As promised:

(*Get data*)
sunset = Sunset[Entity["City", {"NewHaven", "Connecticut", "UnitedStates"}], DateRange[{2015, 1, 1}, {2015, 12, 31}]];
sunrise = Sunrise[Entity["City", {"NewHaven", "Connecticut", "UnitedStates"}],DateRange[{2015, 1, 1}, {2015, 12, 31}]];
month = DateString[#, "MonthName"] & /@ DateRange[{2015, 1, 1}, {2015, 12, 31}];

(*Plot*)
listclean = 24. Mod[#, 1] & /@ QuantityMagnitude@(DateDifference[#[[1]], #[[2]]] & /@ Transpose[{DateString /@ sunrise["Values"],DateString /@ sunset["Values"]}]);

We can now get the total number of daylight hours by:

Total /@ GatherBy[Transpose[{month, listclean}], #[[1]] &][[All, All, 2]]
(*{297.233, 297.667, 370.167, 399.417, 449.15, 453.067, 459.567, 428.1, \
374.583, 343.983, 296.567, 286.883}*)

which give the following plot

ListLinePlot[Total /@ GatherBy[Transpose[{month, listclean}], #[[1]] &][[All, All,2]], PlotMarkers -> {Automatic, 10}, 
 Ticks -> {Transpose[{Range[12], Rotate[#, Pi/2] & /@ GatherBy[Tally@month, #[[1]] &][[All, 1, 1]]}], All}, 
 LabelStyle -> Directive[Bold, Medium]]

This is much faster, does not lead to time outs, so that the results are more reliable and it probably eats up fewer of your cloud credits/API requests.

Cheers,

M.

PS: Note that the Rotate command does not work in the cloud. The rest works just fine. This here should work though:

ListLinePlot[Total /@ GatherBy[Transpose[{month, listclean}], #[[1]] &][[All, All,2]], PlotMarkers -> {Automatic, 10},
Ticks -> {Transpose[{Range[12],DateString[#, "MonthNameShort"] & /@GatherBy[Tally@month, #[[1]] &][[All, 1, 1]]}], All},
LabelStyle -> Directive[Bold, Medium], ImageSize -> Large]

Hi there,

I do not know how to do this in WolframAlpha, but if you get a free Wolfram Cloud account you might try this:

list = Table[{DateValue[DatePlus[DateObject[{2015, 1, 1}], k], "MonthName"], 
   Mod[#, 1] & /@ QuantityMagnitude[(Sunset[Entity["City", {"NewHaven", "Connecticut", "UnitedStates"}], 
        DatePlus[DateObject[{2015, 1, 1}], k]] - Sunrise[Entity["City", {"NewHaven", "Connecticut", "UnitedStates"}], 
        DatePlus[DateObject[{2015, 1, 1}], k]])]}, {k, 1, 365}];
listclean = Select[list, NumberQ[#[[2]]] &];

This gives you the daily data:

ListPlot[24.*Mod[#, 1] & /@ QuantityMagnitude[listclean[[All, 2]]], AxesLabel -> {"Day of the year", "Daylight hours"}, LabelStyle -> Directive[Bold, Medium]]

You can also get the average day light hours per month

daylength = Mean[Mod[#, 1] & /@ QuantityMagnitude[#[[All, 2]]]] & /@ GatherBy[listclean, #[[1]] &]

and then plot it:

ListLinePlot[24. daylength, PlotMarkers -> {Automatic, 10}, 
 Ticks -> {Transpose[{Range[12], Rotate[#, Pi/2] & /@ GatherBy[listclean, #[[1]] &][[All, 1, 1]]}], All}, LabelStyle -> Directive[Bold, Medium]]

In principle you can now also calculate the total number of daylight hours per month:

Total /@ (24. Mod[#, 1] & /@ QuantityMagnitude[#[[All, 2]]] & /@ GatherBy[listclean, #[[1]] &])
(*{297.233, 297.667, 370.167, 399.417, 449.15, 453.067, 459.567, 428.1, \
362.1, 343.983, 296.567, 268.467}*)

The thing is that there were a couple of time outs when I downloaded the data, so that one day in September and two days in December were missing.

ListLinePlot[
 Total /@ (24. Mod[#, 1] & /@ QuantityMagnitude[#[[All, 2]]] & /@ 
    GatherBy[listclean, #[[1]] &]), PlotMarkers -> {Automatic, 10}, 
 Ticks -> {Transpose[{Range[12], Rotate[#, Pi/2] & /@ GatherBy[listclean, #[[1]] &][[All, 1, 1]]}], All}, LabelStyle -> Directive[Bold, Medium]]

Also it it easy to see that some months have fewer days than others.

Cheers,

M.