{ "zh": { "lang_label": "中文", "other_lang_label": "English", "title": "吉隆口岸 2026-08-26 冰岩崩灾害 · ERA5-Land 气象背景", "h1": "吉隆口岸 2026-08-26 冰岩崩灾害 · ERA5-Land 日尺度气象背景", "subtitle": "中尼边境东林藏布(Lhende Khola)流域 · 时间一律为北京时间(UTC+8)", "fact_event": "事发", "fact_event_v": "2026-08-26 10:52 崩塌 · 约 10:59 抵达口岸", "fact_elev": "源区 / 口岸高程", "fact_cover": "ERA5-Land 覆盖", "fact_cover_join": " 至 ", "fact_grid": "网格", "fact_grid_v": "0.1° · 回归用 {box} 格点 · 区域平均加权 {poly} 格点", "banner": "注意:灾害当天(2026-08-26)尚未纳入本页数据。ERA5-Land 近实时版本(ERA5-Land-T)落后实时约 5 天,本页生成于 {generated},档期止于 {end}。下方“事发前条件诊断”因此以档期末日为锚点计算,反映的是崩塌前若干天的条件,而非当天。数据补齐后重跑 notebook 03 即可自动更新。", "map_title": "灾害位置与路径", "map_hint": "底图可切换 · 图层可勾选 · 点击要素查看详情", "base_imagery": "卫星影像 (Esri)", "base_topo": "地形图 (OpenTopoMap)", "layer_polygon": "气象分析区域", "layer_path": "灾害路径", "layer_cells": "ERA5-Land 格点(按高程着色)", "cell_title": "ERA5-Land 格点", "cell_elev": "格点高程", "cell_role_lapse": "气温递减率回归", "cell_role_lapse_v": "参与", "cell_role_lapse_x": "全部 {n} 个格点都参与,回归要的是尽可能大的高程跨度", "cell_role_precip": "降水区域平均", "cell_role_precip_in": "参与,权重 {w},占 {share}%", "cell_role_precip_out": "不参与", "cell_role_precip_x_in": "权重为该格点被分析多边形覆盖的面积比例乘 cos(纬度)", "cell_role_precip_x_out": "该格点在分析多边形之外,与流域降水无关,故不计入。这不代表它没用,它仍参与上面的递减率回归", "poi_elev": "高程约", "poi_localtime": "(北京时间)", "path_source": "来源", "legend_surveyed": "实测河道段 16.9 km", "legend_schematic": "源头示意直线 4.8 km(无河道数据)", "legend_sketch": "下游洪水传播段(人工勾绘)", "legend_polygon": "气象分析区域(路径 4 km 缓冲)", "legend_source": "冰岩崩源区", "legend_port": "吉隆口岸", "legend_elev": "格点填色为高程 1.5 至 5.9 km", "map_note": "几何说明:运移路径的主体(16.9 km)取自实测河道线,不是人工勾绘;上游 4.8 km 因无河道数据以直线示意,下游洪水段为人工勾绘,三者在图上以线型区分。分析区域是沿该路径两侧 4 km 的缓冲区(约 200 km²),不是 DEM 流向提取的真实汇水区,取 4 km 是因为面积与该集水区量级相当,实测 3 至 5 km 之间各诊断量差异在 1% 以内。源区点位取自路径起点,与公开报道的位置相距约 2.4 km,但落在同一 ERA5-Land 格点内。堰塞湖官方通报位于口岸上游约十几公里、海拔约 2950 m 的错坚河与普热普强藏布交汇处,因未公布坐标,本图不标点位。", "range_label": "时间范围", "range_7": "近 1 周", "range_30": "近 1 月", "range_365": "近 1 年", "range_3650": "近 10 年", "range_all": "全部", "range_hint": "也可用滚轮缩放 · 拖拽平移", "chart_temp": "日均气温", "chart_precip": "日降水量", "unit_c": "°C", "unit_mm": "mm", "series_source": "源区 {elev} m(订正)", "series_port": "口岸 {elev} m(订正)", "series_area": "区域平均(未订正)", "series_clim": "源区气候态 {y0}-{y1}", "series_precip": "日降水(区域平均)", "series_precip7": "7 日滑动累计", "event_mark": "08-26 冰岩崩", "no_data": "无数据", "chart_note": "口径:气温为点提取后用 ERA5-Land 自身逐日反演的递减率订正到目标高程,递减率由分析框内 {box} 个格点的 t2m 对高程做最小二乘回归得到(中位数 {lapse} K/km,{ok}% 的日期通过质控)。降水为分析区域内格点按重叠面积与 cos(纬度) 加权的平均,不做高程订正。两图缩放与十字准星联动。", "chart_fail": "图表未能渲染:{msg}。图表库需要联网从 cdnjs 或 jsDelivr 加载,请检查网络后刷新。地图与下方诊断不受影响。", "diag_title": "事发前条件诊断", "diag_anchor": "锚点 {date}", "diag_anchor_event": "(灾害当天)", "diag_anchor_end": "(档期末日)", "diag_pct": "近 {n} 年同一时刻窗口第 {v} 百分位", "diag_pct_day": "近 {n} 年同期第 {v} 百分位", "diag_pct_none": "样本不足", "diag_precip_same_day": "崩塌前当日降水", "diag_precip_same_day_x": "{start} 至 {cutoff}。当日总量 {total} mm,绝大部分落在崩塌之后", "diag_precip_window": "崩塌前 {h} 小时降水", "diag_window_x": "截至 {cutoff},{pct}", "diag_t_source": "源区 {elev} m 当日均气温", "diag_t_source_x": "距平 {anom} °C(相对 {y0}-{y1} 同期)", "diag_fl_day": "当日 0 °C 等温线高度", "diag_fl_day_x": "日历日均值,高于源区 {d} m", "diag_fl_day_x_none": "无有效回归", "diag_fl_window": "崩塌前 {h} 小时等温线高度", "diag_fl_window_x": "高于源区 {d} m,{pct}", "diag_pdd": "前 {d} 天正积温 PDD", "diag_pdd_x": "源区 {elev} m 高程,{pct}", "foot_data_label": "数据", "foot_data": "ERA5-Land(ECMWF / Copernicus C3S,0.1°),取自 reanalysis-era5-land-timeseries 逐小时场,气温与降水同源。两者均在本地按 UTC+08:00 日界聚合:气温取日均、日最高、日最低,降水直接求和。该服务提供的是逐小时增量而非累积量,因此不做解累积,这一点每次运行都会与本地 ERA5-Land 月尺度存档比对复核。格点高程由 ERA5-Land 位势场换算。档期 {start} 至 {end},生成于 {generated}。", "foot_limit_label": "局限", "foot_limit": "ERA5-Land 网格约 9 km,无法分辨实际失稳坡面;本页刻画的是区域尺度背景条件,不构成对触发机制的归因结论。{month} 的数据属 ERA5-Land-T 初始版本,约两个月后会被最终版重算覆盖。", "foot_geom_label": "几何", "foot_geom": "运移路径主体取自实测河道线,上游 4.8 km 与下游洪水段为示意;分析区域为该路径两侧 4 km 缓冲区,非 DEM 汇水区。详见地图下方说明。", "foot_tz": "时间一律为北京时间(UTC+8)。需要联网加载地图瓦片与图表库。" }, "en": { "lang_label": "English", "other_lang_label": "中文", "title": "Gyirong Port rock-ice avalanche, 26 August 2026 · ERA5-Land context", "h1": "Gyirong Port rock-ice avalanche, 26 August 2026 · ERA5-Land daily meteorological context", "subtitle": "Lhende Khola catchment on the China–Nepal border. All times are Beijing time (UTC+8).", "fact_event": "Event", "fact_event_v": "Collapse 2026-08-26 10:52, reached the port about 10:59", "fact_elev": "Source / port elevation", "fact_cover": "ERA5-Land coverage", "fact_cover_join": " to ", "fact_grid": "Grid", "fact_grid_v": "0.1°, {box} cells in the lapse-rate fit, {poly} weighted for the area mean", "banner": "Note. The event day (2026-08-26) is not yet in this page's data. The near-real-time release ERA5-Land-T trails real time by about five days. This page was built on {generated} and the archive ends on {end}. The diagnostics below are therefore anchored on the last archived day rather than on the event itself. Re-run notebook 03 once the data arrives.", "map_title": "Location and runout path", "map_hint": "Switch basemaps, toggle layers, click a feature for detail", "base_imagery": "Satellite imagery (Esri)", "base_topo": "Topographic (OpenTopoMap)", "layer_polygon": "Meteorological analysis area", "layer_path": "Runout path", "layer_cells": "ERA5-Land cells (coloured by elevation)", "cell_title": "ERA5-Land grid cell", "cell_elev": "Cell elevation", "cell_role_lapse": "Lapse-rate regression", "cell_role_lapse_v": "included", "cell_role_lapse_x": "All {n} cells are used. The fit wants the widest possible elevation range.", "cell_role_precip": "Precipitation area mean", "cell_role_precip_in": "included, weight {w}, {share}% of the mean", "cell_role_precip_out": "not included", "cell_role_precip_x_in": "Weight is the fraction of the cell covered by the analysis polygon, times cos(latitude).", "cell_role_precip_x_out": "This cell lies outside the analysis polygon, so it carries no catchment rainfall. It is still used in the lapse-rate regression above.", "poi_elev": "elevation about", "poi_localtime": "(Beijing time)", "path_source": "Source", "legend_surveyed": "Surveyed channel, 16.9 km", "legend_schematic": "Headwater straight line, 4.8 km (no channel data)", "legend_sketch": "Downstream flood reach (hand drawn)", "legend_polygon": "Analysis area (4 km buffer on the path)", "legend_source": "Avalanche source", "legend_port": "Gyirong Port", "legend_elev": "Cell colour spans 1.5 to 5.9 km elevation", "map_note": "Geometry. The main runout reach (16.9 km) follows a surveyed channel line and is not hand drawn. The upper 4.8 km has no channel data and is shown as a straight line. The downstream flood reach is hand drawn. The three are distinguished by line style. The analysis area is a 4 km buffer around that path, about 200 km², and is not a DEM-derived catchment. A 4 km radius was chosen because the resulting area matches the scale of this catchment. Between 3 and 5 km the diagnostics differ by less than 1%. The source marker sits at the start of the supplied path, about 2.4 km from the position given in published reports, but both fall in the same ERA5-Land cell. Chinese authorities placed the barrier lake at the confluence of the Chhochen Khola and the Purepu Tsangpo, roughly ten to twenty kilometres upstream of the port at about 2950 m. No coordinates were released, so it is not marked here.", "range_label": "Range", "range_7": "1 week", "range_30": "1 month", "range_365": "1 year", "range_3650": "10 years", "range_all": "All", "range_hint": "Scroll to zoom, drag to pan", "chart_temp": "Daily mean air temperature", "chart_precip": "Daily precipitation", "unit_c": "°C", "unit_mm": "mm", "series_source": "Source zone, {elev} m (corrected)", "series_port": "Port, {elev} m (corrected)", "series_area": "Area mean (uncorrected)", "series_clim": "Source zone climatology {y0}–{y1}", "series_precip": "Daily rainfall (area mean)", "series_precip7": "7-day running total", "event_mark": "08-26 collapse", "no_data": "no data", "chart_note": "Definitions. Temperature is taken from the nearest grid cell and moved to the target elevation using a lapse rate fitted from ERA5-Land itself. The rate comes from a least-squares fit of t2m against elevation across the {box} cells in the analysis box, with a median of {lapse} K/km. {ok}% of days pass quality control. Precipitation is the area mean over the analysis polygon, weighted by overlap area and cos(latitude), with no elevation correction. The two charts share zoom and crosshair.", "chart_fail": "Charts failed to render. {msg}. The chart library loads from cdnjs or jsDelivr and needs a network connection. The map and the diagnostics below are unaffected.", "diag_title": "Conditions before the collapse", "diag_anchor": "Anchored on {date}", "diag_anchor_event": " (event day)", "diag_anchor_end": " (last archived day)", "diag_pct": "{v}th percentile among the same clock window across {n} years", "diag_pct_day": "{v}th percentile among the same calendar window across {n} years", "diag_pct_none": "too few samples", "diag_precip_same_day": "Same-day rain before the collapse", "diag_precip_same_day_x": "{start} to {cutoff}. The day totalled {total} mm, almost all of it after the collapse.", "diag_precip_window": "Rain in the {h} h before the collapse", "diag_window_x": "Ending {cutoff}. {pct}", "diag_t_source": "Source zone daily mean at {elev} m", "diag_t_source_x": "Anomaly {anom} °C against the {y0}–{y1} mean for the same day", "diag_fl_day": "Freezing level on the event day", "diag_fl_day_x": "Calendar-day mean, {d} m above the source zone", "diag_fl_day_x_none": "no valid fit", "diag_fl_window": "Freezing level in the {h} h before the collapse", "diag_fl_window_x": "{d} m above the source zone. {pct}", "diag_pdd": "Positive degree days over the previous {d} days", "diag_pdd_x": "At the source zone elevation of {elev} m. {pct}", "foot_data_label": "Data", "foot_data": "ERA5-Land (ECMWF / Copernicus C3S, 0.1°), taken as hourly fields from reanalysis-era5-land-timeseries so that temperature and precipitation share one source. Both are aggregated locally onto a UTC+08:00 day boundary. Temperature gives daily mean, maximum and minimum, and precipitation is summed directly. That service delivers hourly increments rather than accumulations, so no de-accumulation is applied. Every run re-checks this against the local ERA5-Land monthly archive. Cell elevation is derived from the ERA5-Land geopotential field. Archive {start} to {end}, built on {generated}.", "foot_limit_label": "Limitations", "foot_limit": "The ERA5-Land grid is about 9 km and cannot resolve the failure slope itself. This page describes regional background conditions and is not an attribution of the trigger. Data for {month} is the preliminary ERA5-Land-T release and will be recomputed in the final release about two months later.", "foot_geom_label": "Geometry", "foot_geom": "The main runout reach follows a surveyed channel line. The upper 4.8 km and the downstream flood reach are schematic. The analysis area is a 4 km buffer around that path and is not a DEM-derived catchment. See the note under the map.", "foot_tz": "All times are Beijing time (UTC+8). Map tiles and the chart library are loaded over the network." } }