Lecture 14 - Field Techniques & Mammalian Ecology

What are assumptions that must be met with the mark-recapture abundance estimates?

-equal sampling effort across sampling events -must allow time between sampling for individuals to redistribute themselves -assumes no new migration into the sampling area

Give an example of an eDNA metabarcoding approach with a semi-aquatic mammal.

A simplified version of what a diet analysis study could be doing is if there were a semi-aquatic mammal, a seal, that is consuming three different fish species at equal ratios (Herron, salmon, and pollock) of one kg each over a given period of time; there will be a mixture of prey DNA in the fecal sample. The DNA has gone through the digestive tract and will be really degraded, so we need to be sure we are only amplifying tiny DNA fragments. We would never be able to identify long fragments because it is being degraded, and the longer it sits in the feces, it will be degraded even more. All the DNA of the species present in the fecal sample will be amplified, which will produce millions of sequencing reads across all the different DNA fragments in the sample. Then, ultimately, the idea will be that we end up with the percentage proportions of the prey consumed. For instance, if the salmon were the main prey, we would expect the percentage of the sequenced reads to be higher for salmon. In this example, though, the data tells us that the seal is using these food sources equally, and thus the sequenced reads are approximately equal in percentages of overall reads for all three fish species. If the primary food source was determined to be salmon, it could tell us a lot about the seal's conservation management. For instance, if there is a crash in the salmon population and this is the main prey type for the seal, we will see a decline in the seal population or a switch to a more reliable food source. This can tell us a lot about the species' ecology and answer a lot of different questions potentially (about monitoring and management).

How can a mammal species be identified through their remains, scratches (or markings), or even burrows?

A way to identify a mammal species from their remains could be by identifying diagnostic characteristics. For instance, some carnivores have baculums, which are variable and can determine a carnivore down to the species level (penis bones can distinguish a carnivore species). Mammalogists need to understand the skull and post-cranial anatomy features to identify a species and potentially extract DNA from their remains. The remains of prey that a carnivore species have abandoned may tell us what predators may be present in the area, as well. Scratches or markings on trees can be indicative of certain species. Evidence of nests or birds or burrows (even if not currently being used) can give us insight into species present.

What are advantages and disadvantages to PIT tagging?

Advantages: -the lifespans are long; perhaps the mammal's entire lifetime (vs. radio collars where batteries die or they malfunction) -they are internally injected, so they are well retained and reliable -they are small in size and will not alter the behavior or appearance of the animal -they are quick to insert, which limits the handling stress time for the mammal -they permanently identify a large range of animals; tend to be only used for smaller mammals though Disadvantages: -PIT tags and electronic readers are expensive -they can migrate away from the site of injection and possibly cause internal damage -they must be either recaptured for the identification or near an antennae devise; the antennae left unattended often is damaged or removed by humans who only adds to their expense -implantation can cause handling stress with the possibility of infection when implanting the device -it is often better for smaller mammals with smaller range limits, or else we would need to set up many monitoring movements of a much larger mammal (not practical or financially feasible)

How can mammal species be identified by their scat?

An example of a non-invasive method is by observing mammal scat or collecting it to perform DNA analyses. By observing the scat in nature, you may not determine the mammal down to an individual level. However, you may be able to narrow it to a group of mammals at minimum (maybe Family or Genus). If we want to obtain more information about that mammal, we can extract DNA from the scat sample. The scat itself does not actually have the DNA of the mammal, per se. However, it has a lot of microbial DNA. As the mammal passes the feces through the colon, some epithelial cells that line the colon are also excreted and surround the scat's exterior portion, but this is not usually really great quality DNA. It is not easy to extract quality DNA from that scat, but we are getting better at extracting from sources that are not necessarily good quality samples.

What is an eDNA sample and why is obtaining them important to biodiversity?

An example of an eDNA sample would be obtaining a five-liter sample of water from a stream to answer whether river otters are using a habitat that has been recently restored and that they have been absent from for a while. We want to monitor when the river otters begin to repopulate this area. We take a water sample over a period of time to see if we can detect when they begin to use the habitat again. We could use a target DNA approach where we are only detecting for river otter DNA in the sample with really targeted primers that look for a particular sequence that is only found in river otter DNA. We will have tons of DNA present in a single sample of water, such as many different microbial genes or aquatic invertebrate species DNA. We could have spores or pollen landing in the water column as well. Other mammals could also be prevalent; many things could be in that single sample. This method is interesting because we have taken a snapshot of the biodiversity at that point in time. These samples should be archived when obtained because many questions that come up down the road that this single sample could answer. It is important not to deplete the sample entirely for the current study for this reason. Decades from now, potentially at that sampling site, biodiversity could be lost in addition to climate change and many other pressures on the environment. eDNA samples may be a way for us to track those trends over a long period of time. As we continue to take these samples down the road, we will start using them in exciting and innovative ways to answer novel questions and monitor biodiversity across time.

How can mammal species be identified by their call recordings?

An important way biologists study bats, to identify what species of bats are present, is through call recordings. In echolocating bats, many species will produce a call that is at a higher frequency (outside of human ear range of hearing), so scientists have to record the call, upload it to a computer, manipulate it so it is something we can hear and measure the range. They can then key into what species are present because the frequency of the call can help identify between species present.

Compare eDNA sampling with camera trap data in terrestrial mammals.

An interesting study compared eDNA sampling to camera trap studies (Leempoel et al., 2020). Camera trap studies use baiting a camera station (not actually trapping the animal) to catch them on camera to determine if they are in the area. The researchers used soil samples and a metabarcoding approach to sequence all of the mammal species present in the sample (for the eDNA) using two different types of primers, one more general and one more specific. They compared the eDNA results with the camera trap data. They determined that the camera traps found three species that eDNA could not identify (raccoon, long-tailed weasel, and domestic cats), and eDNA could find many more species than the camera trap could (~10). Many of the species determined with eDNA were small, rodent-like mammals, which would be difficult to capture using the camera trap method. There was a big overlap of species detected through both methods (~14), which could infer a good consistency across species population. A few domesticated livestock DNA was also found through the eDNA methods, but this could be due to agricultural run-offs and considered contamination of the sample. These mammals do not live in this area.

What are the different types of live traps?

Different types of live traps include Sherman & tomahawk traps (cage traps), pitfall traps (for smaller mammals), and mist-netting (for bats; while feeding on bugs over water sources). All of these methods, particularly cage traps, come with their biases. Some mammals are really good at avoiding traps and are wary of them. Others, as long as there is food bait, they do not care and will keep revising that trap often even though they were caged and presumably have some memory of the event. They do not seem to care as long as they get a reliable food source, but they are biased in these methods.

What is environmental DNA?

Environmental DNA is DNA that is shed from an organism into the environment. A molecular method that tells us about the species' ecology is through this technique and is a way to obtain information about a species DNA without even knowing if the species is present. eDNA can be obtained by taking a water, sediment/soil, or air sample. Air does not necessarily apply to mammals, but eDNA can be obtained in this way for other taxa, such as insects or fungus. The most commonly used for mammals is with water sampling. For instance, identifying an aquatic or semi-aquatic species. Sediment and soil have been used for a long time by microbiologists studying the types of microbes present in the soil. You can also take samples of soil from a river and can isolate DNA from there and amplify all the microbial DNA present within that sample, an eDNA sampling method. eDNA typically uses mitochondrial DNA because there is a higher chance of species detection. DNA may be degrading, and we only have a bit of DNA in a sample, so we will want to use mtDNA because it is more prevalent and will be the DNA type we are most likely to detect. More typically, if we are looking at mammals, eukaryotic organisms, we will target the mtDNA and not the ncDNA because ncDNA has two copies of the genome (maternal/paternal copies). However, there will be a higher number of mtDNA within a cell, depending on the cell type. For instance, brown fat is metabolically active, and there are many mitochondria within this cell type. This is a big area within wildlife biology research that is a rapidly growing field. There is much interest in this area for conservation applications, monitoring species present, and managing them.

What can be determined with eDNA samples at sampling sites?

If we are taking water samples, we are looking for an aquatic or semi-aquatic mammalian species, but we have been able to identify terrestrial species. In areas where there are not many water sources (dryer environments), there may be watering holes with a lot of species richness, and you may be able to determine what that species richness is by obtaining a water sample. Mammals using this may be crossing through the water source or just using it as a drinking source. They will be depositing some DNA along the way, and will be just a matter if we can detect it in our sample. If the mammal is aquatic or semi-aquatic, there is a higher chance they are depositing even more DNA into the water. Soil samples and sediment samples may also be important for terrestrial mammal identification. We are beginning to learn that DNA does not degrade as quickly in the sediments and soil as it does in the water. It seems that the DNA starts to break down in water into smaller pieces over time, and many factors go into this, for instance, the pH of the water, UV exposure, but DNA seems to be more protected in the soil and sediments.

How is ongoing monitoring conducted for mammal species?

It is one thing to put a tag on the ear of a mammal and recapture it periodically or figure out how long its lifespan was, but there are ways we can live-trap a mammal and put devices on it to use for ongoing monitoring. Many devices can collect information about habitat use (location and movements) of a mammal. Some examples are radiotelemetry, passive integrated transponders (PIT tagging), GPS and satellite telemetry, and biotelemetry. Radiotelemetry is a cheap and effective way to monitor mammals and is quite small and light (usually collars that go around the neck). The transmission distances are limited by geography and involve triangulating the location of the animal. PIT tagging is not common with larger mammals but may be better suited for smaller mammals, such as rodents. This does not require power or a battery and is an internal microchip inserted that collects data every time the animal passes near an antenna. The data is then transferred to a computer, and the microchip is usually inserted in areas such as the shoulder blade that has less chance of migrating or moving around. GPS/satellite telemetry is very expensive (up to 10k) and uses large collars or transmitters much heavier. There is no limit to the transmission distance. This method is not in the budget for many wildlife biologists, even though so much information can be obtained because it is not practical based on cost. Biotelemetry can record heart rate, environmental variables, movement, and speed of the mammal. Some sorts of collars can record the physiological characteristics of the mammal and environmental variables. For instance, marine mammal could record water quality characteristics while also recording the mammal's movement and speed. The only problem with these is that the units are often recording data within the unit, and then the unit must be removed from the mammal to retrieve the data to download.

What is a non-invasive method as a field technique?

Non-invasive methods have no cost to the species being sampled and no risk of decreasing fitness or death. The tools we can use to understand a species' ecology may be targeted or may be general and observational. Observation is usually the first step in a larger study. Non-invasive methods may be conducted through trail cameras with motion sensors, infrared flash, or could also be with spotlight searches that use strong lights at night to search for eye shine and movements of mammalian species. Another example of a non-invasive method is by observing mammal scat or collecting it to perform DNA analyses. We can also utilize non-invasive methods through tracks of the mammal, skulls, or other remains, scratches, markings, nests, burrows, and the remains of prey by carnivores. Call recordings (i.e., bat calls) or track plates and track stations are also other forms of non-invasive methods for research on mammals. Hair traps and snares, along with the collection of environmental DNA, are also informative non-invasive methods.

How would we use a live trapping method to obtain population size census estimates or abundance of mammals, and what are some examples?

One of the reasons we may be interested in live trapping a mammal is to get an idea of the census population size estimates and abundance. We may determine census population size estimates and abundance through removal methods (really only possible for small mammals); done a lot for fish and small rodents. What the removal method does is you set many traps out in a target area and then hang on to those animals. On a subsequent night, we set many more traps again while we hold onto the first group of mammals found. With the second set of traps, we can use a statistical method to compare the individuals we caught on the first attempt versus the second attempt. If we caught the same number in the second attempt, there are many more individuals present in the population that we did not capture the first time around. Another way to determine census population size estimates and abundance is through mark-recapture methods to use markers, ear clips, and tags. This is used a bit more common and is also called the Lincoln Index or Lincoln-Peterson Index methods. We will target a specific area (the given habitat we think the mammal is occupying) and randomly set up traps or use a systematic approach to capture individuals. Once captures, they will be marked and then set the animals free. The key to using this method is a statistical model where we will plug in the numbers for the first capture versus a second capture and get abundance estimates of the total number of species in a given area, the specific area we have targeted. Once the first set of animals are set free and redistribute similarly as before (allowing ample time in between capture and recapture), the second sampling effort will be conducted the same as the first (exactly as not to create bias in the experiment). There will be some newly captured individuals and possibly recaptured individuals who are marked from the first capture effort on the second capture effort. This model will give us an estimate of confidence intervals. The more we capture the first time and less new individual s on the second time, the more confidence we have because that implies that we are getting at a larger number of the whole population. We would. I hope to recapture more marked individuals to end up with a lower confidence interval. The other potential issue is assuming there is no new migration into the sampling area, assuming it is a closed population and can be a challenging aspect of using this method. This method is a way we often use to identify the abundance of a group of mammals.

What are the pros and cons with using non-invasive methods in research?

Pros: -There is a no "take" associated with live trapping, which potentially can cause death in the target or non-target species -there is no stress involved in this method (which could decrease fitness), especially for sensitive species -you do not have to locate an elusive species, especially if trap shy and reluctant to enter a trap -it is safer for the field crew; could be some level of risk with the crew handling wildlife -no (or less) permits are required; for invasive methods, you will need scientific collecting permits issued by state wildlife agencies (possibility of trapping or attracting a listed species in your area of study-may be denied) Cons: -The abundance estimates are not possible or maybe challenging -no (or less) metadata; such as age, weight, reproductive status, condition; blood samples can reveal a lot about physiology, genetics, tests for exposure to toxins; some disease are harder or impossible to study (including ectoparasites), and we would need to have invasive techniques and samples to do this -it may be harder to determine range size movements, habitat use -there may be biases based on where you sample; only targeting specific areas of mountain lions, for example, where we think they may have a prime habitat, but actually, this species is using higher elevation habitats than we would have thought; would need more invasive and GPS methods for this NOTE: There are guidelines for the ethical standards for conducting wildlife research - live trapping and handling.

Describe the method and data gathering of radiotelemetry.

Radiotelemetry monitors movement over time of the mammal, especially if they live in remote locations. As opposed to GPS or satellite technologies (not used as frequently), radiotelemetry is using these antenna devices to triangulate where the mammal is in space; the receiver will beep depending on how close the mammal is to the antenna, and then a few people will try and triangulate the location of the mammal. Wolf biologists would use a helicopter to do the radiotelemetry study. A couple of helicopters would fly over different areas to determine when they pick up evidence from the receiver indicating they were in the collared wolf's vicinity.

Compare Sherman traps with Tomahawk traps (cage traps).

Sherman traps are used for small mammals, such as rodents, moles, or shrews. They are not see-through, particularly to prevent attracting predators to the trap. Orange flags are typically placed near these traps so they can be relocated. There is a concern when using these traps because after they are set, it is difficult to find them all, and you may have a mammal in an enclosed area that, if not released within a modest amount of time, could be exposed to elements, such as cold/heat stress and limited food sources. Much logistics must go into this process and ensure it is done ethically and humanely. We want to limit the risk of death to those susceptible to losing heat to their environments. It is also important to include some bedding within the trap so small mammals can make a nest and stay warm because usually the traps are set up and kept out for 24 hours, and they will experience lower temperatures at night. Tomahawk traps are commonly used to trap medium-sized mammals and come in all different shapes and sizes. They are targeted at specific species, typically best practiced to choose a trap that is just the right size for the target species to avoid trapping mammals you are not interested in, trapping a coyote instead of a fox. If you choose a smaller cage, it will allow the fox in, but not the coyote. There is much effort that goes into setting up these traps. Another issue that arises, especially with foxes, is that the mammals may pick up the human scent on the trap or know the trap is foreign to the environment. The trap needs to sit out in the environment for a while and covered with leaves or branches. The trap could be baited but not set to encourage the mammal to enter the trap, and once more interested; the trap could be set. Traps must be revisited every 24 hours or less so that any sensitive species could be let go. GPS could be used in remote locations to obtain data of where exactly they set the trap (instead of using flags).

How is eDNA collected in the field?

The collection of a soil core or sediment sample can be accomplished in the water with a net or collect several liters of water, and the water can be passed through a filter. There are different filter options (pore sizes) to collect different sized particles. For instance, tiny pore size can collect cells present in the water and actual DNA free from the cells (extracellular DNA). This means the cell has already been degraded, and the DNA has already been removed from the nucleus. The DNA would be collected on the filter, brought back to the lab, and extracted from the filter. When using eDNA, there is a concern with contamination. Contamination can occur through field gear, so precautions to prevent this must be implemented.

How can mammal species be identified through eDNA applications?

There are a couple of ways we can use eDNA to identify mammal species presence or absence in an area. We could use PCR assay where we amplify the DNA in a sample using primers that target a specific area of the genome. We can use more general primers to identify a ton of different species (mammals, fish, reptiles, etc.) of a sample and record their sequences or we can just focus on a specific species and only detect if that species is present in our sample with species-specific primers. Species-specific primers that target the genome's area (say for all mammals) will exclude microbes, fish, reptiles, etc. We could just use primers that will only amplify mammalian DNA and will allow us to sequence that DNA to determine the diversity of mammals present in the sample. There is much interest in using eDNA for the early detection of invasive species. An example of a mammal in California that is of interest is nutria, a large semi-aquatic rodent that is very invasive, and by using eDNA, by taking a water sample, we could detect the presence of this species. It can tell us if this invasive species has been successfully expanding in an area. If we have a particularly rare or elusive species (hard to detect in the field), then eDNA can determine if their DNA is present. This means if it is detected, then the species is present somewhere in the area, and we may want to do more targeted fieldwork in this area. We could potentially use this to understand the distribution of a species range-wide because community assemblages can also be used to compare diversity across sampling sites and are typically used for the range-wide study of species. There is a lot of research looking into the relative abundance of species (in early-stage research) to determine what species might be present. If we had a sample from two different sites, for instance, and found more copies of, say, beaver DNA at one site vs. another, this could give us some insight into more beavers present at that site.

How can mammal species be identified with hair traps and snares?

To extract DNA from hair, the root is needed. Most hair does not have any DNA evidence in it, just protein, and we need the living root to be pulled out and will need several hairs to get a good DNA extraction from the mammal. An example of this would be a bear crawling in under a hair trap. The bear has to crawl under the wire, and as it does, it will be brushing up against the abrasive area that will catch a bunch of its hair. Scientists also place structures around the burrows of small rodents, and when they enter the burrow, there is a lent roller at the bottom they will brush up against, and it will remove some of their hair.

How can mammal species be identified with track plates and stations?

Track plates or track stations are a way to get an animal to walk through an area that has been baited, typically with food, to encourage the species of interest to come by, and then their tracks can be recorded. The way this is accomplished is by setting out soot before and adhesive paper with some bait at the end, so the mammal must walk through the soot and deposit its footprints on the adhesive paper when it reaches the bait at the end. Then the species can potentially be determined by the size and characteristics of the print.

What are two approaches methods with eDNA analysis?

Two approaches to the eDNA methodology are 1) targeted species identification and 2) metabarcoding. The targeted species identification approach is used to determine a species' presence or absence in an area. A caveat to this is that we can never be certain if a species is absent even if we did not detect its DNA by sampling the site. This approach is really cheap to execute and can be done quickly in the lab. It may be better for more monitoring and management of the species as a kind of ongoing assessment to determine if the species is using the area of habitat. It is cheaper, especially if primers are available for that species and have a faster turnaround time than the decoding approach. It answers specific targeted questions, such as the presence of a specific species. The metabarcoding approach is made by sequencing all of the species present in that sample within a group of interest, i.e., all mammal species. This is a more costly method because we have to do more in-depth sequencing, which we do not have to do with the target sequence species approach. With metabarcoding, we will get a ton of data, so it is a slower turnaround and more expensive to look at all the reads of DNA for mammals in a given sample. Although we can compare diversity across sampling sites (i.e., diet analysis), we might be able to answer some broader ecological questions.

How would we use a live trapping method to tag mammals, and what are some examples?

Using live trapping to tag mammals could be as simple as marking them with a number so we can record data over its life. We can recapture the animal to continue getting data on it and monitor its reproductive success, weight, and disease status. Also, when found deceased, we can figure out its lifespan and cause of death. Sometimes we tag the mammal when it is relatively young to monitor it across time. We can also use ear tags, "helmets" (although this is not popular with the public), and even tattoos and branding to identify and track individuals. We could also include camera collars or radio collars. Marine mammals are tagged with sensors that record data on dive depths and times spent at the surface (biotelemetry). We can often use different types of satellites or biotelemetry to obtain information about the mammal environment, and the sensors may record data on the dive depths and amount of time spent on land vs. in the water. It provides insight into the daily movements of mammals. We can use radio collars, which bighorn sheep have been candidates for, or camera collars (pretty expensive) that record data on the environment around them (like a go-pro on an animal) but not commonly used because of the expense.

Describe the reflection of the lens in mammals eyes when conducting spotlight searches.

We can identify a mammalian species or at least a group of mammals by the reflection we see in the lens of their eye. This is particularly effective for arboreal and nocturnal species. Animal eye shine is due to this reflection in their eyes because they have a special surface behind the retina called the tapetum lucidum, reflecting light and is responsible for many mammals having a lot better night vision (than say, humans). The color reflected in a photo or in real-time can give you an idea of what mammal might be present. For instance, foxes and rabbits reflect red light, felids reflect green, raccoons reflect yellow, and deer reflect white back. Spotlight searches can help identify the type of species by the reflection of the lens in mammals' eyes.

What are other eDNA sampling possibilities?

We do not necessarily have to take a water sample and filter it or a soil core sample at a small, randomly selected, or targeted site where we think a species might be present. In addition to using, say, track plates (if their tracks cannot determine the species), we could also obtain dirt samples from a track mark and extract the DNA to see if we can determine the species. DNA from their actual track print in the dirt or snow can determine this. To obtain DNA from snow, the snow would be melted down to a water sample. These methods have been accomplished on carnivore mammals. This method can give us much additional information and remove the issue of misidentifying the species from identifying it from tracks or feces visually without analyzing the DNA. Another thing we can do is if there are urine samples in the snow, we can melt that down and get DNA.

How is the live trapping of large mammal species conducted?

When using live-trapping for larger mammals, cage trapping is not as frequently used. It can be challenging to get them into a cage if it is an animal that has been habituated (bear in an urban environment) if it could be easier to capture by luring it into the cage with human food. For mammals inhabiting less urban areas, especially where we would not be able to get a giant trap out into (backcountry), biologists will often opt for using tranquilizer darts, many times from a helicopter. This is not a gentle process and is very stressful for the mammal. Trapping large mammals can be difficult and has a lower success rate. This is often how large ungulates are studied in remote areas, such as bighorn sheep, pronghorns, and mountain goats. They live in areas that humans are unable to access easily so what biologists will do is utilize net guns via helicopter to chase the mammal down and first catch it in the nets prior to tranquilizing because the mammals are really fast and challenging to capture. Then because of the remote area and difficult access, the biologist will hoist the ungulates from the helicopter and move them to a base camp where data will be collected; they will also mask the mammal to ensure they will not be stressed during this process if still semi-coherent.

How is eDNA metabarcoding used for diet analysis?

eDNA metabarcoding can detect all biodiversity in a sample or just all the mammalian diversity. It can also be used to perform a diet analysis of feces to determine a species' diet. It can determine what food sources a given mammal might be using by sequencing all of the plant species, for instance, found in the feces of herbivores. This gives us some insight into the primary source or staples in the diet and whether some mammals have more specialized or generalized diets to certain plant families. It could determine where the herbivore is spending most of its time or more focused on a particular plant species. It will tell us a lot about the ecology of these mammals.