feat: enhance intensity map visualization by adjusting station opacity based on P-wave arrival time

app.py

@@ -694,9 +694,14 @@ def get_intensity_color(intensity):
 
 def create_intensity_map(
         pga_list, target_names, epicenter_lat=None, epicenter_lon=None,
-        selected_stations=None
+        selected_stations=None, duration=None, first_pick=None
 ):
-    """使用 Plotly 創建互動式震度分布地圖（合併輸入測站與預測震度）"""
+    """使用 Plotly 創建互動式震度分布地圖（合併輸入測站與預測震度）
+
+    輸入測站的透明度根據 P 波到時是否在時間窗內調整：
+    - P 波在時間窗內：較不透明 (opacity=0.9)
+    - P 波在時間窗外：較透明 (opacity=0.3)
+    """
 
     # 按震度等級分組資料
     intensity_groups = {
@@ -736,36 +741,77 @@ def create_intensity_map(
     # 創建 Plotly 地圖
     fig = go.Figure()
 
-    # 【底層】添加輸入測站（灰色無填充圓圈，不搶視覺焦點）
+    # 【底層】添加輸入測站（根據 P 波時間點是否在時間窗內調整透明度）
     if selected_stations:
-        input_station_lats = []
-        input_station_lons = []
-        input_station_texts = []
+        # 分離 P 波在時間窗內和時間窗外的測站
+        stations_in_window = {"lat": [], "lon": [], "text": []}
+        stations_out_window = {"lat": [], "lon": [], "text": []}
+
+        # 計算時間窗範圍
+        end_time = first_pick + duration if first_pick is not None and duration is not None else None
 
         for station_data in selected_stations:
-            input_station_lats.append(station_data["latitude"])
-            input_station_lons.append(station_data["longitude"])
-            input_station_texts.append(
-                f"{station_data['station']}<br>"
+            lat = station_data["latitude"]
+            lon = station_data["longitude"]
+            station_name = station_data["station"]
+            p_arrival_time = station_data.get("p_arrival_time")
+
+            # 判斷 P 波是否在時間窗內
+            in_window = False
+            if end_time is not None and p_arrival_time is not None:
+                in_window = (0 <= p_arrival_time <= end_time)
+
+            hover_text = (
+                f"{station_name}<br>"
                 f"輸入測站<br>"
-                f"位置: ({station_data['latitude']:.3f}, {station_data['longitude']:.3f})"
+                f"P 波到時: {p_arrival_time:.2f}s<br>" if p_arrival_time is not None else f"{station_name}<br>輸入測站<br>"
+                f"位置: ({lat:.3f}, {lon:.3f})"
             )
 
-        fig.add_trace(
-            go.Scattermap(
-                lat=input_station_lats,
-                lon=input_station_lons,
-                mode="markers",
-                marker=dict(
-                    size=8,
-                    color="rgba(128, 128, 128, 0.7)",  # 半透明灰色
-                ),
-                text=input_station_texts,
-                hovertemplate="%{text}<extra></extra>",
-                name="輸入測站",
-                showlegend=True,
+            if in_window:
+                stations_in_window["lat"].append(lat)
+                stations_in_window["lon"].append(lon)
+                stations_in_window["text"].append(hover_text)
+            else:
+                stations_out_window["lat"].append(lat)
+                stations_out_window["lon"].append(lon)
+                stations_out_window["text"].append(hover_text)
+
+        # 添加時間窗內的測站（較不透明）
+        if stations_in_window["lat"]:
+            fig.add_trace(
+                go.Scattermap(
+                    lat=stations_in_window["lat"],
+                    lon=stations_in_window["lon"],
+                    mode="markers",
+                    marker=dict(
+                        size=8,
+                        color="rgba(128, 128, 128, 0.9)",  # 較不透明
+                    ),
+                    text=stations_in_window["text"],
+                    hovertemplate="%{text}<extra></extra>",
+                    name="輸入測站 (P波在窗內)",
+                    showlegend=True,
+                )
+            )
+
+        # 添加時間窗外的測站（較透明）
+        if stations_out_window["lat"]:
+            fig.add_trace(
+                go.Scattermap(
+                    lat=stations_out_window["lat"],
+                    lon=stations_out_window["lon"],
+                    mode="markers",
+                    marker=dict(
+                        size=8,
+                        color="rgba(128, 128, 128, 0.3)",  # 較透明
+                    ),
+                    text=stations_out_window["text"],
+                    hovertemplate="%{text}<extra></extra>",
+                    name="輸入測站 (P波在窗外)",
+                    showlegend=True,
+                )
             )
-        )
 
     # 【頂層】添加各震度等級的測站（預測結果）
     intensity_labels = ["0", "1", "2", "3", "4", "5-", "5+", "6-", "6+", "7"]
@@ -945,7 +991,7 @@ def step2_extract_and_plot_waveforms(cached_stream, cached_stations, event_name,
 
 # ============ 步驟 3：執行模型推論（使用快取的波形）============
 def step3_predict_intensity(cached_waveforms, cached_station_info, cached_stations,
-                            event_name):
+                            event_name, duration):
     """
     步驟 3：執行震度預測
 
@@ -963,6 +1009,7 @@ def step3_predict_intensity(cached_waveforms, cached_station_info, cached_statio
 
         epicenter_lat = earthquake_metadata[event_name]["epicenter_lat"]
         epicenter_lon = earthquake_metadata[event_name]["epicenter_lon"]
+        first_pick = earthquake_metadata[event_name]["first_pick"]
 
         logger.info("[步驟 3] 開始模型推論...")
 
@@ -1046,15 +1093,13 @@ def step3_predict_intensity(cached_waveforms, cached_station_info, cached_statio
         target_names = all_target_names
 
         # 繪製互動式地圖（固定高度 800）- 合併輸入測站與預測震度
+        # 根據 P 波到時是否在時間窗內調整輸入測站透明度
         intensity_map = create_intensity_map(
             pga_list, target_names, epicenter_lat, epicenter_lon,
-            selected_stations=cached_stations
+            selected_stations=cached_stations, duration=duration, first_pick=first_pick
         )
 
-        # 生成警報文字報告 (duration from waveform length: 3000 samples / 100 Hz = 30 seconds)
-        duration = cached_waveforms[0].shape[
-                       0] / 100 if cached_waveforms is not None and len(
-            cached_waveforms) > 0 else 30
+        # 生成警報文字報告
         alert_report = generate_earthquake_alert_report(
             pga_list, target_names, event_name, duration
         )
@@ -1100,6 +1145,7 @@ with gr.Blocks(title="TTSAM 震度預測系統", fill_height=True) as demo:
             max_lines=7,
             interactive=False,
             show_copy_button=False,
+            autoscroll=False,
         )
 
         waveform_plot = gr.Plot(
@@ -1140,7 +1186,7 @@ with gr.Blocks(title="TTSAM 震度預測系統", fill_height=True) as demo:
         outputs=[cached_waveforms, cached_station_info, waveform_plot]
     ).then(  # 鏈式觸發步驟 3
         fn=step3_predict_intensity,
-        inputs=[cached_waveforms, cached_station_info, cached_stations, event_dropdown],
+        inputs=[cached_waveforms, cached_station_info, cached_stations, event_dropdown, duration_slider],
         outputs=[predicted_intensity_map, alert_textbox]
     )
 
@@ -1151,7 +1197,7 @@ with gr.Blocks(title="TTSAM 震度預測系統", fill_height=True) as demo:
         outputs=[cached_waveforms, cached_station_info, waveform_plot]
     ).then(  # 鏈式觸發步驟 3
         fn=step3_predict_intensity,
-        inputs=[cached_waveforms, cached_station_info, cached_stations, event_dropdown],
+        inputs=[cached_waveforms, cached_station_info, cached_stations, event_dropdown, duration_slider],
         outputs=[predicted_intensity_map, alert_textbox]
     )
 
@@ -1170,7 +1216,7 @@ with gr.Blocks(title="TTSAM 震度預測系統", fill_height=True) as demo:
         outputs=[cached_waveforms, cached_station_info, waveform_plot]
     ).then(
         fn=step3_predict_intensity,
-        inputs=[cached_waveforms, cached_station_info, cached_stations, event_dropdown],
+        inputs=[cached_waveforms, cached_station_info, cached_stations, event_dropdown, duration_slider],
         outputs=[predicted_intensity_map, alert_textbox]
     )