Seven threads, one question: how do cities reshape the planet? My work answers it through the lenses of land change, infrastructure, biodiversity, human–environment interactions, health, the spatial models that hold them together, and the geospatial craft underneath all of it.
Urban land is the single most concentrated form of human transformation of the planet, and yet long-term high-resolution maps of where it has actually grown remain scarce. My work reconstructs where urban land has actually appeared over the past three decades, using long-run satellite records, and projects it forward under a range of development futures. The aim is to keep regional differences visible: a single global growth curve hides the places where the change is fastest and the consequences heaviest.
Ongoing work covers a global ensemble forecast of urban land-cover conversion, and a reconstruction of urban expansion in Bangladesh from the mid-1980s with projections to 2050. Both are in preparation. A study of urban growth around protected areas is in revision.
My doctoral work produced the first global-scale road network growth model, calibrated against historical road expansion and projected forward under multiple development scenarios. AGU's Eos magazine featured the work shortly after publication.
Ongoing work extends the model in two directions. A manuscript on hierarchical road network growth is in revision. In preparation are a study of how infrastructure siting decisions shape future hazard exposure, and a forecast of healthcare access in India and Bangladesh that treats road and urban growth as one coupled process.
Protected areas are the main instrument the world has for holding onto habitat, and this thread asks how well that instrument is actually working while cities and roads grow around it. The questions run from a single delta to the whole terrestrial network: which habitats are being consumed, which species ranges are contracting fastest, and whether a boundary on a map still means what it did thirty years ago. The recurring finding is that the pressure now sits at the edge rather than inside the line — reserves are not being built over so much as surrounded, and their connection to everything around them is quietly severed.
Two manuscripts are with journals. One, with Burak Güneralp and Lee A. Fitzgerald, is under review and covers the combined impacts of road development and urban expansion on biodiversity in Brazil, Indonesia, and Nigeria. A second, with Güneralp, is in revision and measures how urban growth isolated the global protected-area network between 1992 and 2020. Two further studies are in preparation: one on how city size scales with habitat loss, and one on urban expansion and biodiversity loss in Bangladesh to 2050.
This is the umbrella that ties the previous three together. Cities, roads, and infrastructure are not external to ecosystems — they are coupled with them. The questions I keep returning to are about where that coupling produces conflict (habitat loss, hazard exposure, lost services) and where it produces opportunity (compact development, accessibility, conservation gains).
Two studies are in preparation: rapid-intensification hurricane hazard and coastal exposure in the Gulf of Mexico, and built environment and socioeconomic exposure to flooding in Nigeria. A study of the combined impacts of road development and urban expansion is under review.
A parallel thread on spatial methods for public health. Earlier work mapped COVID-19 spread across multiple geographies, examined how geospatial analyses are used in research on homeless populations, and produced systematic reviews on the role of GIS in pandemic decision-making.
Ongoing work is a forecast of healthcare access in India and Bangladesh through 2050, in preparation. It projects travel time to primary, secondary, and tertiary care using a coupled urban and road growth model rather than treating the two as independent.
Across all of the themes above, the question of how we model spatial processes shows up again and again. My work in this thread brings together machine learning, statistical, and process-based approaches: hedonic and regression models for housing and accessibility, supervised and unsupervised ML for land-cover classification, statistical and AI-based flood susceptibility modeling, and large-scale process models for urban–transportation coevolution.
In preparation is a probabilistic assessment of access to hospitals and cyclone shelters that carries flood-model uncertainty through to network failure rather than reporting a single map. A global ensemble forecast of urban land-cover conversion is also in preparation, and the road network growth model is in revision.
Every result on this page rests on a chain of geospatial decisions that rarely make it into the abstract: which product to trust, what a 300-metre pixel actually means on the ground, how to partition overlapping buffers so no hectare is counted twice, and whether the answer survives a change of threshold. This thread is that craft treated as the subject rather than the plumbing.
Ongoing work supports the studies above: cross-dataset validation and buffer partitioning for the protected-area manuscript in revision, and the susceptibility modelling and uncertainty handling behind the flood-access study in preparation.