Alaska sunlight peak hours

[Weldman]

I hate starting new threads, but this one deserves a discussion not hardly found anywhere. In Alaska if one hypothetically went there with PV system in the summer months how many peak hours of sunlight would one get. Is there areas to watch out for such as sections of Alaska to avoid. Have heard it's as bad as Washington state on the West side of Washington state due to clouds and such and have heard the sun never goes down in summer time too.
My thinking is even the sun isn't out full shining that the hours it does stay out when cloudy, it makes up the difference lost of full sunlight, or is this a endeavor one would have to embark on to find out and report on?

[Ampster]

The answer can be found on PV Watts. A lot of the answer depends on where you are located, what direction the panels face and what degree of tilt you give them.

[J.P.M.]

If by "peak hours" you mean how much solar energy a PV system might be exposed to over a certain period of time, averaged over many years, Ampster's got this one mostly right.

But to maybe back up a little, pretty much regardless of location on the planet, the sun is above the horizon for pretty close to half of the 8,760 hours in a year. All 8,760/2 hours happen in 6 months at the poles, and the sun is above the horizon for 12 hours or very close to it every day for locations on the equator. The amount of daylight at any other specific location and date is then f(latitude).

The amount of insolation (the time integrated solar irradiance) a surface will see over any time period at a location is a function of the surface orientation (including and accounting for any surface motions such as "tracking") and local weather conditions including how cloudy it is and how clean the atmosphere is (with the last two as quantified by something called the clearness index).

In lower latitudes, a PV system's out on a cloudy day will rarely, if ever, get even close to the output of that same system on even a partly cloudy day. Usually, system output on "completely": cloudy days (with a clearness index of, say, something like 0.10, will be maybe 10-15 % of the output on completely clear days (that will have a clearness index of, say, ~ 0.70 or a bit higher).

At higher latitudes, it's common for albedo (reflected irradiance from the ground and surrounding buildings, etc.) to increase cloudy day output ratios, especially for arrays at higher tilts so that clear day system production totals can be higher than otherwise expected, but still nowhere near that of even a partly cloudy day.

Models such as PVWatts and others mostly use other models and some (but surprising small amounts of) historical measured data to estimate irradiance at locations to use for modeling possible solar device performance at a location.

See the PVWatts help screens for further information about solar resource data and why models are design tools, not performance predictors and the difference between modeling and prediction.

Take what you want of the above. Scrap the rest.

[Weldman]

If by "peak hours" you mean how much solar energy a PV system might be exposed to over a certain period of time, averaged over many years, Ampster's got this one mostly right.

But to maybe back up a little, pretty much regardless of location on the planet, the sun is above the horizon for pretty close to half of the 8,760 hours in a year. All 8,760/2 hours happen in 6 months at the poles, and the sun is above the horizon for 12 hours or very close to it every day for locations on the equator. The amount of daylight at any other specific location and date is then f(latitude).

The amount of insolation (the time integrated solar irradiance) a surface will see over any time period at a location is a function of the surface orientation (including and accounting for any surface motions such as "tracking") and local weather conditions including how cloudy it is and how clean the atmosphere is (with the last two as quantified by something called the clearness index).

In lower latitudes, a PV system's out on a cloudy day will rarely, if ever, get even close to the output of that same system on even a partly cloudy day. Usually, system output on "completely": cloudy days (with a clearness index of, say, something like 0.10, will be maybe 10-15 % of the output on completely clear days (that will have a clearness index of, say, ~ 0.70 or a bit higher).

At higher latitudes, it's common for albedo (reflected irradiance from the ground and surrounding buildings, etc.) to increase cloudy day output ratios, especially for arrays at higher tilts so that clear day system production totals can be higher than otherwise expected, but still nowhere near that of even a partly cloudy day.

Models such as PVWatts and others mostly use other models and some (but surprising small amounts of) historical measured data to estimate irradiance at locations to use for modeling possible solar device performance at a location.

See the PVWatts help screens for further information about solar resource data and why models are design tools, not performance predictors and the difference between modeling and prediction.

Take what you want of the above. Scrap the rest.

Well I ran the numbers and it's hard to believe that Seattle Washington gets more kWh a year than Anchorage or Fairbanks Alaska. I wonder if this website is accounting for the dark winters and that's why it's factoring in less for Alaska. Trying to do calculations based on yearly averages doesn't help when it comes to a place that has 6 plus months of darkness a year. Looks like this is going to require information directly from locals.

[littleharbor]

I've always wondered about those long days of summer in Alaska. Does the sun not travel around in a circle, above the horizon? Although extreme weather would be detrimental to a tracker, wouldn't a dual axis tracker really increase harvest in summer?

[J.P.M.]

I've always wondered about those long days of summer in Alaska. Does the sun not travel around in a circle, above the horizon? Although extreme weather would be detrimental to a tracker, wouldn't a dual axis tracker really increase harvest in summer?

If the sun doesn't set on a particular date at a particular location, the solar azimuth angle goes through 360 degrees at that location.

In general, under the impossible condition of always clear skies (or think of an array located just above the earth's atmosphere), the farther away from the equator a location is, the higher will be the ratio of annual total tracking array harvest to annual total stationary array harvest. Local weather and climate will of course modify the annual totals.

[J.P.M.]

Well I ran the numbers and it's hard to believe that Seattle Washington gets more kWh a year than Anchorage or Fairbanks Alaska. I wonder if this website is accounting for the dark winters and that's why it's factoring in less for Alaska. Trying to do calculations based on yearly averages doesn't help when it comes to a place that has 6 plus months of darkness a year. Looks like this is going to require information directly from locals.

Well, to the extent the model produces a reasonable representation of the probable long term average output of a modeled PV array at a location, believe it.

Some days/months years, Seattle will get more or less annual insolation than other places. Some times not.

PVwatts is a model. Run the numbers for best array orientations for each location.

Also, are you referring to kWh as in annual modeled P.O.A. irradiance or annual modeled energy harvested per array STC kWh ? Different locations will have different optimal orientations. What orientation(s) are you using ?

[Ampster]

The sun travels through more of the atmosphere the farther you are from the equator and that may have some effect as well.

[Weldman]

Well, to the extent the model produces a reasonable representation of the probable long term average output of a modeled PV array at a location, believe it.

Some days/months years, Seattle will get more or less annual insolation than other places. Some times not.

PVwatts is a model. Run the numbers for best array orientations for each location.

Also, are you referring to kWh as in annual modeled P.O.A. irradiance or annual modeled energy harvested per array STC kWh ? Different locations will have different optimal orientations. What orientation(s) are you using ?

I had it oriented 180 degrees for both places with tilt at 30 degrees for Seattle and I went with 0 to 45 degrees trying to beat Seattle’s output and just fell short. I was hoping I could find some locals on what real life numbers are. I was on the annual models, but seems I need to pour a drink and divulge more into smaller details to fine tune it.

[bob-n]

This may or may not matter to you, depending on your situation, but if you generate more energy than you use, you need to store or sell it. Storage costs money and is rarely practical for very long terms, like Alaska winters. With a site in Alaska, you're going to be producing too much in summer and nothing in winter, so relying on netmetering, You get less for selling energy than you pay for buying energy. You may want to check with the Alaska power company(s) to find out what they pay for excess energy you generate vs what they charge for energy you use. The ratio varies wildly from place to place.

It would be really cool if someone tabulated the buy/sell terms across the country, for easy access. For sure, it would be hard to keep up-to-date.

Does anyone know if netmetering terms vs location exists online?

[Weldman]

This may or may not matter to you, depending on your situation, but if you generate more energy than you use, you need to store or sell it. Storage costs money and is rarely practical for very long terms, like Alaska winters. With a site in Alaska, you're going to be producing too much in summer and nothing in winter, so relying on netmetering, You get less for selling energy than you pay for buying energy. You may want to check with the Alaska power company(s) to find out what they pay for excess energy you generate vs what they charge for energy you use. The ratio varies wildly from place to place.

It would be really cool if someone tabulated the buy/sell terms across the country, for easy access. For sure, it would be hard to keep up-to-date.

Does anyone know if netmetering terms vs location exists online?

I know of a friend of a friend that might hypothetically go there with a RV loaded down with 1.2 kw worth of panels to batteries that hold 440 ah @ 24v. This shall be fun to see where it leads to. So there won't be any power company to call up or any back feeding into any grids, just straight up off grid. If no one posts on boots on the ground results I guess I will get this friend of a friend to post results if they hypothetically go...

[SunEagle]

I know of a friend of a friend that might hypothetically go there with a RV loaded down with 1.2 kw worth of panels to batteries that hold 440 ah @ 24v. This shall be fun to see where it leads to. So there won't be any power company to call up or any back feeding into any grids, just straight up off grid. If no one posts on boots on the ground results I guess I will get this friend of a friend to post results if they hypothetically go...

Interesting. I just came back from an RV event and sat through a seminar given by 4 people that boon dock a lot and have added Lithium chemistry batteries to their rigs. While neither indicated an ROI they all said that the ability to live off grid anywhere they could get their RV was worth the cost of the solar/battery system.

One pair have a youtube site called RVGeeks. They are an interesting pair and were very honest with the audience about what not to do concerning a Lithium battery system for their RV.

[J.P.M.]

The sun travels through more of the atmosphere the farther you are from the equator and that may have some effect as well.

That is a good point and technically correct, but it may not have as much effect in GHI or POA irradiance as one might think for many applications, but particularly for this situation of a comparison between 2 more northerly and relatively similar latitudes.

There are bigger fish to fry first, like location clearness indices.

The effect you write of is known as the air mass or air mass correction and used to first find the approx. path length of beam radiation through the atmosphere relative to the depth of the atmosphere at a location. A zenith angle of 0 deg. meaning the sun is directly overhead has an air mass of 1.0. The air mass is then used to estimate how much of the beam irradiance is attenuated as f(path length), and then how much of the remaining beam is turned into diffuse, partly as f(path length). To a 1st approx., air mass at any time is roughly the value of the Secant (=1/Cos) of the solar zenith angle at a location and a specific time and date. So, when the solar zenith angle is 30 deg. ( meaning the solar elevation angle = 60 deg.), the air mass is ~ 1/Cos 30 = 1/.866 = 1.15.

It can be calc'd for any location at any time and summed (integrated over any time period), but is usually part of an estimate of instantaneous direct normal irradiance with those values then time integrated as needed for the analysis.

For the comparison of Seattle and, say, Fairbanks, on an annual basis, and also somewhat dependent on water vapor, particulates, mixed gases and dust in the atmosphere, a ballpark number for the annual effect of air mass and its effects on the GHI difference between Seattle and, say, Fairbanks is probably of the order of a % or so, maybe less. Not discounting it, but it might have a small enough effect, particularly without good data and some assurance of the atmos. constituents, particularly as f(altitude), that it may be better to leave it out or use 1 % unless the accuracy of the rest of the data warrants it. Otherwise, my guess is the effect of path length for this application will get lost in the noise of the rest of the irradiance calcs.

Also, and FWIW, the effect of path length on attenuation of irradiance at locations farther from the equator is usually somewhat less than one might expect or often less than calculated, mostly because of the atmosphere's generally lower temp. at higher latitudes that reduces the atmosphere's capacity to carry water vapor, with water vapor being one of the main contributors to attenuation (and scattering) of irradiance by the atmosphere.

For those interested, SAM (PVWatts on steroids) will give a fair approx. of the atmospheric path length at a location by the hour over a year's time.

Take what you want of the above. Scrap the rest.

[Ampster]

A simple rule of thumb that works for me when I am fishing on the ocean is to use more sunscreen the closer I am to the equator.

[J.P.M.]

A simple rule of thumb that works for me when I am fishing on the ocean is to use more sunscreen the closer I am to the equator.

A seminal contribution to the body of knowledge related to solar resource assessment.

I could only add a general comment to include sunscreen use at high altitude locations and/or places with high albedo pretty much regardless of latitude.

As I often write, take what you want of my stuff. Scrap the rest.