As many volunteers may be aware, Snapshot Wisconsin operates continuously and year-round. This is distinct from many other studies and monitoring efforts focused upon wildlife that typically evaluate population status during a particular time of year and look at changes between years. One belief that underpins Snapshot Wisconsin is that because the environment consistently changes across the year, how animals move, behave, live, die also changes. Consistent and continuous data collection provides the project much richer insights into animal habitat associations, and it also gives the project the ability to evaluate which time of year deserves most focus–when should we be monitoring things?

As discussed in previous posts, we heavily rely on spatial data produced by processing satellite imagery in order to quantify species’ habitats and estimate or predict where animals are. Recall that two important sensors or satellites are Landsat, which has fine spatial resolution but captures imagery of Wisconsin less frequently, and Modis (on the Aqua and Terra Satellites) that has coarser spatial resolution, but captures an image of Wisconsin daily. Data produced from Modis imagery is incredibly useful for capturing, say, the timing of larger scale phenomena like big snow events, or the onset of either long-term snow coverage in the winter or green up in the spring. The behaviors and activity of animal species are often connected to the timing of these environmental cues (or others, like temperature, or the length of daylight).

One animal species that is particularly sensitive to seasonality is the black bear: bears spend the winter in dens in a state of torpor. In brown bears, previous research has suggested that the timing of bear den entry is sensitive to environmental factors, while the timing of ending torpor is more related to individual physiology. One thing we are interested in is whether bear behavior (out and about, or in torpor) exhibits any correlation with the variation in the timing of plant green-up and senescence across space, and whether “mismatch” between when bears exit dens and when plants green up (plants like sedges are an important food source for bears early in the year) seems to have any population consequences.

M2E25L59-59R383B331

M2E38L99-99R390B371

M2E58L169-169R399B388

M2E84L231-230R399B435

Snapshot Wisconsin cameras capture bears growing and moving across seasons. 

While Wisconsin vegetation greens up from late winter to mid-summer, bears also become increasingly present across the landscape.

BUT…..while satellite imagery provides an excellent overhead synopsis of plant activity, it is not always clear which plants are making the image green. Bears–and many other animal species–primarily eat green matter at ground level rather than leaves at the tops of trees. Snapshot Wisconsin’s cameras provide a ground view that we can relate to satellite images to get a sense of what airborne imagery is responding to.  In the long term, this will allow improved estimates of where animals are and at what time.

As many volunteers know, one of the primary purposes of Snapshot Wisconsin is to get a better handle on where species are located throughout the state. We now have enough pictures classified (and a reasonable handle on the effects of error within the classification process) to be able to start the process.

The essence of this exercise is to model associations between 1) locations where animals are and are not found on trail cameras, and 2) environmental or spatial characteristics at those camera locations. As noted in a previous blog post, in order to effectively map predictions about animal distribution, we need to have spatially explicit environmental variables, many of which are derived from satellite imagery. Examples of environmental variables include land cover, seasonal patterns in plant productivity/greenness, snow cover, and the intensity of night-lighting, which is a good index for human activity.

Below are a collection of maps resulting from our first attempt at mapping statewide species distributions based on Snapshot Wisconsin data.

A couple things to note:

1. How we think about commonness or rarity depends on the species being considered. For example, “less common” for turkey may mean an area is visited by only a few turkeys over a couple weeks, while “less common” for bear may mean that the area has not had a bear visit for a year or more.
2. There are some imperfections. In particular, the tip of the Bayfield peninsula tends to exhibit some patterns that are probably wrong. Mapping will continue to be an iterative process based on our best metrics.

The below maps are accurate enough to be useful, but there’s always room for improvement. We hope that volunteers find this first round of analysis interesting, and maybe even useful for classifying.

Astute contributors to Snapshot Wisconsin may have noted that one of the primary partners on the project is NASA. Yes, that NASA, known for space-flights and Neil Armstrong. For many people, the involvement of an air and space agency with a wildlife monitoring project may not be intuitive. Here’s how NASA and Snapshot Wisconsin work together:

NASA doesn’t just focus on our sun, solar system, and broader universe. It also has a dedicated Earth Science branch that houses research related to our atmosphere, weather, energy cycling, and ecosystems. This branch aims to predict change over time in, for example, energy cycling or biodiversity resources.

Some things, like animal population processes, are incredibly difficult to track across large spatial areas. Even with all the Snapshot Wisconsin cameras out on the ground, the total physical area in which we are observing animals is fairly limited – a point where each camera is, with a lot of space in between points. What we need to do is fill in the data gaps between camera locations. In other words, we need to be able to make predictions about areas where we don’t have observations. And this is where NASA comes into play.

NASA maintains a number of satellites that orbit Earth. These satellites carry on-board sensors that record light reflectance off of the Earth’s surface at different wavelengths. The images these satellites take of the Earth’s surface can be used to determine, for example, the locations of different landcover types (forest, wetland, prairie, etc.), or where leaves are growing or senescent, or where and when there is snow-cover. What’s great about these sensors is that they take photos regularly, and over large continuous spaces, so we can collect these data from our trail camera locations AND the spaces in between them.

Two of the more important sensors for our research are the Landsat and MODIS sensors. Landsat images have a spatial resolution of 30 meters (think of this as a pixel size – each pixel is 30 m by 30 m) and a temporal resolution (i.e., gap between flyovers) of 16 days. MODIS images have a spatial resolution of 250 – 500 m, and a temporal resolution of 1-2 days. These sensors are complementary—MODIS’s greater temporal resolution makes it more useful for detecting temporal environmental changes like plant green-up, while Landsat’s greater spatial resolution makes it more useful for detailed mapping of relatively static environmental attributes, like the location of forests, wetlands, and prairies.

Landcover maps of the same area derived from Landsat (left) and MODIS (right) satellite images. Modified from Schmidt et al. 2012.

How do we use satellite data and trail camera data together? We determine the association between the number of animals we count in trail camera photos and a series of environmental variables taken from satellite data. Understanding these associations gives us an idea of why animals might be more or less abundant in some places than in others, and allows us to suggest actions managers might take. For example, we might find that prairie chickens are highly associated with prairies but not with forests, and so we might suggest removing trees that are encroaching upon prairie land  in order to increase prairie chicken numbers.

Without images collected from space, it would be incredibly difficult to reliably predict and map the distribution and abundance of species.

How seasonality influences where animals are located is a major scientific focus of Snapshot Wisconsin.  Animals move around, so their distribution in space changes through time. Animals also tend to have different phenophases – characteristics associated with specific seasonal or annual environmental phenomena (think spring or fall migration) that can influence how often they are detected by cameras.

A classic phenophase found in mammal species is torpor or hibernation, and bears provide an excellent local Wisconsin example.  Whether bears are ‘true hibernators’ or not is often debated. Compared to many smaller mammals that drop their body temperature to near freezing, bears scarcely depress their body temperature at all.  Furthermore, small mammals tend to sporadically emerge from hibernation throughout the winter to pass metabolic waste and eat a little…bears do not.  Most amazingly, bears can go all winter without eating or moving without substantial muscle atrophy: unlike humans or most other mammals, bears can synthesize new proteins out of the nitrogen contained in their urea. While other animals wake up to pass waste, bears turn this waste into new muscle tissue.

One way in which Snapshot Wisconsin can contribute to our understanding of bear ecology beyond measures of bear distribution or cub production is by providing information regarding the timing of when bears seem to be active.  Bears are clearly more regularly photographed during summer and early autumn than during winter:

The pictures provide an indirect cue as to how bears are behaving and the timing of their hibernation.  This information can be used to evaluate hypotheses regarding variation in bear hibernation behavior–for example, there is some evidence that bears living in close proximity to humans and human food sources enter hibernation later than bears primarily consuming wild foods.

Seasonal patterns in wildlife images can also provide useful population-level information. Let’s visualize the photographic rates of a species that does not practice hibernation, deer:

There is a less pronounced seasonal pattern (a large drop in June reflects a slightly unbalanced effort across the year–many cameras were being set up at this time in 2015). Still, there is a drop in late winter and early spring that is consistent with what we know about the annual population dynamics of deer in the state–most adult deer die during late winter, and late-winter abundance should always be lower than pre-winter abundance (for comparison, one can find estimates of deer abundance based upon harvest metrics here and here).  The number of pictures taken during any one time period is subject to a lot of variation beyond changes in population size, and we do not imply that a 50% increase or decrease in the number of pictures corresponds to a 50% increase or decrease in the number of animals.  However, the count and timing of pictures is an input for statistical models that can correct for other sources of variation and be used to formally estimate things like changes in population size.