Part 1: Albedo

    (1) Annual-average albedo plot.

1. Suspicious albedos. Do you see any values greater than 1.0 or undefined values? Should there be any such values, and if there are, what might account for them?
   
   Not on this plot. (However, if we'd calculated albedo by dividing reflected solar by incoming solar radiation for each month separately, and then averaged the months to get an annual average, we would have seen some undefined values at sufficiently high latitudes. That's because there is no incoming solar radiation during some months of the year at those latitudes, and because it's not possible to divide by zero, the albedo at those latitudes and months would be undefined. Trying to include undefined values in the annual average would also make the annual average for the affected latitudes undefined. Calculating the annual average albedo using the strategy above avoids that problem.)
   
   
2. Global, annual-average albedo. Calculate the area-weighted global mean value of annual-average albedo and compare it to the value given in the global heat budget diagrams provided in class with Lab Activity #5. Are they very different? (If so, what might account for the difference?)

The area-weighted, annual, global average is 0.317 (31.7%). This is a little higher than what we calculated from the data in the annual, global average energy budget figure in Lab Activity #5. The values might differ because we calculated annual average albedo for 1987 only, whereas we don't know what years were used to determine the values in the figure in Lab Activity #5. It's also possible that the averaging calculations were done a little differently in the two cases.

3. Spatial patterns of albedo. On the (unweighted) annual-average albedo plot, where does albedo tend to be relatively low and relatively high, compared to immediately surrounding areas? (Focus on larger scale areas and deemphasize small, individual spots.) Pose some hypotheses about what might account for some of these variations.
   
   
   • Highest: Greenland and Antarctica.
     
     
   • Next-highest (but distinctly lower albedo):
   • narrow strip around much of the globe just north of the equator
   • the Arctic Ocean, North Africa, southeastern China, and areas just offshore of Antarctica
   • much of the midlatitudes and lower parts of the high latitudes
   • low latitudes in the eastern South and North Pacific Ocean, eastern South Atlantic Ocean, and (a hint, anyway) eastern Indian Ocean
   • a kind of "wing" shape extending from the western equatorial Pacific Ocean, and similarly in the South Atlantic Ocean
     
     
   • Lowest: tropical and subtropical oceans outside of the areas listed above
     
     
   Some possible explanations:
   • Snow and ice on Greenland and Antarctic (snow and ice have high albedo, and those areas have permanent ice caps)
   • Ocean surface has low albedo, so all ocean areas are dark except where there are clouds (which have high albedo) for at least part of the year. The more persistent the cloudiness is, the higher the albedo will be in the annual average.
   • Deserts in North Africa consist of relatively high albedo sand and rock
   • Southeastern China has significant cloudiness for at least part of the year
   • Midlatitude land areas have significant cloudiness for at least part of the year
   • Upper midlatitude and lower high latitude land areas have snow cover for at least part of the year
     
     

1. Comparison with spatial patterns of precipitation. Open a child window containing the annual-average albedo, then hide the radiation budget terms layer. Now, from the "Climatology" data library, drag the monthly-average precipitation data set onto the layer list. Calculate and plot a new field: annual-average precipitation for 1987. Does this plot help you test any of your hypotheses in (1)(c)? If so, how?
   
   
   This plot supports the idea that the narrow strip of high albedo just north of the equator (and more broadly near the equator in the western Pacific Ocean and Indian Ocean, plus part of the "wing" extending southeastward from the western equatorial Pacific) are due to clouds. (Clouds are necessary to produce precipitation.) However, not all clouds produce precipitation, so the relative lack of precipitation over some ocean areas that have moderately high albedo doesn't rule out the attribution of that albedo to clouds.

    (2) Annual-average clear-sky albedo plot. If your child window is still open, close it. Now, repeat the plot in (1), but this time calculate the annual-average albedo by first calculating the annual-average reflected solar radiation from the earth under "clear sky" conditions (that is, with clouds removed).

1. Spatial patterns of clear-sky albedo. Where does the clear-sky albedo tend to be relatively low and relatively high? Does this plot help you test any of your hypotheses about the reasons for spatial variations in albedo in (1)(c) above, and if so, how?
   
   
   • Highest:
     • Very high latitudes, including Greenland (just glimmers of this amongst the missing values)
   • Next-highest:
     • North Africa and Saudi Arabia
   • Low but not the lowest:
     • Land areas generally
   • Lowest:
     • All ocean areas
       
       
   This confirms that the higher albedo over North Africa (and presumably Saudi Arabia, too) is due to the nature of the ground there, not due to clouds. It also confirms that any part of the oceans with relatively higher albedo must be due to clouds. It further offers a few hints of confirmation that the high albedo over Greenland and Antarctica and the Arctic Ocean are potentially due to snow and ice, though high albedoes in those areas could be due to either or both of snow/ice and clouds, so removing clouds (if any in those areas) doesn't necessarily prove that the high albedo is due to snow and ice. Finally, the moderately high albedo over southeastern China must be due to clouds.
   
   
2. Comparison with spatial patterns of vegetation. Open a new child window with the clear-sky albedo plot in it, then hide the clear-sky albedo layer. Next, from the "Climatology" data library, drag the "Terrestrial Biomes" data set onto the layer list, and plot the "Dominant Vegetation" field. Comparing this plot with the clear-sky albedo plot in the child window, would you say that this plot helps you test any of your hypotheses in (1)(c)? If so, how?
   
   
   • Not shown here, but North Africa and Saudi Arabia are shown to consist of desert, where the unvegetated, sand and rock ground surface has relatively high albedo.
     
     

    (3) Animations of monthly-average albedo plots. In a Web browser, access a 12-month animation (movie) of individual, monthly-average albedo plots from the G/M/O 405 class backup Web site at: http://funnel.sfsu.edu/courses/gmo405/S18/MyWorldPlots/Albedo/Albedo_1987.gif.

Open a separate window in the browser (pull down the "File" menu and select "New..." or "New Window") and access a second movie, a 12-month animation of individual, monthly-average clear-sky albedo plots at: http://funnel.sfsu.edu/courses/gmo405/S18/MyWorldPlots/Albedo/AlbedoClear_1987.gif.

1. What temporal patterns (that is, patterns of variation over time) do you see in each animation? Pose hypotheses about what might cause them. Do the two animations together help you test any of your hypotheses? If so, how?

The border between high and low albedo over land at some high and middle latitudes shifts north in spring and summer and back south fall and winter in the Northern Hemisphere. (There's not enough midlatitude land area in the Southern Hemisphere to show this.) This suggests a seasonal variation in snow/ice cover, and possibly leaf cover of deciduous trees in some areas, at middle and lower high latitudes. There's a similar seasonal northward and southward shift in the albedo border at midlatitudes over the oceans in both hemispheres, but it's not as distinct, which suggests cloud patterns that are semi-persistent (rather than highly persistent throughout a given month, as snow and ice would tend to be). The clear-sky animation clarifies the distinct seasonal variation in albedo over middle and high latitude land areas (but it disappears over oceans). Over oceans, the strip of high albedo just north of the equator is very persistent from month to month, varying by only a few degrees of latitude. The moderately higher albedo over North Africa doesn't move.

2. Do any of these patterns help you test any of your hypotheses in (1)(a)? If so, how?
   
   Confirms that higher albedo areas over oceans are due to clouds. Distinct seasonal shift in albedo over middle and high latitudes suggests that snow and ice cover over land in those areas raises the annual average albedo in those areas, with the higher latitudes have the higher annual average because the snow and ice persists there longer than at middle latitudes. (Can't rule out the possibility that clouds contribute to these differences too, though, because clouds and snow & ice have similar albedoes, so if both are present, removing one won't change the albedo much.)
   
   

Part 2: Greenhouse Effect

How do these three values compare with the ones in the heat budget figures provided in Lab Activity #5: "Long-Term Average Heat Budgets for the Earth's Atmosphere and Surface"? [Note: to make this comparison, you'll have take into account the fact that the heat budget numbers that appear in the figures provided in class are not fluxes but rather percentages of the incoming solar radiation flux. Hence, you'll have to apply those percentages to the long-term, global average insolation to get the heat budget figures as fluxes instead of percentages.] If the two sets of figures seem significantly different, can you think of any reasons to explain those differences?

[Skipping the details here, but there are some differences, especially in the surface radiative emission flux.]

    (5) Annual-average greenhouse effect and greenhouse increase.

1. Comparison of global-average greenhouse effect calculations. Is the area-weighted, global-average greenhouse increase consistent with the global-average surface temperature and the effective radiative temperature for the earth (as viewed from space) that we got in Lab Activity #4: "Introduction to the Earth's Heat Budget"?
   

   The area-weighted, annual and global-average greenhouse increase calculated from My World data is 35°C. In Lab Activity #4 we calculated a greenhouse increase of 33.8°C or so. These aren't too far apart.

2. Spatial patterns in the greenhouse effect. What spatial/geographic patterns do you see in the annual-average greenhouse effect? Pose hypotheses to try to account for some of them.
   
   
   • High values:
     • Narrow strip of high greenhouse effect just north of the equator (broader in western Pacific and Indian Ocean, spanning the equator).
     • Most of Africa
     • "Wing" extending from western equatorial Pacific and Atlantic to the southeast. Less distinct "wing" extending from the same areas to the northeast.
   • Low values:
     • high latitudes (especially over Greenland and Antarctica)
     • high altitudes (elevated areas) including the Tibetan Plateau, the Andes Mts., the Rocky Mts., and much of Greenland and Antarctica)
       
       
   Areas with persistent cloudiness might have high values (since clouds contribute to the greenhouse effect).
   Areas at high elevations have less atmosphere above them than lower elevation areas (and hence fewer total greenhouse gases between the earth's surface and space), so they might have a smaller greenhouse increase.
   No ideas for subtropical Africa. It's got little cloud cover, and it's not obvious that it might have a lot of water vapor in the air (which would enhance the greenhouse increase) because those are desert areas. (Deserts aren't necessarily low in water vapor--just low in precipitation and cloudiness--but most deserts are short on all three.)
   
   
3. Possible correlations between the greenhouse effect and other quantities. Plot your choice of any one or more of the following:
   • Topography (pay attention to land only). (In the "World" data library, drag the "Elevation and Bathymetry" data set onto the layer list and plot the elevation and bathymetry field. Divide this field by the mean elevation above sea level of land, 677 meters, to "normalize" the land elevations.)
   • Annual-average precipitable water. (In the "Climatology" data library, drag the "MonAvg_Precipitable_H2O_Vapor" data set onto the layer list. Calculate a new field: the annual-average precipitable water. Find out what this is by consulting the documentation. Divide the new field by the mean, unweighted global average value to get a new, "normalized" field of annual-average precipitable water.)
   • Annual-average specific humidity. (Same procedure as for precipitable water.)
   • Annual-average surface temperature. (In the Energy Budget data library, .... )
   
   Do any of these four quantities seem to be relatively well correlated spatially with the greenhouse effect? (The term "correlated" can be defined in a mathematically rigorous way, but here all we want is a subjective sense of whether or not the plot patterns resemble each other closely.) Of these four, can you think of some that might be well correlated with each other?
   
   What physical connections do you think there might be between each pair of well-correlated plots that would make the correlations more than accidental?
   
   
   • Topography correlates noticeable with some areas of low greenhouse effect and greenhouse increase (but not all). (Higher elevations should have a weaker greenhouse effect because there is less total amount of greenhouse gases between the surface and space.)
   • Precipitable water correlates well with high greenhouse effect and greenhouse increase. (More water vapor in a column of air should make the greenhouse effect stronger.)
   • Specific humidity correlates somewhat, but not as obviously as precipitable water, to high greenhouse effect and increase. (Not surprising, since precipitable water accounts for all water in a column in the atmosphere, whereas specific humidity tells us only about the amount of water vapor in the at the earth's surface. The former is probably a better measure of how strongly water vapor will contribute to the greenhouse effect.)
   • Temperature correlates with greenhouse effect, but we have to be careful about attributing differences in temperature to differences in greenhouse effect. Yes, we might expect stronger greenhouse effect to produce higher temperatures, but comparison applies at a given place, not necessarily between different places, such as low latitudes vs. high latitudes. Temperature differences between low and high latitudes can be accounted for to a significant degree by differences in solar absorption, not differences in the strength of the greenhouse effect, though the latter can certainly contribute (perhaps reinforcing the difference due to solar absorption in many areas).