How to become a conservation data scientist

Dr. Dalia Conde’s incredible journey from wildlife biology to zoo data champion

I met Dr. Dalia Conde for the first time in July at our staff retreat in Minnesota. Her personality is as big as her passion for her work. Picture a biologist who has tagged jaguars in the rainforest, a conservation scientist advocating the value of data for her work, and a leader casually riding her bicycle through the office with a huge smile on her face. Even if you weren’t already committed to conservation, you couldn’t help but be inspired by her infectious energy and vision.

From orangutans and bats to wildlife biologist

She was fascinated by wildlife from an early age and credits the iconic National Geographic photo of a child with an orangutan as being the initial spark. She says, “Seeing a woman with a child working in the rainforest made such an impression on me. I immediately went to my mother and demanded to know why I was growing up in Mexico City instead of in the rainforest.” Dalia later decided to study biology at the University of Mexico, although on a trial basis at first. As she puts it, “I thought I would try studying biology for six months and, if I didn’t like it, I’d study cinema or something else.”

Fortunately for science, one of her early field expeditions as a student volunteer sealed the deal. The project she worked on was studying the differences of seed dispersal in the forest by bats versus birds. Since they were also looking at differences of bat diversity between different forest regions, she spent time traveling from the rainforest to coffee plantations. The impact of deforestation left a deep impression on her.

What really struck me was seeing how the forest was being eaten. It’s only in the protected areas that I experienced the beauty of the jungle; the noises and the sounds; the species and the colors; and the intensity of life because there’s life everywhere, every second – it’s just so alive. And then seeing how that was going to disappear and be lost.

–Dr. Dalia Conde, Director of Science, Species360

The contrast was even more marked when flying out of the jungle to pick up supplies in Guatemala – the forest completely disappears by the time you reach the city. She describes her experience – “You start seeing roads like veins with fishbones of settlements sprouting from them, and it looks like the jungle is being eaten by the roads like a disease. People start moving there and the forest is getting lost. The argument at the time was that its good for the economy since road bring development, but it made no sense to me since the people who move there are still very poor.” The trip really opened her eyes to the complexity of conservation work – it’s not just a biological issue but also a socioeconomic and political one.

A quick intro to zoo data, courtesy of the jaguars

Dalia’s journey from conservation science to zoo data was a bit of a winding path but no less fascinating. Her academic advisor, Rodrigo Medellin, sent her on several field expeditions. This led to an with an opportunity with an NGO to work on a project to reintroduce pronghorn antelope from the southern USA to Mexico. Selecting the right antelope populations for reintroduction involved looking at similarity of habitat but also genetics. Since part of the work involved determining which antelope were genetically closer, it was the first time she began to work with species data from zoos. She partnered with Oliver Ryder, Kleberg Endowed Director of Conservation Genetics at San Diego Zoo, on the study. Even though she also spent time doing field analysis as part of the project, she was already beginning to realize how difficult it is to get data from the field.

When it was time for her to decide on her Ph.D. thesis, she hadn’t forgotten her earlier experience of seeing the forest disappear. She wanted to develop a model of deforestation drivers and decided to focus on the impact of road development on the forest ecosystem and habitat. Measuring deforestation alone wouldn’t tell the whole story, so she decided to concentrate on the top predator in the ecoystem, the jaguar, to measure the impact on habitat.

Because jaguars are so elusive and hard to capture, there’s not a lot of field data on them. Zoo data on the species helped to fill the knowledge gap. However, it wasn’t until she encountered her own AHA moment that Dalia really became convinced of the value of zoo data. Typically, habitat models don’t take gender differences into account, but the zoo data showed marked differences in behavior between male and female jaguars. Once she adjusted the habitat models to account for this, she saw a huge difference between the male and female jaguars in the wild as well.

If you take all the data points from the GPS trackers on the jaguars and analyze them without the gender difference in mind, you lose a lot of information and get the wrong answer. That’s when I realized just how valuable insights from zoos can be.

–Dr. Dalia Conde, Director of Science, Species360

The leap into biodiversity big data, courtesy of the zoos

Her introduction to large scale zoo data came later, as part of her work at the Max-Planck Institute for Demographic Research. When she started to explore how demographic aging models could be applied to wildlife for extinction risk assessments, she quickly encountered significant data gaps in threatened species – the basic aggregate data needed for analysis just didn’t exist. There was no way to discover answers to even basic questions like how many threatened species are in captive populations. One of the researchers at the center introduced her to ZIMs data which became the basis of her 2011 paper, “An emerging role of zoos to conserve biodiversity”, which she coauthored with Nate Flesness. Having Nate as the data expert was very important for credibility. She says, “I didn’t want to publish without having the expert on the data involved because you can make wrong assumptions.”

Although the paper was a success and frequently cited, it wasn’t smooth sailing for her follow-on research work. There was a lot of opposition around the value of zoo data for conservation. “I had a lot of people telling me that [zoo] data is useless because it’s from captive animals, there’s a lot of quality issues, you can’t use this data for animals in the wild,” she shares. She was even turned down for an ERC grant because the reviewers felt that there was enough data available on wild populations, negating the need for zoo data. Rather than slowing her down, it galvanized her to address the challenge head on – she set out to map the actual knowledge that exists for every species.

For some species, if we are able to address the uncertainties between data from zoo and wild populations, we can take advantage of the wealth of data collected by Species360 members to conserve wild populations.

–Dr. Dalia Conde, Director of Science, Species360

Mapping new insights with DISKo

The central question – how much demographic data exists for any given species—led to DISKo (Demographic Index of Species Knowledge). DISKo is a collaborative venture to standardize data across 22 databases to create an index of basic demographic knowledge for each of the world’s vertebrates. The index will be key to supporting research and conservation planning in the future.

As Dalia and her team (Lionel Jouvet, Johanna Stärk and Rita Silva) continue work on expanding the index, they are discovering an unprecedented opportunity for ZIMs data to plug vital knowledge gaps. By starting with more accurate base data, the models they are developing will improve conservation planning in areas like the IUCN Red List Assessment for threatened species and TRAFFIC illegal wildlife trade monitoring. Future goals for DISKo include integration of habitat information along with demographic data for deeper predictive models. This is especially critical in light of rapid habitat loss in many parts of the world.

Species360 data has incredible value for conservation science. With 42 years of sharing data for 21k species in over 90 countries, the potential impact is huge, especially if we can help a global organization like IUCN with the knowledge we have here.

–Dr. Dalia Conde, Director of Science, Species360

Can we learn from the past for the future?

As scientists predict that extinction rates of species are much higher than previously thought, what we know versus don’t know is clearly critical. Decision and planning models are built using best guess assumptions when data is unknown or unavailable. However, the slightest bias in assumptions can lead to vastly different—even wrong—outcomes, like Dalia’s experience with male versus female jaguars. That’s why plugging knowledge gaps with real data, like ZIMs, is essential for smarter conservation in the future. Looking back at the poster child for extinction illustrates just how just how significant the smallest unknown can be:

When the dodo bird went extinct, the response was probably ‘who cares – why should we stop development to save one species?’ It took us almost 200 years to discover that the native calvaria tree depended on the dodos for germinating their seeds. That’s why there were no seedlings of the calvaria tree on the island to be found. The tree is important to the whole ecosystem of the island, for climate, replenishing the aquifers, for other bird species.

–Dr. Dalia Conde, Director of Science, Species360

Originally published on the Species360 blog on July 27, 2017