This is information from Table 1.

In other words, measured or calculated sunlight intensities.

The invisible component for sunlight that is responsible for sunburn and breaks down organic compounds. UV light has a shorter wavelength than visible light. See this link for more information.

This study measured how chemical and biological properties of the water in Toolik Lake and its surrounding streams change over time.

This paper is cited in the supplemental information because the authors used the same analysis methods in the two papers for measuring water chemistry parameters.

This study used a model to measure the amount of permafrost decrease that occurred between 1850 and 2002 and found that there was enough change to potentially impact landscape, hydrology, and ecosystems.

This video, made by people who work with primary author Rose Cory, explains what photodegradation is.

In other words, over the area where rain water flows into the river and lake system we studied.

It is currently unknown which processes are important for converting dissolved organic carbon from lakes, rivers, and streams into carbon dioxide.

Carbon release from permafrost has been a hot topic in the news, because of its potential importance for climate change. For more information, see this article from Cambridge University and this article from the Woods Hole Research Center.

Mineralization refers to the break down of organic molecules to inorganic molecules. For the purposes of this paper, it means conversion of organic carbon to carbon dioxide.

In other words, organic carbon is more available for microbes to eat and turn into carbon dioxide (bacterial respiration) after it has been partially broken down by sunlight. This process happens enough to be important when considering the total amount of bacterial respiration.

N represents the number of samples tested.

Quantum yield is a measurement of how much of a chemical reacts when exposed to light. For example, in this study it would be the amount of carbon dioxide produced divided by the amount of UV sunlight absorbed by dissolved organic carbon.

These are the sites marked on the map in figure S1 (see Figure 1 annotations).

Here, the authors validate their results by comparing them with previous studies.

In Toolik Lake, they found at least as much sunlight-driven dissolved organic carbon (DOC) degradation (relative to microbial-driven DOC degradation) as has been previously measured in other locations.

The amount of bacterial respiration measured by the authors was lower than in previous studies in other areas, and the reactivity of the dissolved organic carbon with sunlight was higher. Therefore, sunlight-driven degradation was more important in this study than in others that took place in other locations.

Previous studies have found that because some of the dissolved organic carbon (DOC) in the Arctic comes from melting permafrost, it has little prior exposure to sunlight and can thus be degraded by sunlight more easily than DOC in other locations. The data in this paper are consistent with this previous research.

Because most of the surface water in the study area is shallow, and there is little shade, sunlight-driven dissolved organic carbon (DOC) degradation is more important than in deep lakes, streams, and forests. The authors suggest that photodegradation of DOC may also be important in other areas with shallow surface water and little shade.

As briefly discussed earlier in the paper, the authors used their calculated daily areal rated of dissolved organic carbon (DOC) processing to calculate the total amount of each type of DOC processing that occurs in their entire study area.

They calculated separate rates for three different types of water bodies.

The authors were able to use their dissolved organic carbon (DOC) degradation rates for the different types of water bodies to calculate the relative importance of sunlight- and microbial-driven DOC degradation in the whole study area (row labeled "sum" in Table 2).

Final tally: Sunlight is responsible for 75% of dissolved organic carbon degradation and 55% of carbon dioxide production in the Kuparuk River basin.

The day in the spring when all of the ice in a lake has melted.

This study showed that carbon from Arctic soils can be released as carbon dioxide from Arctic surface waters.

This letter to the editor makes an argument that transport of permafrost carbon into surface waters and beyond is important to consider when modeling the carbon cycle in the Arctic.

This review examines the impacts of climate change on the carbon cycle in the Arctic.

This study reported a higher and more reliable estimate of the total amount of carbon stored in Arctic permafrost relative to previous studies. They included deeper soils and additional pools of carbon.

This study found that on a decadal time scale, melting permafrost is a large source of carbon to the atmosphere, even when increases in carbon dioxide use by new plant life is taken into consideration.

This study showed that dissolved organic carbon (DOC) that is exposed to sunlight is more susceptible to being broken down by microbial respiration than DOC that is kept in the dark (i.e., photostimulated bacterial respiration is important).

The authors cite this paper heavily in their supplemental information. Many of the same methods were used in the two papers, including those for field sampling, light absorption measurements, and measurements of photostimulated bacterial respiration.

This review describes the pathways for permafrost carbon to enter the atmosphere and the changes in Arctic ecosystems that could be caused by permafrost thawing. They conclude that the overall impact of thawing permafrost will be to increase atmospheric carbon and global climate change.

This paper uses a model to measure the feedback between global warming and permafrost melting (i.e., how much additional warming will occur because warming is causing additional carbon dioxide and methane to enter the atmosphere).

This study found that sunlight converted dissolved organic carbon, nitrogen, and phosphorous into forms that are more easily consumed by bacteria and plankton in a lake in Norway. The bacteria and plankton were able to be more productive when grown in sunlight-exposed waters.

This paper shows that both sunlight and microbial transformation of dissolved organic carbon (DOC) is important to consider when examining the structure of DOC remaining in surface water. Additionally, it is important to consider how long the DOC has been present in the surface water.

This study is very similar to the one you just read. These authors studied sunlight-driven dissolved organic carbon degradation in Swedish lakes and found that 12% of the carbon dioxide released by the lakes was from photodegradation.

This study determined the effect of a specific type of sunlight-driven degradation of dissolved organic carbon (DOC) on the availability of DOC to be eaten by bacteria.

The authors cite this paper in the supplemental information because they used the same methods for measuring oxygen consumption by bacteria in water samples.

The authors determined how important sunlight is for breaking down organic carbon by comparing how much microbes can degrade in the dark and how much all the sunlight-driven processes can degrade.

Measurement methods are summarized here and provided in detail in Reference 11.

Photostimulated bacterial respiration was measured by exposing some sterile water samples to sunlight and keeping some in the dark, then adding bacteria and measuring the difference in how much carbon dioxide was produced in the bacteria in the sunlight-exposed samples versus the bacteria in samples that were kept in the dark.

Areal rate refers to the amount of something that happens over an amount of surface area. For example the rate of bacterial respiration is measured as the amount of oxygen released per day, but the areal rate would be measured as the amount of oxygen released per square meter per day.

This refers to how we interpret our own behavior as well as the behaviors of the people we interact with.

Remediation means trying to resolve a problem once it has already occurred. In this case, the author is referring to remediating violence.

Baseline measurements are measurements that report on the sample population prior to the experiment. In this case, prior to the start of the jobs summer jobs program, 20% of the youth in the study had been arrested and 20% had been victims of violent activity.

By operative mechanism, the author is referring to the thing or factor that helps us understand why this summer jobs program helps reduce violent crime.

Empirical literature refers to articles that are based on data that has been collected in a field (real-world) or laboratory setting. Because science is a process that builds upon itself, scientists are always looking to see what the existing empirical literature establishes so that a study that presents something new can be designed.

Because we can not guarantee external validity, the author suggests that studies similar to this one take place in other cities to better identify why summer jobs programs affect youth participation in violence.

Tertiary prevention refers to prevention that occurs after young people are already out of school. This study differs by focusing on young people as they are in school.

The possibilities raised by these findings suggest that prevention is a cost efficient way to decrease youth involvement in violence. Prevention is more cost effective than remediation.

In this experimental study, outcome measures refer to violent activity as well as other types of crime and schooling outcomes. This paper reports specifically on whether summer jobs have any affect on violent activity.

What is a disturbance, in this context? In ecology, a disturbance is anything that changes the ecosystem. For example, a forest fire, tornado, human activity (building more houses in an area that used to be a grassland), etc.

How do you think Landsat-like data will influence future forest change around the world? How could Landsat-like data influence global phenomena, such as climate change?

In this paragraph, Hansen and colleagues structure information to show how public global imaging data sets such as Landsat data can be used to address environmental, economic, and social issues.

Not being allowed to do a particular activity (in this case, cut down forests) for a set amount of time

To clarify, subtropical forests changed the most in terms of gain and loss, yet when looking at all subtropical regions as a whole, there isn't a clear trend as there is for tropical forests.

What might be causing these large changes in forest cover in the subtropics?

Aggregate dynamic refers to the combination of forest loss and gain.

Why do boreal forests grow slower than other forest types? For starters, boreal forests have shorter growing seasons. Also, boreal forests have only conifer tree species that do not grow as quickly as deciduous trees.

The authors' results and analyzed data are a great jumping off point for governments that want to better understand how their country's land is changing with time.

These results can then be used to implement policies that manage or protect land where needed.

To look at the maps of forest change data from Hansen and colleagues, click here.

For more information about Sweden's 2005 Cyclone Gudrun, click here.

Why is carbon storage important? Find out by following this link to explore carbon "sequestration" (just a fancy term for "storage") and how this process can help curb global warming.

Hansen and colleagues are referring to previous work conducted by the Brazilian government, which looked at forest loss in the Amazon.

This report from the Brazilian government shows that Brazil's forest protection laws are effective (i.e., the rate of forest loss in the Amazon is steadily declining).

What are important ecosystem services that forests provide?

First, let's break this down into what an ecosystem service is. These services include any benefit that an ecosystem can provide to people.

So, what can forests provide that benefit people? Here are some examples: Forests provide timber, store carbon, purify air and water, and provide space for recreation (e.g., hiking in the woods!).

When comparing across different climate domains (i.e., tropical, subtropical, temperate, or boreal), Hansen and colleagues found that only the tropics showed a clear change in the rate of forest loss during the time that they surveyed (2000–12).

In this case, forests are decreasing rapidly in the tropics and in the next sentence we find out exactly where this decrease is taking place; mainly in Malaysia, Paraguay, Bolivia, Zambia, and Angola.

Take a look at Google Maps to find the location of these countries.

Richness is simply the number of different species.

Biodiversity refers to the diversity of biology (a.k.a. the number of different species).

Putting this all together, we can determine that forests, compared with other types of ecosystems (e.g., deserts), have a lot of different species of plants, animals, bacteria, etc. (this is especially the case in tropical forests).

Hansen and colleagues are building off of previous scientific research that addressed this same issue of trying to figure out how forests are changing around the world.

The main difference between this study and previous work is that technology has improved in such a way that we can now look at the entire globe at a very fine resolution to determine where and how our forests are changing through time.

Hansen and colleagues looked at satellite images of Earth (see "Landsat data") for every year from 2000 to 2012 and determined how much of Earth is covered in forest, and whether forests increased or decreased in size from year to year.

This method involves a lot of computer processing, because the raw satellite images need to be screened and assessed before they can be used.

Hansen and colleagues performed the computational processing in Google's Earth Engine.

Describes where things are in relation to each other. In this case, the satellite images of Earth provide a clear picture of where forests are located and distributed.

Imagine having spatially explicit data of Earth's surface, all collected at the same time. This is what satellite-based imaging systems do. By examining these pictures over time, we are able to map forest loss and gain. Imagine trying to do this task from the ground. It would be very hard indeed.

This sentence is referring to how Hansen and colleagues used Landsat data (e.g., the satellite images of Earth), which is all collected in exactly the same way each time. Thus, there are no inconsistencies between one sampling event (e.g., photo) and another sampling event.

Tropical: Areas near Earth's equator that are warm/hot year-round with consistent or seasonal rainfall.

Subtropical: Areas with hot and humid summers and mild winters.

Temperate: Areas with four seasons (summer, fall/autumn, winter, and spring) divided mainly by differences in temperature.

Boreal: Subarctic areas with long, cold winters and short, cool summers.

Check out where these climate domains/zones are around the world on this map.

"Deforestation dynamics" refers to changes (i.e., dynamics) in forests due to cutting down trees and replacing them with nonforest land uses, such as agriculture or development (houses, buildings, etc.).

This wording implies that Hansen and colleagues ran a statistical model to determine whether the loss or gain in forest cover over time was more or less than what you would expect if forest cover did not change.

The tropics experienced a clear increasing rate of forest loss, expressed in units of forest area loss per year, whereas other climate domains (e.g., temperate, boreal, subtropical) all lost and gained forest cover

However, when you add all of the subtropical regions of Earth together, for instance, there isn't a clear net loss or gain in forest cover. This is because of the fact that most forest change in the subtropics is due to forestry land uses where trees are grown as a crop. In forestry land uses, trees are continuously grown and cut down to make products such as lumber and paper.

Hansen and colleagues use linear models (y=mx+b) to test whether the trends in forest loss in Indonesia and Brazil are both significant through time. What did they find?

Many subtropical forests are treated like crops (e.g., corn, soybean). As such, the trees are cut down and harvested for timber and then replanted for the next harvest season.

Thus, this climate domain experienced the most forest loss while also having the lowest ratio of loss to gain (because the trees that were harvested were subsequently replanted).

To see an example of this, check out an article on forest landowners in Mississippi here.

This statement refers to "short rotation forestry," which is a type of forestry that densely plants fast-growing tree species (e.g., poplar trees).

Once these trees reach a certain size (e.g., stems that are 10–20 cm in diameter at breast height), they are then cut down and harvested for lumber, pulp, and paper products, or energy.

The trees then regrow from the stumps, sending up new trunks. This process of cutting down a tree to stimulate regrowth is called "coppice."

Out of all countries, Russia lost the most forest cover from 2000 to 2012. Why do you think Russia has lost more forest cover than any other country?

Look at this forest loss on this map (click on Forest Cover Loss 2000-2014).

By looking at the forest loss with or without subsequent forest gain, Hansen and colleagues can determine whether a country may have forestry programs or some other type of deforestation.

To explain, forestry programs treat forests as a crop in that the trees are harvested and subsequently replanted/regrown until the next harvest, and so on. Thus, countries that experience forest loss and subsequent gain likely have forestry programs.

Countries that have forest loss without much forest gain probably do not have forestry programs, and thus the deforestation is caused by some other factor. What could be some other causes of deforestation?

Check out this Landsat time-lapse video from 1984-2012 of Brazil's Amazon Basin.

In this context, the authors are describing the satellite images—these images are widespread (e.g., systemic) in that they photograph the entire surface of Earth.

The authors used Google Earth Engine to process the Landsat images.

The Google Earth Engine is a cloud platform, meaning that a network of thousands of computers works together to perform a task that a single computer would take years to do.

Similarly, Google Cloud provides the same features. Google Cloud allows you to store, manage, and process information on computer servers that are accessed through the Google Cloud website.

Cloud computing is especially helpful for processing large amounts of data/information.

Hansen and colleagues processed 700,000 images of Earth. Processing this information through the Google Earth Engine with 10,000 computers took approximately 15 days. If the authors only had one computer to work with, these calculations would have taken a few years!

Hansen and colleagues are illustrating how the work they conducted in this study can be applied and used by governments around the globe to monitor changes in land cover over time.

Also, to clarify, "time-series data sets" refers to how their data (forest cover measurements) were collected, i.e., in this case in a series of collections through time (one collection of satellite images of Earth for each year from 2000 to 2012).

Science is collaborative and support from scientific research can come from many different sources. Check out which organizations helped to fund this research.

Gordon and Betty Moore Foundation

U.S. Geological Survey

Google, Inc.

David and Lucile Packard Foundation

NASA

US Agency for International Development

This means that previous research studies have found that an individual's happiness/life satisfaction can be statistically predicted, based on how conservative or liberal that person is; the more conservative people are, the happier they say they are.

In a meta-analysis, Jost, Glaser, Kruglanski, and Sulloway (2003) found that political conservatism was associated with a number of individual differences (such as openness to experience and fear of threat and loss) that relate to the motivation to manage uncertainty and threat.

Heine, Kitayama, and Hamamura (2007) show support for the idea that people from Western/individualist cultures self-enhance significantly more than people from Eastern/collectivist cultures.

These results indicate that the relationship between political conservatism and life satisfaction can be statistically explained by self-deceptive enhancement.

With this kind of mediation analysis, the effect is said to be significant when the 95% confidence interval (in this case, from 0.03 to 0.07) does not include zero within its range. See this website for some more information.

The results from this analysis indicate that, as political conservatism (according to voting record) increases, use of positive affect words decreases.

There is no statistical relationship between political conservatism and use of negative affect words, joviality words, or sadness words.

Critical thinking: Why do you think there was a relationship for positive affect, but not joviality?

The researchers examined whether or not an individual's usage of positive/negative words and emoticons in tweets could be statistically predicted based on that individual's assumed political party affiliation (either Democrat or Republican).

This study is essentially the same as the one that examined photographs of members of Congress (Study 2), but this one uses photographs from a business setting, rather than a political one.

You can check out these "legislative scorecards" yourself, here.

The authors point out that previous research would suggest that self-reported happiness should match behavioral indications of happiness; the fact that their studies show the opposite finding, they argue, can be explained through self-enhancement.

Connects to Vision and Change Core Competency #1: ability to apply the process of science.

The authors look beyond their own conclusions and consider additional questions that should be investigated in the future.

Connects to Vision and Change Core Competency #6: ability to understand the relationship between science and society.

The authors relate their findings to the link between psychological research and public policy.

The experiments in this article also use nonrandomized groups (i.e., participants cannot be randomly assigned to be a member of the Democratic or Republican party).

How might you run a similar study using randomized groups? What would this add to our understanding of happiness?

This connects to Science Practices for AP Biology Practice 3: The student can engage in scientific questioning to extend thinking or to guide investigations within the context of the AP course.

In this paper, the authors argue that positive self-biases (overly positive self-evaluations, illusion of personal control, and unrealistic optimism) are characteristic of normal human thought.

In addition, the authors say that these positive illusions are adaptive and beneficial for mental health, in that they promote more satisfying connections with others, greater productivity, and greater resilience in response to negative feedback.

This research, by Napier and Jost (2008), found that the happiness gap between conservatives and liberals could be explained by the fact that political conservatives are more likely than liberals to rationalize inequality.

In one study, the researchers looked at U.S. economic inequality over time (from 1974 to 2004), and found that, in general, more inequality was associated with lower well-being, but that conservatives seemed to be somewhat buffered from this effect. As inequality goes up, well-being goes down a lot for liberals, but only a little for conservatives.

Other research (for example, Jost and Hunyady, 2002) also shows that conservatives are generally more likely to view the status quo (or, the way things are now) as being fair.

So, the implication is that conservatives seem to be less bothered by inequality.

The argument here is that if research relies too much on just one way of doing things (in this case, using self-report), then this single methodology might be overlooking an important issue.

For example, Zou and Schimmack (2013) argue that self-report measures should look at multiple raters; that is, in addition to asking participants about themselves, researchers should also validate those ratings by asking the same questions to people (family, friends) who know the participant well.

Connects to NGSS Practice 1 (Asking questions [for science] and defining problems [for engineering]) and Practice 3 (Planning and carrying out investigations).

How would you design a study to evaluate this idea? How do the researchers go about testing their hypotheses?

This is the main hypothesis for Study 1. The authors predict that the reason why conservatives report higher levels of well-being is that, compared with liberals, conservatives are more likely to self-enhance.

For example, research by Harker and Keltner (2001) showed that, compared with non-Duchenne smiles, displays of Duchenne smiling in yearbook photographs were associated with higher levels of personal well-being at ages 21, 27, 43, and 52.

Democrats tended to use one negative word for every 13.65 positive words. This negative word usage is less frequent than Republicans, who used one negative word for every 11.5 positive words.

That is, the results indicate that although conservatives were not significantly more likely than liberals to not use the orbicularis oculi at all, conservatives were likely to display less muscle activation around the eyes, compared with liberals.

Here, Nisbett and Wilson review evidence that supports the idea that people generally have relatively little insight into their own psychological/cognitive processes. When it comes to the process of a stimulus causing/influencing a response, people can be unaware of the stimulus itself, unaware of the response, or unaware of how the stimulus influenced the response.

The authors argue that although people can attempt to report on the mechanisms that underlie their thoughts, decisions, etc., these self-reported explanations are not necessarily accurate, and they rely more on people's lay theories about how cognition should work, as opposed to how it actually does work.

Here, the authors report on the "happier-than-average effect," which shows that greater self-enhancement (either because of individual differences or experimental manipulation) leads to greater reports of subjective well-being.

Hibbing, Smith, and Alford argue that individuals' reactions, both physiological and psychological, to negative events and stimuli underlie a major part of the differences between political liberals and conservatives. Their argument centers around the idea that, compared with liberals, conservatives display greater responses to negativity.

This theoretical paper examines the potential for differences between high self-esteem and "optimal" self-esteem. Kernis makes the argument that high self-esteem can be either fragile (nonoptimal) or secure (optimal), depending on factors such as the defensiveness and stability of that self-esteem.

In his book, Gross National Happiness, Brooks examines the "happiness gap" in the United States and argues that values such as faith, hard work, and individual liberty cause happiness, whereas factors like secularism and an overreliance on government promote unhappiness.

The book focuses on culture, politics, religion, and economics. Brooks's conclusion includes suggestions on how the U.S. government can help facilitate happiness.