This post is written by Sara Hendery, Communications Coordinator for the Feed the Future Innovation Lab for Integrated Pest Management.

The majority of the world’s food comes from just three crops — rice, wheat, and maize — and they alone suffer 16 percent losses from invasive species.

In a time when both international trade and human mobility is rapidly increasing, so is the spread of invasive species, leading to devastating effects in global food production. Invasive species cause countries billions of dollars in losses every year but are especially burdensome for rural communities in the developing world that depend on natural resources for survival.

Abhijin Adiga leads an invasive pest modeling project for Virginia Tech’s Feed the Future Innovation Lab for Integrated Pest Management. Key to combatting the threat of invasive species, he says, is knowing where and when they will arrive.  

“With advances in ‘big data’ techniques and high-performance computing, complex phenomena such as invasive species spread can be modeled and studied extensively,” said Adiga, who is a Research Assistant Professor for the Biocomplexity Institute at the University of Virginia. “Such tools can help policymakers make informed and timely decisions about trade and where resources should be placed.”

Adiga’s team models the spread of one particular invasive pest — Tuta Absoluta. Native to South America, the pest has now reached Africa, Asia, Europe, Central America, and the Caribbean, where it has gained notoriety for wiping out 100 percent of tomato crops, subsequently increasing tomato prices and decreasing farmer income.

“With no known ‘silver bullet’ for combating the pest, simply being prepared for its arrival will make a major difference for farmers,” Adiga says. “The consequences of invasive species spread are often not known until after their arrival.”

Instead of using ecological niche modeling, a modeling approach typically used to estimate the geographical suitability distribution of an insect, Adiga’s team applies a novel modeling approach. The system factors in human movement, an aspect usually considered in infectious disease modeling. After research revealed Tuta absoluta predominately spreads through human-mediated tomato shipments, it was critical that Adiga’s team adapt the model.  

Adiga discusses further the application and value of invasive species modeling:

How does your team model the spread of invasive species?

Our objective is to study the multi-pathway spread of invasive species, both naturally as well as human-mediated. To this end, we look at the pest’s spread as an epidemiological process over a network. The network is composed of locations and two locations are linked if there is a possibility of the pest directly moving from one to the other. Naturally, if the two locations are close to each other, the pest can fly or be carried by wind — making adjacent locations linked to each other. However, it is possible that locations hundreds of miles apart are also linked. For example, pest-infested host crops can be transported from one market to another.

What findings has your invasive pest modeling work revealed thus far?

Our studies establish a high correlation between trade flows and spatio-temporal spread patterns of Tuta Absoluta. Our models indicate that major production areas close to urban centers are highly vulnerable to infestations. This is because — due to high demand for tomatoes — urban centers attract imports from several regions (domestic as well as international). This increases the probability of introducing the pest into this area. Our models show that domestic spread is quite rapid; within two to four years of introduction, Tuta Absoluta can spread all over a country due to high volume of domestic trade.

What is the value of invasive pest modeling data for our present and future world, and what challenges do you face?

Current state-of-the-art modeling approaches mainly focus on long-term suitability of the establishment of pests. There is a great need for methods, however, that provide possible spatio-temporal spread scenarios for emerging pests. Such information is crucial for monitoring efforts, prevention of introduction, and mitigating the spread if the pest does get introduced.

Our models require a lot of quality data, which is generally not available — particularly in regions where USAID operates. In such situations, extensive sensitivity analysis and uncertainty quantification is necessary for robust predictions.

What is the best-case scenario for how your pest modeling work will be applied?

The number of invasions from alien species has only been increasing in recent times. This is mainly due to globalization. In particular, our food systems are increasingly comprised of concentrated production centers and long-distance transportation of commodities, making them very vulnerable to attacks. Not only that, but elements such as climate change can facilitate invasive species spread even further.

Simply knowing where Tuta Asboluta will arrive and when has aided us in conducting awareness and management workshops around the globe to help communities prepare, training farmers and researchers on the use of such technologies as pheromone traps. My team also received an additional grant from the U.S. Department of Agriculture to map the pest’s entry into the U.S., safeguarding the country’s billion-dollar tomato industry.