Dr. Dalia A. Conde’s longstanding relationship with conservation ecology began to change when she walked into an “oasis” of technology in Portugal in 2018.
Surrounded by developers and programmers at the Digital Infrastructures for Research conference, her belief in HPC’s utility for greater ecological understanding was confirmed and reinvigorated.
Since 1996, she had traversed the fields of Africa, the jungles of Mexico, and the deserts of North Mexico tracking and collecting individual datum on elephants, bats, and jaguars (for which she was recognized with the Women in Discovery award, alongside the likes of Jane Goodall and Sylvia Earle).
However, when she devised the idea of synthesizing those datasets — generated by herself and the world’s many independent explorers, ecologists, and zoologists — into one index, which would include every known species on the planet, she was met with skepticism: “It’s impossible!” she was told.
Four years later, with the fruitful project underway, we sat down to speak with her about her transition from fieldwork to HPC and her challenges and efforts in crossing into uncharted digital terrain.
What was your path to HPC?
It was not on purpose — it just happened.
When I finished my PhD, I worked with jaguars in the rain forest, and the situation was not good given that the area where we worked was an area with a lot of organized crime from drug dealers. It was in the middle of the Maya Forest, and I was pregnant, and my husband had just got a job. I knew I was not going to do field work for a while with the jaguars for obvious reasons: When you become a mom, you think differently.
Then, I started getting really passionate about working with many species. I’ve worked on species conservation since day one in my career — since 1996. It’s a long path, so I don't know where to start.
But what I can tell you is that all this started when I wanted to really gather a lot of different types of data, and I wanted to develop an index of knowledge for every species on the planet. And when I started, everybody told me that it was impossible. That was in 2012. And I said, “No, it has to be possible. Why would it be impossible with all this technology that’s been developed?”
So, I did the first index, which was only for demographic knowledge. (That was published in Proceedings of the National Academy of Sciences.) And that required working with some programmers, but it was not big data. You don’t need HPC to do those type of analyses.
And from then on, I needed to do more types of analytics in order to understand species survival and mortality patterns. And of course, you need to look at a lot of competing models to see which fits better to the data, which can best represent biological process. And the only way you can do that is with HPC.
So, that’s how we got started.
And I've been very lucky because we have the Abacus 2.0 high-performance computer at the University of Southern Denmark. I got support from the university to use HPC — and, actually, they were quite happy as well because they were like, “Oh, you're the only person not doing what everyone else is.”
Was there a big moment in your transition from field work to big data analysis?
When I started working on this, I was invited to the conference of the Digital Infrastructures for Research in Portugal, and I could only be at that conference for one day (because the next thing I had to go to — on the illegal wildlife trade — was happening the same week).
But, for the day I was there, it was like being in an oasis. It was magic.
"I have a lot of women inspirations from Wings World Quest, as well. I have a lot of mentors there.: ~ Dr. Conde
I saw the different technologies that were being developed by all these people. They’re brilliant. They’re working on technologies like machine learning and artificial intelligence to gather a lot of different types of information. You know, high volumes, high variety — the three V's of data.
And then it was like, “What I'm trying to do is totally feasible. It’s totally feasible.”
I realized I was living in the Middle Ages and [the HPC world] was in the Renaissance, you know.
And that was cool — being introduced to these communities of computer scientists and people that do HPC. It was then that I discovered what I had discovered was only a tiny bit of the array of possibilities.
What were some of the challenges you faced?
At the beginning, I was doing some of the programming, and it was so limited. Then we started working with a programmer, which was a very good interaction.
But, still, he was not at the level we need now. He was great, and I'm really grateful. But now, we need to get into HPC. We need to integrate data from many different open databases. We need to deal with many issues where you need people with expertise on artificial intelligence and on text mining, et cetera.
"Now I'm getting data from all these databases, and I'm not working with one species; I'm working with more than six thousand... I feel I'm exploring the digital world now, mapping it." ~ Dr. Conde
Many people ask “Why do you need HPC? We do images, we do things that are logged.” Yes, but we have to run thousands of simulations and models, and for that, you need a lot of computational power.
And I will be very honest. I am still not working with the right crowd. I think the programs we’re doing with HPC are very bad, because we’re biologists doing the programming. We’re not proper programmers. And that's why, for me, it’s so important to go to PASC21 [an upcoming computing conference at which she’ll speak on exploring and mapping data landscapes] — to meet programmers.
Because we’re trying to launch initiatives to see if I can get more support. For some reason, the breach between policymakers, biology, and computer science is so large, and everybody's so into their own world. There’s this huge opportunity right now.
How does your current work compare to your earlier field work with jaguars?
Well, it's very different type of work. Of course, I miss the field work. But the level of insight that I get now working with HPC is unbelievable.
Of course, this level of insight comes from people that are doing the work in the field. You know, people that do the work in zoos and aquariums and rescue centers and sanctuaries.
Now I'm getting data from all these databases, and I'm not working with one species; I'm working with more than six thousand. So, it’s hard to compare, but I'm very excited. I feel I'm exploring the digital world now, mapping it.
But instead of the fields in Africa or the jungle in Mexico or the deserts of North Mexico or Belize, I’m exploring the knowledge of species in the digital world. And the only way to do this is to put all these experts from [the HPC] community in your expedition.
We need the experts. We need the Fridtjof Nansen (He’s the one who discovered the North Pole.) of machine learning or artificial intelligence, so we can do this exploration and make a system that is alive.
Are there any mentors or figures who inspired you or played a significant role in your journey?
Yes. Yes, one was Norman Myers. He was a conservation biologist from England, and he used to come to the university where I did my PhD, Duke University, every year and he was a man that worked across so many areas of knowledge.
His focus was on biodiversity conservation, but he’d look not only at the biology, but also the policy and political instabilities. He’d look at the subsidies that had effects on fisheries, you know — the subsidies that are bad for the economy and bad for the environment (but that we keep doing anyway). He’d look at what we now call environmental refugees.
He looked at many, many areas — the connectivity — of this knowledge. He was not mapping the knowledge, but he was looking at these intersections.
And I have a lot of women inspirations from Wings World Quest, as well. I have a lot of mentors there.