数据与方法
本站是研究演示：展示海平面上升与沿海洪水在美国东海岸怎样触及公园和儿童游乐场。结果是筛查级估算，不是防洪安全评估，不是关闭或重新开放的决定，也不是官方统计。据此行动前请现场核实。
水位
静水位 = 潮位站 MHHW（1983–2001 基准期）+ NOAA 公布的 1992→2005 相对海平面偏移 + NOAA 2022 情景（相对 2005）+ 事件高出 MHHW 的量。全部换算到 NAVD88 米。
- 事件：每日高潮（全年为 MHHW；选了月份时为该月天文潮预报的最高潮）、小洪水高度、GEV 极值 1/10/100 年一遇（可选月份，体现季节差异）、本站历史最高事件。
- “小洪水”事件是高出 MHHW 的量等于今天 NOAA 小洪水阈值的水位，随海平面一起抬升；“高潮洪水天数”则用固定阈值，直接取 NOAA 展望。
- 历史最高事件是实测值，已含 1992 年到事发年的海平面上升；放到未来年份前先扣除这段（1992→2005 用 NOAA 偏移，之后用中情景）。总览地图与公园视图同一口径。
- 情景取 NOAA 中位数；地图与年份表不按月份，月份只作用于公园视图。
- 极值分布按平稳处理，随海平面整体平移；高潮洪水天数直接取 NOAA 年代展望，不再叠加。
- 播放的过程线：纽约历史最高事件为 Sandy 实测序列平移；每日高潮为天文潮预报；其余为示意形状，峰值来自 NOAA 统计。
淹没
NOAA SLR Viewer 3 m DEM 上的"静水 + 海洋连通"计算（与 NOAA SLR Viewer 同一方法）。每个格子记录它与海连通所需的最低水位。
- 不含：波浪与爬高、降雨与排水倒灌、地下水上升、防洪墙/闸门/新抬升工程、地面沉降的空间差异、动力过程与持续时间。
- DEM 采集时间早于部分新工程，纽约东河三处尤其如此。
设施
- 官方：NYC Parks 公园边界与儿童游乐区（DCA）、NYC OTI 建筑轮廓与屋顶高。
- OSM：OpenStreetMap 要素（游乐场、球场、树、长椅、步道等），完整度因城市而异。
- 推断：游乐场内无逐件设备记录时生成的占位设备，位置与类型非实测。
- 场景植被：OSM 的单棵树、行道树、树林和灌木按位置生成，只作背景，不计入设施；树林里的树是按面积均匀撒点，不是实测位置。
- 地面高程取自 DEM；"关键高度"为暂定值；检查任务依据 NYC Parks《Design and Planning for Flood Resiliency》(2017)，实际任务由管理方核定。
不确定性
- 80% 单侧余量：按 NOAA SLR Viewer 的置信方法（NOAA 2010《Mapping Inundation Uncertainty》），合成 DEM 源激光雷达误差与 VDatum 潮汐基准误差，σ = √(σDEM² + σ基准²)，余量为 ±0.84σ（单侧 80% 置信）。
- 游乐设备部件高度取自 CPSC《Public Playground Safety Handbook》（2025）与厂商规格，从铺装面量起。"已到""未到"表示至少 80% 可能；"可能到"介于 20% 与 80% 之间。只有下限的部件（秋千座底）不判"已到"。
- 区间不含情景差异；NOAA 2022 情景本身没有发生概率。
- 水深显示到 0.1 m。结果为筛查级，需现场核实。
- 垂直放大同时作用于地形、水深、设施与建筑高度。
东海岸总览
NOAA SLR Viewer 1、2、3、4、6、10 ft 连通淹没范围与 OSM 公园、自然保护地与游乐场叠加，先扣掉今天高潮时已是水面的部分（NOAA 0 ft 档），只算新增淹没的陆地。某一档新增淹没达到陆地的 10% 或 1 英亩（NOAA 制图的最小单元），该公园在这一档算“被淹”；新增淹没不足此数但超过 50 m² 的，算“边缘进水”（空心点）。按最近潮位站的情景换算年份。门槛是惯例取值（10% 见 MTC、UCS），不是物理量；每档的淹没比例显示在公园提示框里。NOAA 还为每档发布置信度栅格，分很可能湿（10 次中 8 次以上判对）、不确定、很可能干三类。只数很可能湿的格得到“很可能”被淹的档位；再加上不确定、且在 10 ft 与海连通的格，得到“可能”被淹的档位。被淹个数下方的区间是这两种算法的个数，提示框写出两者各从哪一档开始。OSM 不是官方清单，计数不是官方统计。地图放大后的蓝色淹没范围取 NOAA SLR Viewer 对应档位（0.5 ft 一档，最高 10 ft）：用地图中心最近潮位站的静水位，向下取档；只显示与海连通的淹没，不显示低洼未连通区。淡色是今天高潮时的水面（NOAA 0 ft 档），蓝色是超出它的新增淹没。点任意公园点，打开预先切好的场景（地形、OSM 要素、NAIP 影像）；没有预切的部分由浏览器现场生成，只用 OSM、NOAA 与 USGS 数据。
来源
Data & methods
This site is a research demo. It shows how sea-level rise and coastal floods reach parks and children's playgrounds on the US East Coast. The results are screening-level estimates. They are not a flood-safety assessment, a closure or reopening decision, or an official statistic. Check on site before you act on them.
Water level
Still-water level = tide-station MHHW (1983–2001 epoch) + NOAA's published 1992→2005 relative sea-level offset + NOAA 2022 scenario (relative to 2005) + the event height above MHHW. Everything is converted to NAVD88 metres.
- Events: daily high tide (MHHW over the year; with a month, that month's highest predicted tide), minor-flood height, GEV 1-, 10- and 100-year extremes (with a month option for seasonal differences), and the highest event on record at the station.
- The “Minor flood” event is a water level as far above MHHW as today's NOAA minor flood threshold, raised with sea level. The high-tide flood-day counts use the fixed threshold, straight from the NOAA outlook.
- The record event is an observation that already contains the sea-level rise from 1992 to its own year; that rise is removed before it is placed on a future year (NOAA offset for 1992→2005, the Intermediate scenario after). The map and the park view use the same rule.
- Scenarios are NOAA medians. The map and the years table use annual levels; the month applies to the park view only.
- Extreme-value distributions are treated as stationary and shift with mean sea level; high-tide flood days come straight from the NOAA decadal outlook and are not stacked again.
- Playback hydrographs: the New York record event is the observed Sandy series shifted; the daily high tide is the astronomical prediction; the others are schematic shapes with peaks from NOAA statistics.
Inundation
"Still water + ocean connectivity" on the NOAA SLR Viewer 3 m DEM (the same method as the NOAA SLR Viewer). Each cell stores the lowest water level at which it connects to the ocean.
- Not included: waves and run-up, rainfall and drainage backflow, groundwater rise, floodwalls, gates and newly raised works, spatial differences in land subsidence, dynamics and duration.
- The DEM was collected before some new works, especially at the three East River sites in New York.
Facilities
- Official: NYC Parks property boundaries and Children's Play Areas (DCA), NYC OTI building footprints and roof heights.
- OSM: OpenStreetMap features (playgrounds, pitches, trees, benches, paths, etc.); completeness varies by city.
- Inferred: placeholder equipment generated inside a playground when there is no item-level record; position and type are not surveyed.
- Scenery vegetation: OSM single trees, tree rows, woods and scrub, drawn as background only and not counted as facilities. Trees inside woods are spread evenly over the area, not surveyed positions.
- Ground elevation comes from the DEM; "critical heights" are provisional; inspection tasks follow NYC Parks, Design and Planning for Flood Resiliency (2017), and actual tasks are set by the park manager.
Uncertainty
- 80% one-sided margin: follows the NOAA SLR Viewer confidence method (NOAA 2010, Mapping Inundation Uncertainty), combining the source-lidar error of the DEM with the VDatum tidal-datum error, σ = √(σDEM² + σdatum²), margin ±0.84σ (one-sided 80% confidence).
- Play-equipment part heights come from the CPSC Public Playground Safety Handbook (2025) and manufacturer specifications, measured from the play surface. "Reached" and "Not reached" mean at least 80% likely; "Possible" is between 20% and 80%. Parts with only a minimum height (belt seat bottom) are never marked "Reached".
- The band does not include scenario differences; the NOAA 2022 scenarios themselves have no probabilities.
- Depths are shown to 0.1 m. Results are screening-level and need field checks.
- Vertical exaggeration applies to terrain, water depth, facilities and building heights alike.
East Coast overview
NOAA SLR Viewer connected-inundation extents at 1, 2, 3, 4, 6 and 10 ft are overlaid on OSM parks, nature reserves and playgrounds. Water already there at today's high tide (the NOAA 0 ft level) is removed first, so only newly flooded land counts. A park is flooded at a level when the new flooding covers 10% of its land or 1 acre (the NOAA mapping minimum unit); less than that but over 50 m² is edge flooding (hollow point). Years come from the scenarios of the nearest tide station. The thresholds are conventions (10%: MTC, UCS), not physical quantities; the share flooded at each level is shown in the park's tooltip. NOAA also publishes a confidence raster for each level: likely wet (right in more than 8 of 10 cases), low confidence, likely dry. Counting only likely-wet cells gives the level where a park is likely flooded; adding low-confidence cells that are ocean-connected at 10 ft gives the level where it is possibly flooded. The range under the flooded count is the count by each of these two, and the tooltip gives the level where each starts. OSM is not an official inventory, so the counts are not official statistics. When you zoom in, the blue flooded area is the NOAA SLR Viewer level (0.5 ft steps, 10 ft at most) for the still water at the tide station nearest the map centre, rounded down. It shows ocean-connected flooding only, not low-lying unconnected land. The pale fill is today's water at high tide (the NOAA 0 ft level); blue is the flooding beyond it. Click any park point to open its precut scene (terrain, OSM features, NAIP image); any part not precut is built by the browser on the spot, using only OSM, NOAA and USGS data.
Sources