LATAM: Daily mobility data for cities, metro areas, districts, provinces, and states

The datasets are split by census block, cities, counties, districts, provinces, and states. The typical dataset includes the below fields. Column numbers, Data attribute, Description 1, device_id, hashed anonymized unique id per moving device 2, origin_geoid, geohash id of the origin grid cell 3, destination_geoid, geohash id of the destination grid cell 4, origin_lat, origin latitude with 4-to-5 decimal precision 5, origin_long, origin longitude with 4-to-5 decimal precision 6, destination_lat, destination latitude with 5-to-6 decimal precision 7, destination_lon, destination longitude with 5-to-6 decimal precision 8, start_timestamp, start timestamp / local time 9, end_timestamp, end timestamp / local time 10, origin_shape_zone, customer provided origin shape id, zone or census block id 11, destination_shape_zone, customer provided destination shape id, zone or census block id 12, trip_distance, inferred distance traveled in meters, as the crow flies 13, trip_duration, inferred duration of the trip in seconds 14, trip_speed, inferred speed of the trip in meters per second 15, hour_of_day, hour of day of trip start (0-23) 16, time_period, time period of trip start (morning, afternoon, evening, night) 17, day_of_week, day of week of trip start(mon, tue, wed, thu, fri, sat, sun) 18, year, year of trip start 19, iso_week, iso week of the trip 20, iso_week_start_date, start date of the iso week 21, iso_week_end_date, end date of the iso week 22, travel_mode, mode of travel (walking, driving, bicycling, etc) 23, trip_event, trip or segment events (start, route, end, start-end) 24, trip_id, trip identifier (unique for each batch of results) 25, origin_city_block_id, census block id for the trip origin point 26, destination_city_block_id, census block id for the trip destination point 27, origin_city_block_name, census block name for the trip origin point 28, destination_city_block_name, census block name for the trip destination point 29, trip_scaled_ratio, ratio used to scale up each trip, for example, a trip_scaled_ratio value of 10 means that 1 original trip was scaled up to 10 trips 30, route_geojson, geojson line representing trip route trajectory or geometry The datasets can be processed and enhanced to also include places, POI visitation patterns, hour-of-day patterns, weekday patterns, weekend patterns, dwell time inferences, and macro movement trends. The dataset is delivered as gzipped CSV archive files that are uploaded to your AWS s3 bucket upon request.

数据集按人口普查区块（census block）、城市、县、区、省及州划分。典型数据集包含以下字段：

列号 | 数据属性 | 描述
1 | device_id | 每个移动设备的匿名哈希唯一标识
2 | origin_geoid | 起点网格单元的地理哈希标识（geohash ID）
3 | destination_geoid | 终点网格单元的地理哈希标识（geohash ID）
4 | origin_lat | 起点纬度，精度为4至5位小数
5 | origin_long | 起点经度，精度为4至5位小数
6 | destination_lat | 终点纬度，精度为5至6位小数
7 | destination_lon | 终点经度，精度为5至6位小数
8 | start_timestamp | 起点时间戳/本地时间
9 | end_timestamp | 终点时间戳/本地时间
10 | origin_shape_zone | 客户提供的起点形状ID、区域或人口普查区块ID
11 | destination_shape_zone | 客户提供的终点形状ID、区域或人口普查区块ID
12 | trip_distance | 推断的直线距离（单位：米）
13 | trip_duration | 推断的行程时长（单位：秒）
14 | trip_speed | 推断的行程速度（单位：米/秒）
15 | hour_of_day | 行程开始的小时（0-23）
16 | time_period | 行程开始的时间段（早晨、下午、晚上、夜间）
17 | day_of_week | 行程开始的星期几（周一、周二、周三、周四、周五、周六、周日）
18 | year | 行程开始的年份
19 | iso_week | 行程的ISO周数
20 | iso_week_start_date | ISO周的开始日期
21 | iso_week_end_date | ISO周的结束日期
22 | travel_mode | 出行方式（步行、驾驶、骑行等）
23 | trip_event | 行程或路段事件（开始、路线、结束、起止）
24 | trip_id | 行程标识（每批结果唯一）
25 | origin_city_block_id | 行程起点的人口普查区块ID
26 | destination_city_block_id | 行程终点的人口普查区块ID
27 | origin_city_block_name | 行程起点的人口普查区块名称
28 | destination_city_block_name | 行程终点的人口普查区块名称
29 | trip_scaled_ratio | 用于放大每个行程的比例（例如，值为10表示将1条原始行程放大为10条）
30 | route_geojson | 表示行程路线轨迹或几何形状的GeoJSON线（GeoJSON line）

数据集可通过处理与增强，进一步包含地点、兴趣点（POI）访问模式、时段模式、工作日模式、周末模式、停留时间（dwell time）推断及宏观移动趋势。数据集以压缩CSV归档文件（gzipped CSV archive files）形式交付，可根据请求上传至您的AWS S3存储桶。

搜集汇总

背景与挑战

背景概述

该数据集提供拉丁美洲城市、大都市区、区县、省份和州的日常移动性数据，包含行程起终点地理坐标、时间戳、距离、持续时间、出行模式等多维字段。数据可按人口普查区块等层级分割，支持进一步处理以分析地点访问模式、时间趋势和宏观移动行为，并以gzipped CSV格式通过AWS S3交付。