5 ms·

Using Weather Data to Determine Your Perfect Vacation Destination

15 points by rishi 16y ago

brc 16y ago
This is a good idea in principle but hampered by lack of usable data. For one thing, average temperature is not much use for a holiday. Places that are weather dependent (tropical islands, beach holidays, ski holidays) generally will have the right temperature at the time of year you're thinking of going, but the actual experience of the trip is less temperature related than precipitation and wind related. A holiday on a tropical island with 5 days of rain is miserable even if it's warm. A skiing holiday with a blizzard isn't much fun, and being on a cloudy beach is only half as much fun as being on a sunny one. Unless you could mine a probability of sunshine in a reasonable forecast period - and this depends on your travel timeframe - the data is always going to be problematic. That's not a slight on the author, it's a problem inherent in weather prediction. And lets face it, some of the biggest computing power in the world goes into weather prediction. The cynic in me says you can get just as much information on potential destinations just looking at a globe and picking a hemisphere / latitude based on time of year, and gridding down to a set of locations that way.
- moontear 16y ago
  I totally agree with you. Whilst the basic notion that travel sites could do a better job at giving you suggestions based on your preferences - and temperature definitely being one factor - temperature is not everything. The author did a nice job collecting all the temperature information, but for an end user the simple names of the destinations don't mean much (where exactly is "Metehara" without asking Google?). I wouldn't find myself just looking for a temperature and in accordance to that look for destinations. Much more important for a holiday is the price range, so if I'd use the Excel file to find my perfect destination I would still have to find all the prices associated with the destinations (e.g. traveling to Ethiopia / Metehara will be quite expensive). All in all: Good start. Incorporate this data within a larger travel search engine and let the user have one more option to narrow down the search.
  - anish2424 16y ago
    I'm thinking of making all the locations that get outputted hyperlinked to Google Maps. What I'd really like is for the outputted list of destinations to be outputted onto Google Maps as pin drops - but I couldn't find any help on how to do that online. Any ideas?
- anish2424 16y ago
  Very good point brc. The next iteration of the spreadsheet will include average precipitation by location. Hopefully that can be used a gauge and be more helpful supplemented with average temperature. However, I understand that even with all the historical data simply being presented, it isn't really being used for predictive purposes. Nevertheless, hopefully it helps a few people to start thinking of destinations they might not have otherwise thought of.
pyre 16y ago
I'm curious why the Department of Energy has weather data. I thought that the NOAA was the 'weather agency.'
- interknot 16y ago
  The data in question appears to be simulated and intended for use in developing energy calculations for buildings: http://apps1.eere.energy.gov/buildings/energyplus/weatherdata_sources.cfm http://apps1.eere.energy.gov/buildings/energyplus/weatherdat... That said, DOE is probably keenly interested in gathering reliable weather data so that they can keep the power grid working.

15 points by rishi 16y ago

brc 16y ago

This is a good idea in principle but hampered by lack of usable data. For one thing, average temperature is not much use for a holiday. Places that are weather dependent (tropical islands, beach holidays, ski holidays) generally will have the right temperature at the time of year you're thinking of going, but the actual experience of the trip is less temperature related than precipitation and wind related. A holiday on a tropical island with 5 days of rain is miserable even if it's warm. A skiing holiday with a blizzard isn't much fun, and being on a cloudy beach is only half as much fun as being on a sunny one. Unless you could mine a probability of sunshine in a reasonable forecast period - and this depends on your travel timeframe - the data is always going to be problematic. That's not a slight on the author, it's a problem inherent in weather prediction. And lets face it, some of the biggest computing power in the world goes into weather prediction. The cynic in me says you can get just as much information on potential destinations just looking at a globe and picking a hemisphere / latitude based on time of year, and gridding down to a set of locations that way.

moontear 16y ago
I totally agree with you. Whilst the basic notion that travel sites could do a better job at giving you suggestions based on your preferences - and temperature definitely being one factor - temperature is not everything. The author did a nice job collecting all the temperature information, but for an end user the simple names of the destinations don't mean much (where exactly is "Metehara" without asking Google?). I wouldn't find myself just looking for a temperature and in accordance to that look for destinations. Much more important for a holiday is the price range, so if I'd use the Excel file to find my perfect destination I would still have to find all the prices associated with the destinations (e.g. traveling to Ethiopia / Metehara will be quite expensive). All in all: Good start. Incorporate this data within a larger travel search engine and let the user have one more option to narrow down the search.
- anish2424 16y ago
  I'm thinking of making all the locations that get outputted hyperlinked to Google Maps. What I'd really like is for the outputted list of destinations to be outputted onto Google Maps as pin drops - but I couldn't find any help on how to do that online. Any ideas?
anish2424 16y ago
Very good point brc. The next iteration of the spreadsheet will include average precipitation by location. Hopefully that can be used a gauge and be more helpful supplemented with average temperature. However, I understand that even with all the historical data simply being presented, it isn't really being used for predictive purposes. Nevertheless, hopefully it helps a few people to start thinking of destinations they might not have otherwise thought of.

pyre 16y ago

I'm curious why the Department of Energy has weather data. I thought that the NOAA was the 'weather agency.'

interknot 16y ago
The data in question appears to be simulated and intended for use in developing energy calculations for buildings: http://apps1.eere.energy.gov/buildings/energyplus/weatherdata_sources.cfm http://apps1.eere.energy.gov/buildings/energyplus/weatherdat... That said, DOE is probably keenly interested in gathering reliable weather data so that they can keep the power grid working.