Create app.py

app.py

@@ -0,0 +1,327 @@
+# app.py
+import pandas as pd
+import numpy as np
+import plotly.express as px
+import plotly.graph_objects as go
+import plotly.subplots as sp
+import gradio as gr
+from datetime import datetime
+
+# Constants
+NASA_DATA_URL = "https://data.giss.nasa.gov/gistemp/tabledata_v4/GLB.Ts+dSST.csv"
+CURRENT_YEAR = datetime.now().year
+
+def load_and_process_data():
+    """Load and process NASA temperature data with robust error handling"""
+    try:
+        # Read NASA data with metadata handling
+        df = pd.read_csv(
+            NASA_DATA_URL,
+            skiprows=1,
+            na_values=['***', '****', '*****', '******']
+        )
+        
+        # Clean and reshape data
+        df = df[df['Year'] >= 1880]  # Reliable data starts from 1880
+        df = df[['Year'] + [str(m) for m in range(1,13)]]
+        df.columns = ['Year'] + [f"{i:02d}" for i in range(1,13)]
+        
+        # Melt to long format and handle dates
+        df = df.melt(id_vars='Year', var_name='Month', value_name='Anomaly')
+        df['Date'] = pd.to_datetime(df['Year'].astype(str) + '-' + df['Month'], format='%Y-%m')
+        df = df.dropna(subset=['Anomaly'])
+        df['Anomaly'] = df['Anomaly'].astype(float)
+        df['Decade'] = (df['Year'] // 10) * 10
+        
+        # Calculate rolling averages
+        df['5yr_avg'] = df['Anomaly'].rolling(60, min_periods=1).mean()
+        df['10yr_avg'] = df['Anomaly'].rolling(120, min_periods=1).mean()
+        
+        return df
+    
+    except Exception as e:
+        print(f"Data loading error: {e}")
+        return pd.DataFrame()
+
+def create_time_series_plot(df, show_uncertainty=False):
+    """Create interactive time series plot with advanced features"""
+    if df.empty:
+        return go.Figure()
+    
+    fig = go.Figure()
+    
+    # Add monthly anomalies
+    fig.add_trace(go.Scatter(
+        x=df['Date'], 
+        y=df['Anomaly'],
+        mode='markers',
+        marker=dict(size=3, opacity=0.3, color='lightgray'),
+        name='Monthly Anomaly'
+    ))
+    
+    # Add trend lines
+    fig.add_trace(go.Scatter(
+        x=df['Date'], 
+        y=df['5yr_avg'],
+        mode='lines',
+        line=dict(width=2, color='blue'),
+        name='5-Year Average'
+    ))
+    
+    fig.add_trace(go.Scatter(
+        x=df['Date'], 
+        y=df['10yr_avg'],
+        mode='lines',
+        line=dict(width=3, color='red'),
+        name='10-Year Average'
+    ))
+    
+    # Add uncertainty bands if requested
+    if show_uncertainty:
+        rolling_std = df['Anomaly'].rolling(120).std()
+        fig.add_trace(go.Scatter(
+            x=df['Date'],
+            y=df['10yr_avg'] + rolling_std,
+            fill=None,
+            mode='lines',
+            line=dict(width=0),
+            showlegend=False
+        ))
+        
+        fig.add_trace(go.Scatter(
+            x=df['Date'],
+            y=df['10yr_avg'] - rolling_std,
+            fill='tonexty',
+            mode='lines',
+            line=dict(width=0),
+            name='Uncertainty'
+        ))
+    
+    # Add reference line at 0°C
+    fig.add_hline(y=0, line_dash="dash", line_color="black")
+    
+    # Add significant warming markers
+    recent = df[df['Year'] >= 2000]
+    if not recent.empty:
+        fig.add_trace(go.Scatter(
+            x=recent['Date'],
+            y=recent['10yr_avg'],
+            mode='markers',
+            marker=dict(size=8, color='red'),
+            name='Post-2000 Average'
+        ))
+    
+    # Layout enhancements
+    fig.update_layout(
+        title='Global Temperature Anomalies (1880-Present)',
+        xaxis_title='Year',
+        yaxis_title='Temperature Anomaly (°C)',
+        hovermode='x unified',
+        template='plotly_dark',
+        height=600,
+        annotations=[
+            dict(
+                x=0.01, y=0.01,
+                xref="paper", yref="paper",
+                text="Data Source: NASA GISS",
+                showarrow=False
+            )
+        ]
+    )
+    
+    return fig
+
+def create_heatmap(df, decade_range=(1950, CURRENT_YEAR)):
+    """Create decadal heatmap visualization"""
+    if df.empty:
+        return go.Figure()
+    
+    # Filter and aggregate data
+    filtered = df[df['Decade'].between(decade_range[0], decade_range[1])]
+    if filtered.empty:
+        return go.Figure()
+    
+    pivot_df = filtered.pivot_table(
+        index='Decade',
+        columns='Month',
+        values='Anomaly',
+        aggfunc='mean'
+    )
+    
+    # Create heatmap
+    fig = px.imshow(
+        pivot_df,
+        labels=dict(x="Month", y="Decade", color="Anomaly"),
+        x=['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 
+           'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'],
+        color_continuous_scale='RdBu_r',
+        aspect="auto"
+    )
+    
+    # Add annotations
+    for i, row in enumerate(pivot_df.values):
+        for j, value in enumerate(row):
+            fig.add_annotation(
+                x=j, y=i,
+                text=f"{value:.1f}",
+                showarrow=False,
+                font=dict(color='black' if abs(value) < 0.8 else 'white')
+            )
+    
+    # Layout enhancements
+    fig.update_layout(
+        title=f'Decadal Temperature Anomalies ({decade_range[0]}-{decade_range[1]})',
+        xaxis_title="",
+        yaxis_title="Decade",
+        coloraxis_colorbar=dict(title="Anomaly (°C)"),
+        height=600
+    )
+    
+    return fig
+
+def create_regional_comparison(df):
+    """Create regional comparison visualization (mock data)"""
+    # In a real implementation, this would use actual regional data
+    regions = {
+        'Arctic': 2.5,
+        'Global': 1.2,
+        'Europe': 1.5,
+        'Asia': 1.3,
+        'North America': 1.4,
+        'Africa': 1.1,
+        'South America': 1.0,
+        'Oceania': 1.1,
+        'Antarctica': 0.8
+    }
+    
+    fig = go.Figure(go.Bar(
+        x=list(regions.values()),
+        y=list(regions.keys()),
+        orientation='h',
+        marker_color='crimson'
+    ))
+    
+    fig.update_layout(
+        title='Regional Warming Rates (Since Pre-Industrial)',
+        xaxis_title='Temperature Increase (°C)',
+        yaxis_title='Region',
+        template='plotly_dark',
+        height=500
+    )
+    
+    return fig
+
+def create_dashboard():
+    """Create Gradio dashboard with multiple visualizations"""
+    df = load_and_process_data()
+    
+    with gr.Blocks(title="NASA Climate Viz", theme=gr.themes.Soft()) as demo:
+        gr.Markdown("# 🌍 Earth's Surface Temperature Analysis")
+        gr.Markdown("### Visualization of NASA's Global Temperature Data")
+        
+        with gr.Row():
+            gr.Markdown("""
+            **Data Source**: [NASA Goddard Institute for Space Studies](https://data.giss.nasa.gov/gistemp/)  
+            **Last Update**: {CURRENT_YEAR}  
+            **Base Period**: 1951-1980
+            """)
+        
+        with gr.Tab("Time Series Analysis"):
+            gr.Markdown("## Global Temperature Anomalies Over Time")
+            with gr.Row():
+                show_uncertainty = gr.Checkbox(label="Show Uncertainty Bands")
+                year_range = gr.RangeSlider(
+                    minimum=1880,
+                    maximum=CURRENT_YEAR,
+                    value=[1950, CURRENT_YEAR],
+                    label="Year Range"
+                )
+            time_series = gr.Plot()
+        
+        with gr.Tab("Decadal Heatmap"):
+            gr.Markdown("## Monthly Anomalies by Decade")
+            decade_range = gr.RangeSlider(
+                minimum=1880,
+                maximum=CURRENT_YEAR - (CURRENT_YEAR % 10),
+                value=[1950, CURRENT_YEAR - (CURRENT_YEAR % 10)],
+                step=10,
+                label="Decade Range"
+            )
+            heatmap = gr.Plot()
+        
+        with gr.Tab("Regional Comparison"):
+            gr.Markdown("## Regional Warming Patterns")
+            gr.Markdown("*Note: Regional data shown is for demonstration purposes*")
+            region_plot = gr.Plot()
+        
+        with gr.Tab("Data Insights"):
+            gr.Markdown("## Key Climate Observations")
+            if not df.empty:
+                latest = df[df['Year'] == df['Year'].max()]
+                hottest_year = df.groupby('Year')['Anomaly'].mean().idxmax()
+                hottest_decade = df.groupby('Decade')['Anomaly'].mean().idxmax()
+                
+                insights = f"""
+                - 📈 **Current Decade ({CURRENT_YEAR//10*10}s)**: {df[df['Decade'] == CURRENT_YEAR//10*10]['Anomaly'].mean():.2f}°C anomaly
+                - 🔥 **Hottest Year**: {hottest_year} ({df[df['Year'] == hottest_year]['Anomaly'].mean():.2f}°C)
+                - 🌡️ **Recent Temperature**: {latest['Anomaly'].mean():.2f}°C above baseline
+                - 📅 **Long-term Trend**: {df['Anomaly'].mean():.2f}°C average anomaly since 1880
+                - ⏱️ **Acceleration**: Warming rate has doubled since 1980
+                """
+            else:
+                insights = "Data not available"
+            
+            gr.Markdown(insights)
+            gr.Markdown("### Temperature Change Since 1880")
+            if not df.empty:
+                change_df = df.groupby('Year').agg({'Anomaly': 'mean'}).reset_index()
+                change_df['Change'] = change_df['Anomaly'].cumsum()
+                change_plot = px.area(change_df, x='Year', y='Change')
+                gr.Plot(change_plot)
+        
+        # Event handling
+        show_uncertainty.change(
+            fn=lambda u, y: create_time_series_plot(
+                df[(df['Year'] >= y[0]) & (df['Year'] <= y[1])],
+                u
+            ),
+            inputs=[show_uncertainty, year_range],
+            outputs=time_series
+        )
+        
+        year_range.change(
+            fn=lambda y, u: create_time_series_plot(
+                df[(df['Year'] >= y[0]) & (df['Year'] <= y[1])],
+                u
+            ),
+            inputs=[year_range, show_uncertainty],
+            outputs=time_series
+        )
+        
+        decade_range.change(
+            fn=lambda dr: create_heatmap(df, dr),
+            inputs=decade_range,
+            outputs=heatmap
+        )
+        
+        # Initial renders
+        demo.load(
+            fn=lambda: create_time_series_plot(df[(df['Year'] >= 1950) & (df['Year'] <= CURRENT_YEAR)]),
+            outputs=time_series
+        )
+        
+        demo.load(
+            fn=lambda: create_heatmap(df, [1950, CURRENT_YEAR]),
+            outputs=heatmap
+        )
+        
+        demo.load(
+            fn=create_regional_comparison,
+            outputs=region_plot
+        )
+    
+    return demo
+
+if __name__ == "__main__":
+    dashboard = create_dashboard()
+    dashboard.launch()