Webinar presenters Amanda Bixler and Dr. Andrew Laurence answered a number of your questions after their presentation, Forensic Palynology and the Search for Geolocation. Here are just a few of their responses.
Audience Question: an earlier slide stated that it’s not admissible in court if results can help focus an investigation that uncovers evidence that is admissible what later happens with the pollen evidence. Is it completely discarded?
Dr. Andrew Laurence: To be honest, I actually don’t know the answer that question usually once I submit the record, that’s it on my end Amanda. Do you know what happened with Bella?
Amanda Bixler: I don’t believe that it was it. Was it the pollen just basically help them prioritize and manage those leads. So once it came in that used other investigative resources to then confirm Bella’s identity.
Audience Question: Is it possible to take pollen samples in an autopsy situation from the body such as lungs, airways, or other areas to help trace where a person may have been before they became deceased.
Dr. Andrew Laurence: Absolutely and that’s exactly what they do in the United Kingdom. So they have a forensic palynologist there and the vast majority of our cases are actually taking pollen from the morgue. Taking swabs from the nasal cavity and they have actually gotten from the lungs.
Audience Question: Is anyone training to become a forensic palynologist to increase the number of personnel. And how long is that program?
Dr. Andrew Laurence: Unfortunately, there are no actual programs for forensic palynology. So I think way back at the beginning I kind of mentioned most, well all forensic palynologists are either geologist our archaeologist by training. We’ve all been kind of recruited along the way. So I was recruited in grad school while working my dissertation in archaeology. Shannon was also same thing, only in geology. She was recruited as a student brought in through our step(?) program. Palynology itself is typically Ph.D. graduate-level work. Most people end up going working for the oil industry where all the money’s at. But occasionally we get people like us that are willing to work for the government because we want to work and serve the public rather than the private sector. We are actually in the process of hiring a third person and we are moving along with that. So, I would recommend if anyone’s interested in going down this route because we desperately need more people all is to find a program of probably like LSU, Louisiana State University, is probably your best bet within the United States to study palynology at the graduate level. And I kind of go from there.
Audience Question: Are you aware of any local law enforcement agency, so by local I’m assuming they mean US law enforcement agencies or even private labs that are looking to expand their forensic palynology capabilities?
Dr. Andrew Laurence: Private, yes. I don’t know about the local like local PDs. Part of the issue is again. There aren’t that many palynologists in the world to begin with and most gravitate towards the oil industry. So, the chemicals we use are quite dangerous and you have to have licenses to even buy them and use them. So that is also a limiting factor. In fact, a lot of universities that used to have pollen programs that ended up falling to the wayside because of that reason. We just couldn’t keep up with certification and the cost involved with those chemicals. For us, we have all that just because we are with the federal government. I’ve seen a few labs, in general, are already set up to handle that type of stuff. So yeah, there are a lot of limiting factors involved. And number one is the problem with personnel. Unfortunately, pollen can’t be automated. They tried a number of times both in the private sector and the government and it just can’t be done with our current technology. So you are limited to two actual analysts doing the work got it.
Audience Question: And actually that was you touched on another question that we had whether or not the polling is being identified manually by yourself of if you have software or something like that that does it, and it sounds like it all has to be done manually by people like you.
Dr. Andrew Laurence: Yes, so there have been three main tracks that have tried for automation. First is DNA which seems like that would be a good route. However, there are a lot of problems with that first is, pollen actually only has pollen in its cytoplasm. The outer wall, now if you see that picture on the screen, there’s no organic material left. The pollen we look at is just an empty shell. Pollen loses its DNA within a few hours to a few days after deposition in the real world. So if you actually Google this, you’ll see all sorts of people saying, “Oh, we can do this,” they can’t. Because they’re they are only looking at laboratory samples. You know, pollen taken directly from the plant under laboratory conditions. Of course, that’s going to have the DNA as soon as it’s out in the real world, it loses it within a few hours. The reason being to remember, these are flying sex sells the point is to get this DNA out of itself and onto a plant so most as soon as it impacts another object it gets ejected out of it. That’s just the way that pollen works. Allergies are actually caused by the beta 1 receptors and starch on the inside of pollen not the exterior. I’m talking about respiratory allergies, not skin allergies. So the pollen itself is actually too big to do anything to you internally. So what’s happening is it breaks open in the environment? It’s releasing the cytoplasm into the air and that’s what you’re breathing in that’s causing an allergic reaction. So that is basically how pollen lose its DNA in the real world. Number two is DNA can only do presence/absence which is meaningless for geolocation. So remember, it’s all about percentages. The pollen assemblage is all about each pollen grain compared to one another and what percentage is because pollen isn’t made equally. So, you can have the same species list, of course with DNA you get down the species-level identification. However, you say white pine, pinus strobus, as an example. That exists all the way from New England, including Canada, all the way down to Florida all the way west to here in Illinois. So finding that in your sample doesn’t mean anything and that’s a huge huge area. However, the percentage of that is meaningful because some parts, if you find five percent of that pollen, that means you’re in let’s say, I don’t know Ohio, Northern Kentucky. You find 10% you’re in like New York and so forth. So that’s another limiting factor and the other issues some pollen grains are designed to just crack open like an egg, they’re called the CTC pollen types. They don’t have any apertures at all. And so as soon as it hits an object, it’s evolved to just crack open and lose its DNA. So that’s a problem right there. Image recognition has also been tried. However, again, that’s something that works wonderful with 98% plus success in laboratory conditions from home straight from the plant. However, as I had to kind of mentioned what the Marcia King case the pollen is in the real world is beat up, damaged and computers actually cannot tell the difference between damage and actual features even with all the machine learning algorithms and neural networks that have been tried. Misidentification rate is extraordinarily high in humans out-compete easily. I mean their identification rate drops down about 30% in the real world. Other people have tried doing chemical analysis figuring that the wall structure is defined by lovely chemicals in it, that also turned out to be a bust. So pollen is wasting similar chemically. The only way to kind of get down to it is to do a species-level a hundred percent of the time do SEM, LM, and then TEM which unfortunately is incredibly costly and time-consuming. And to do a 200-pollen grain count you’re looking at one year worth of time to do that as opposed to human that can do that in a single afternoon.
Audience Question: What is the difference between NamUs and NCMEC?
Amanda Bixler: So NamUs is an online database that allows user input from the public to law enforcement to medical examiners, coroners for missing and unidentified persons cases. It allows them to have a repository for all of the biometric collection information that law enforcement gets for the dentals, DNA, fingerprints, and it houses it all in one location so that when these leads come in law enforcement can look to this information to see what’s available for any potential comparisons or what missing persons are out there, unidentified persons are out there that potentially could match their victim or their suspect. For NCMEC, we are the National Clearinghouse for all missing child cases. So in a child 17 or under, not all but most children under 17 get reported to us. So we provide assistance to law enforcement across the country for any of the resources that we have here at NCMEC. As well as unidentified cases that are believed to be children. You saw a couple of our cases that we presented on today. So we assist law enforcement and medical examiners and coroners with Jane and John Doe cases that are believed to be children or teenage age range victims.
Audience Question: The question comes in from Jennifer and Jennifer works and trains cadaver canines. Her question is has pollen been associated with odor scent-detection K9s to link at pollen source article with the police with a specific crime scene, and she’s specifically looking at scent associations with non-cadaver sources to try to increase the efficacy of searches?
Dr. Andrew Laurence: To the best of my knowledge, no, it’s never been tried before.
Audience Question: Does my agency needs some sort of MOU with Customs and Border Patrol in order to be able to work with you on analyzing samples? Are there specific case types other than missing person or homicide cases that we can submit samples for?
Dr. Andrew Laurence: So, no MOUs are needed. We just have a form they need to fill out a request for forensic examination form, which if you were to email me see my contact information there I can get that to you. We do like to vet cases as they come in because not all cases are actually good for pollen analysis. So just to avoid people chip stuff in that we get in like well, we probably can’t do a whole lot with this or help you out too much and just chip it back like that. Talk about it, say, “Okay, what is you have, what do you want to know?”, and as for other types of cases absolutely we can work on so we do lots of smuggling, persons of interest just for our own CBP casework. It’s just the focus of this presentation was missing persons and trafficking. But yes, we could analyze all sorts of stuff
Amanda Bixler: And I’ll throw in, one of the things that NCMEC assists these agencies with when they have the missing or the unidentified cases and they’re looking to do pollen analysis. We’re familiar with Andy’s process of what he’ll need to know, what background information things like that so that’s one of the ways that NCMEC can assist on those child cases to get them to Andy and his industry and for evaluation to see if it is a good candidate and then assisting with that process getting that evidence from the agency to Andy.
Audience Question: A number of the audience have asked variations on why pollen evidence is not admissible in court. I’m assuming Andy since you’re not an attorney, you probably can’t get too much into those details. But again, you know what they say about assuming. You probably aren’t really familiar with the legalistic details, but if there’s anything that you can share or Amanda you can share that might help answer those questions.
Dr. Andrew Laurence: So it needs to go through what’s called the Daubert criteria, which is basically a test of the court looks at to say okay. This is an actual science and it can be used in court. Basically, it’s the court system kind of putting a rubber stamp on the sign saying yes, this is good to go. We just haven’t done that. We probably could because other countries do it which is kind of like the first step of saying okay, it has been used and yes, there are tons and tons of court cases out there from the UK, Britain, New Zealand, in particular where people have been prosecuted based on pollen evidence. It’s just we haven’t pushed to do that. We kind of like it in the intelligence world for right now, since there’s only two of us and it makes our lives easier and our turnaround time much faster. One of the biggest issues that we encounter, in court, the science side is usually being tested based on the methods. That’s what you will where the challenge comes in. And with pollen, because we’re dealing with biological samples, it changes quite a bit. So, for example, say you wipe down a suspect’s shoes. Well, we have our standard method that we use for analyzing a baby wipe taken from a smooth surface like a shoe. However, if there is some type of shoe polish on that shoe, it could react negatively with our chemicals which then requires us to essentially make up new methods on the fly to deal with that new variation and go from there. As it stands as long as we’re in the intel world we can do that. No problems whatsoever and deal with any type of issue. When you’re dealing with court, it makes it. It’s more difficult to do that. It can be done but it adds a lot more layers of red tape because now you have to justify everything to the point where the jury can be convinced of what you did was an appropriate action, as opposed to just being able to do the science to get the result out. Pollen sample is different. I mean, I can’t stress that enough how many times we’ve done the same thing over and over again same types of samples and every single time it’s required modifications of the method because of something like literally shoe polish has been a problem. Some shoes will have it some don’t and depends on the brand. Weapons we encounter that with different oils on the firearms themselves, some are problematic some are not. So, there’s a lot of variation that we have to deal with our analysis, to begin with
Audience Question: You mentioned that you use references. What are some of those references are there books with pictures of pollen grains, for example?
Dr. Andrew Laurence: Yes, there are. We also have our own column reference collection that was taken from vouchered specimen which means plants that we know exactly where they came from, what they are, identified by an actual botanist who stamped their name on and say, “Yes, this is the actual species.” We have several thousand of those, both Shannon and I are also affiliates at the Field Museum here in Chicago so we can go over there use their pollen reference collection as well as their library. Though we have our own library of those pollen reference books. As well as published sources saying okay where plants grow what are different vegetation zones where they exist? So yes a combination of books and actual pollen slides.
Click Here to Watch a Recording of Forensic Palynology and the Search for Geolocation.