Maximum Temperature (tasmax)**: °C  
+- **Daily Minimum Temperature (tasmin)**: °C  
+- **Daily Precipitation (pr)**: mm  
+- **Daily Relative Humidity (hurs)**: %  
+- **Daily Wind Speed (sfcWind)**: 2 m level, m/s  
+- **Daily Solar Radiation (rsds)**: W/m²  
+
+---
 
 ## Installation
-cavapy can be installed with pip. Ensure that you are not using a python version > 13. 
+cavapy can be installed with pip. Ensure that you are not using a python version > 3. 
 
 ```
 conda create -n test python=3.11
@@ -69,6 +87,8 @@ Since bias-correction requires both the historical run of the CORDEX model and t
 It takes about 10 minutes to run each of the tasks below. For bigger areas/country, the computational time increases. For example, for Zambia it takes about 30 minutes.
 
 ### Bias-corrected climate projections
+**By default all available climate variables are used. You can specify a subset with the variable argument**
+
 ```
 import cavapy
 Togo_climate_data = cavapy.get_climate_data(country="Togo", cordex_domain="AFR-22", rcp="rcp26", gcm="MPI", rcm="REMO", years_up_to=2030, obs=False, bias_correction=True, historical=False)

cavapy-0.1.3.dist-info/RECORD

@@ -0,0 +1,5 @@
+cavapy.py,sha256=i9ULBOrL4NrtG13KFl7g9JEytsXJkF0Vepudebw7MEY,24377
+cavapy-0.1.3.dist-info/LICENSE,sha256=1etyG4_n-Tb3yoNMwQ38g_WxXFQ4E_ZCjZc-AGYPc9U,1151
+cavapy-0.1.3.dist-info/METADATA,sha256=M2KwoqRuQFd7SAubXRnwFKSJaVci1HfU1Q-BwOoT9Vo,6002
+cavapy-0.1.3.dist-info/WHEEL,sha256=RaoafKOydTQ7I_I3JTrPCg6kUmTgtm4BornzOqyEfJ8,88
+cavapy-0.1.3.dist-info/RECORD,,

cavapy.py

@@ -161,7 +161,7 @@ def get_climate_data(
     else:
         variables = VALID_VARIABLES
 
-    _validate_urls(gcm, rcm, rcp, remote, cordex_domain, obs)
+    _validate_urls(gcm, rcm, rcp, remote, cordex_domain, obs, historical, bias_correction)
 
     bbox = _geo_localize(country, xlim, ylim, buffer, cordex_domain)
 
@@ -206,6 +206,8 @@ def _validate_urls(
     remote: bool = True,
     cordex_domain: str = None,
     obs: bool = False,
+    historical: bool = False,
+    bias_correction: bool = False,
 ):
     # Load the data
     log = logger.getChild("URL-validation")
@@ -219,6 +221,11 @@ def _validate_urls(
         # Set the column to use based on whether the data is remote or local
         column_to_use = "location" if remote else "hub"
 
+        # Define which experiments we need
+        experiments = [rcp]
+        if historical or bias_correction:
+            experiments.append("historical")
+
         # Filter the data based on the conditions
         filtered_data = data[
             lambda x: (
@@ -226,27 +233,19 @@ def _validate_urls(
                 & (x["domain"] == cordex_domain)
                 & (x["model"].str.contains(gcm, na=False))
                 & (x["rcm"].str.contains(rcm, na=False))
-                & (x["experiment"].isin([rcp, "historical"]))
+                & (x["experiment"].isin(experiments))
             )
         ][["experiment", column_to_use]]
 
         # Extract the column values as a list
-        num_rows = filtered_data.shape[0]
-        column_values = filtered_data[column_to_use]
-
-        if num_rows == 1:
-            # Log the output for one row
-            row1 = column_values.iloc[0]
-            log_proj = logger.getChild("URL-validation-projections")
-            log_proj.info(f"{row1}")
-        else:
-            # Log the output for two rows
-            row1 = column_values.iloc[0]
-            row2 = column_values.iloc[1]
-            log_hist = logger.getChild("URL-validation-historical")
-            log_proj = logger.getChild("URL-validation-projections")
-            log_hist.info(f"{row1}")
-            log_proj.info(f"{row2}")
+        for _, row in filtered_data.iterrows():
+            if row["experiment"] == "historical":
+                log_hist = logger.getChild("URL-validation-historical")
+                log_hist.info(f"{row[column_to_use]}")
+            else:
+                log_proj = logger.getChild("URL-validation-projections")
+                log_proj.info(f"{row[column_to_use]}")
+
     else:  # when obs is True
         log_obs = logger.getChild("URL-validation-observations")
         log_obs.info(f"{ERA5_DATA_REMOTE_URL}")
@@ -417,12 +416,18 @@ def _climate_data_for_variable(
     )
     data = pd.read_csv(inventory_csv_url)
     column_to_use = "location" if remote else "hub"
+    
+    # Filter data based on whether we need historical data
+    experiments = [rcp]
+    if historical or bias_correction:
+        experiments.append("historical")
+        
     filtered_data = data[
         lambda x: (x["activity"].str.contains("FAO", na=False))
         & (x["domain"] == cordex_domain)
         & (x["model"].str.contains(gcm, na=False))
         & (x["rcm"].str.contains(rcm, na=False))
-        & (x["experiment"].isin([rcp, "historical"]))
+        & (x["experiment"].isin(experiments))
     ][["experiment", column_to_use]]
 
     future_obs = None
@@ -454,12 +459,17 @@ def _climate_data_for_variable(
 
         # Add the downloaded models to the DataFrame
         filtered_data["models"] = downloaded_models
-        hist = (
-            filtered_data["models"].iloc[0].interpolate_na(dim="time", method="linear")
-        )
-        proj = (
-            filtered_data["models"].iloc[1].interpolate_na(dim="time", method="linear")
-        )
+        
+        if historical or bias_correction:
+            hist = filtered_data[filtered_data["experiment"] == "historical"]["models"].iloc[0]
+            proj = filtered_data[filtered_data["experiment"] == rcp]["models"].iloc[0]
+            
+            hist = hist.interpolate_na(dim="time", method="linear")
+            proj = proj.interpolate_na(dim="time", method="linear")
+        else:
+            proj = filtered_data["models"].iloc[0]
+            proj = proj.interpolate_na(dim="time", method="linear")
+            
         if bias_correction and historical:
             # Load observations for bias correction
             ref = future_obs.result()
@@ -581,6 +591,14 @@ def _download_data(
     else:
         log.info(f"Establishing connection to CORDEX data for {variable}")
         ds_var = xr.open_dataset(url)[variable]
+        
+        # Check if time dimension has a prefix, indicating variable is not available
+        time_dims = [dim for dim in ds_var.dims if dim.startswith('time_')]
+        if time_dims:
+            msg = f"Variable {variable} is not available for this model: {url}"
+            log.exception(msg)
+            raise ValueError(msg)
+            
         log.info(f"Connection to CORDEX data for {variable} has been established")
         ds_cropped = ds_var.sel(
             longitude=slice(bbox["xlim"][0], bbox["xlim"][1]),
@@ -636,14 +654,15 @@ def _download_data(
 
 if __name__ == "__main__":
     data = get_climate_data(
-        country="Zambia",
+        country="Togo",
+        variables=["hurs"],
         cordex_domain="AFR-22",
         rcp="rcp26",
         gcm="MPI",
-        rcm="REMO",
+        rcm="Reg",
         years_up_to=2030,
         obs=False,
-        bias_correction=True,
+        bias_correction=False,
         historical=False,
     )
     print(data)

cavapy-0.1.1.dist-info/RECORD

@@ -1,5 +0,0 @@
-cavapy.py,sha256=dYDRyrrKn560xp4gGOL73ctn8YZLqrtkMw6631YMbkI,23557
-cavapy-0.1.1.dist-info/LICENSE,sha256=1etyG4_n-Tb3yoNMwQ38g_WxXFQ4E_ZCjZc-AGYPc9U,1151
-cavapy-0.1.1.dist-info/METADATA,sha256=PB0uKvdF9W0vli69Vf4KKdrOaOLMFnhwtitXi0od7KE,4732
-cavapy-0.1.1.dist-info/WHEEL,sha256=RaoafKOydTQ7I_I3JTrPCg6kUmTgtm4BornzOqyEfJ8,88
-cavapy-0.1.1.dist-info/RECORD,,