https://en.wikipedia.org/wiki/Great-circle_distance

下面是一个使用JSON数据的很好的实现: https://github.com/agap/llttz

public TimeZone nearestTimeZone(Location node) {
    double bestDistance = Double.MAX_VALUE;
    Location bestGuess = timeZones.get(0);

    for (Location current : timeZones.subList(1, timeZones.size())) {
        double newDistance = distanceInKilometers(node, current);

        if (newDistance < bestDistance) {
            bestDistance = newDistance;
            bestGuess = current;
        }
    }

    return java.util.TimeZone.getTimeZone(bestGuess.getZone());
}

  protected double distanceInKilometers(final double latFrom, final double lonFrom, final double latTo, final double lonTo) {
        final double meridianLength = 111.1;
        return meridianLength * centralAngle(latFrom, lonFrom, latTo, lonTo);
    }

    protected double centralAngle(final Location from, final Location to) {
        return centralAngle(from.getLatitude(), from.getLongitude(), to.getLatitude(), to.getLongitude());
    }

    protected double centralAngle(final double latFrom, final double lonFrom, final double latTo, final double lonTo) {
        final double latFromRad = toRadians(latFrom),
                lonFromRad = toRadians(lonFrom),
                latToRad   = toRadians(latTo),
                lonToRad   = toRadians(lonTo);

        final double centralAngle = toDegrees(acos(sin(latFromRad) * sin(latToRad) + cos(latFromRad) * cos(latToRad) * cos(lonToRad - lonFromRad)));

        return centralAngle <= 180.0 ? centralAngle : (360.0 - centralAngle);
    }

    protected double distanceInKilometers(final Location from, final Location to) {
        return distanceInKilometers(from.getLatitude(), from.getLongitude(), to.getLatitude(), to.getLongitude());
    }
}

It's indeed important to recognize that this a more complicated problem than most would suspect. In practice many of us are also willing to accept a working set of code that works for "as many cases as possible", where at least its fatal issues can be identified and minimized collectively. So I post this with all of that and the spirit of the OP in mind. Finally, for practical value to others who are trying to convert GPS to timezone with the end goal of having a location-sensitive time object (and more importantly to help advance the quality of average implementations with time objects that follow from this wiki) here is what I generated in Python (please feel free to edit):

import pytz
from datetime import datetime
from tzwhere import tzwhere

def timezoned_unixtime(latitude, longitude, dt):
    tzw = tzwhere.tzwhere()
    timezone_str = tzw.tzNameAt(latitude, longitude)
    timezone = pytz.timezone(timezone_str)
    timezone_aware_datetime = timezone.localize(dt, is_dst=None)
    unix_time = (timezone_aware_datetime - datetime(1970, 1, 1, tzinfo=pytz.utc)).total_seconds()
    return unix_time

dt = datetime(year=2017, month=1, day=17, hour=12, minute=0, second=0)
print timezoned_unixtime(latitude=40.747854, longitude=-74.004733, dt=dt)

下面是如何使用谷歌的脚本编辑器来获取gsheet中的timezoneName和timeZoneId。

步骤1。获取谷歌的时区API的API键

步骤2。创建一个新的gsheet。在“工具”菜单下单击“脚本编辑器”。添加如下代码:

function getTimezone(lat, long) {  
  var apiKey = 'INSERTAPIKEYHERE'
  var url = 'https://maps.googleapis.com/maps/api/timezone/json?location=' + lat + ',' + long + '&timestamp=1331161200&key=' + apiKey 
  var response = UrlFetchApp.fetch(url);
  var data = JSON.parse(response.getContentText());
  return data["timeZoneName"];
}

步骤3。保存并发布getTimezone()函数，并如上图所示使用它。

https://en.wikipedia.org/wiki/Great-circle_distance

下面是一个使用JSON数据的很好的实现: https://github.com/agap/llttz

public TimeZone nearestTimeZone(Location node) {
    double bestDistance = Double.MAX_VALUE;
    Location bestGuess = timeZones.get(0);

    for (Location current : timeZones.subList(1, timeZones.size())) {
        double newDistance = distanceInKilometers(node, current);

        if (newDistance < bestDistance) {
            bestDistance = newDistance;
            bestGuess = current;
        }
    }

    return java.util.TimeZone.getTimeZone(bestGuess.getZone());
}

  protected double distanceInKilometers(final double latFrom, final double lonFrom, final double latTo, final double lonTo) {
        final double meridianLength = 111.1;
        return meridianLength * centralAngle(latFrom, lonFrom, latTo, lonTo);
    }

    protected double centralAngle(final Location from, final Location to) {
        return centralAngle(from.getLatitude(), from.getLongitude(), to.getLatitude(), to.getLongitude());
    }

    protected double centralAngle(final double latFrom, final double lonFrom, final double latTo, final double lonTo) {
        final double latFromRad = toRadians(latFrom),
                lonFromRad = toRadians(lonFrom),
                latToRad   = toRadians(latTo),
                lonToRad   = toRadians(lonTo);

        final double centralAngle = toDegrees(acos(sin(latFromRad) * sin(latToRad) + cos(latFromRad) * cos(latToRad) * cos(lonToRad - lonFromRad)));

        return centralAngle <= 180.0 ? centralAngle : (360.0 - centralAngle);
    }

    protected double distanceInKilometers(final Location from, final Location to) {
        return distanceInKilometers(from.getLatitude(), from.getLongitude(), to.getLatitude(), to.getLongitude());
    }
}

我写了一个包https://github.com/ringsaturn/tzf支持获取时区在Go&Python和非常快:

package main

import (
    "fmt"

    "github.com/ringsaturn/tzf"
)

func main() {
    finder, err := tzf.NewDefaultFinder()
    if err != nil {
        panic(err)
    }
    fmt.Println(finder.GetTimezoneName(116.6386, 40.0786))
}

Python https://github.com/ringsaturn/tzfpy sample:

from tzfpy import get_tz

print(get_tz(121.4737, 31.2305))

Rust https://github.com/ringsaturn/tzf-rs样品:

use tzf_rs::DefaultFinder;

fn main() {
    let finder = DefaultFinder::new();

    print!("{:?}\n", DefaultFinder.get_tz_name(116.3883, 39.9289));
}

对于我们这些使用Javascript并希望通过谷歌api从邮政编码获取时区的人来说，这里有一个方法。

通过地理定位获取lat/lng 通过这个参数获取时区 进入时区API。 此处使用Luxon进行时区转换。

注意:我的理解是，邮政编码在不同的国家并不是唯一的，所以这可能最适合在美国使用。

const googleMapsClient; // instantiate your client here
const zipcode = '90210'
const myDateThatNeedsTZAdjustment; // define your date that needs adjusting
// fetch lat/lng from google api by zipcode
const geocodeResponse = await googleMapsClient.geocode({ address: zipcode }).asPromise();
if (geocodeResponse.json.status === 'OK') {
  lat = geocodeResponse.json.results[0].geometry.location.lat;
  lng = geocodeResponse.json.results[0].geometry.location.lng;
} else {
  console.log('Geocode was not successful for the following reason: ' + status);
}

// prepare lat/lng and timestamp of profile created_at to fetch time zone
const location = `${lat},${lng}`;
const timestamp = new Date().valueOf() / 1000;
const timezoneResponse = await googleMapsClient
  .timezone({ location: location, timestamp: timestamp })
  .asPromise();

const timeZoneId = timezoneResponse.json.timeZoneId;
// adjust by setting timezone
const timezoneAdjustedDate = DateTime.fromJSDate(
  myDateThatNeedsTZAdjustment
).setZone(timeZoneId);