About

Thomas Gillespie is a Professor in the Department of Geography at UCLA and serves as Co-Chair of the Environmental Science and Engineering (D.Env.) Program at the Institute of the Environment and Sustainability. His research interests focus on using geographic information systems (GIS) and remote sensing data to predict patterns of species richness and rarity for plants and birds at a regional spatial scale. Gillespie's botanical research involves surveying tropical dry forests in biodiversity hotspots such as Wallacea, Sundaland, Indo-Burma, Mesoamerica, New Caledonia, and the Caribbean, providing comparative floristic and structural data to inform conservation priorities, natural resource management, and tropical ecology. His faunal research has primarily centered on tropical bird communities, with publications also addressing mammal and herpetofauna diversity, aiming to model species distributions and extinction probabilities in fragmented landscapes under various environmental change scenarios. His remote sensing research encompasses both airborne and spaceborne sensors, particularly high-resolution data from Landsat and IKONOS satellites, to develop algorithms predicting the distribution and abundance of endangered species and to assess landscape metrics in habitat fragments. Gillespie's work contributes to understanding biodiversity patterns, habitat disturbance, and conservation strategies, with a focus on tropical dry forests and California ecosystems.

Research topics

• Ecology
• Geography
• Biology
• Environmental science
• Computer Science
• Geology
• Cartography
• Climatology
• Agroforestry
• Physical geography
• Environmental resource management
• Oceanography
• Remote sensing

Selected publications

• Species distribution modeling for conservation science: new predictor layers, reproducible code, and an evaluation of California protected areas

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-01-25 · 1 citations

  preprintOpen access

  Aim Our study provides foundational resources for future SDMing: methods for generating fine-scale, equal-area predictor datasets and best-practice SDM guidelines. We also provide reproducible code to streamline their implementation. Location Southwestern North America Methods Using over 215,000 research-grade iNaturalist occurrence records for 127 species of conservation concern or scientific interest in California and surrounding area, we quantified and compared SDM performance between two predictor datasets that differ in their source of bioclimatic data, spatial resolution, and coordinate reference system: one generated using ClimateNA software (resolution = 300 x 300 m; NAD83/California Albers) and the other using existing WorldClim data (varying resolution = ~669-797 x 926 m; WGS84). We also compared two modeling algorithms (MaxEnt vs Random Forests), and two background point selection strategies (random points vs weighted points accounting for sampling effort). As an example application, we used SDM predictions to evaluate the conservation value of different protected area types within California. Results ClimateNA outperformed WorldClim for 94% of species, Random Forests outperformed MaxEnt for 87%, and random background points outperformed weighted background points for 100%. All differences were statistically significant. Together, the ClimateNA dataset, Random Forests, and random background points achieved highest performance for 86% of species. Using this best-performing set of models, we found that regional parks, county parks, state beaches, and open spaces in California were highest in multi-species suitability, while larger protected areas, such as national parks and national forests, generally exhibited surprisingly low suitability. Substantial spatial biases intrinsic to SDMing with unprojected predictor datasets (e.g., WGS84) are described, along with clear solutions using equal-area predictor datasets. Main conclusions Considerable disparity was observed among the performance of common SDM methods. This study highlights the importance of fine-scale, equal-area predictor datasets and best-practice guidelines, and demonstrates how SDMs can provide critical insights into protected area planning.

  PublisherOA PDFDOI

• Genome assembly for the Sierra Nevada Parnassian ( <i>Parnassius behrii</i> ) and a brief review of butterfly genome sizes

  Journal of Heredity · 2025-12-16

  articleOpen access

  The Sierra Nevada Parnassian (Parnassius behrii W. H. Edwards, 1870) (Lepidoptera: Papilionidae) is a high-elevation specialist butterfly endemic to the Sierra Nevada, California. We present a genome assembly for P. behrii, representing the first major genomic resource for the Parnassius phoebus species complex. The assembly consists of two haplotypes, 1.59 Gb and 1.46 Gb in length, with contig N50 values of 10.93 Mb and 11.84 Mb, scaffold N50 values of 52.56 Mb and 51.90 Mb, scaffold L50 values of 13 and 14, and BUSCO completeness scores of 98.7% and 94.4%, respectively. Both haplotypes are highly contiguous, with 31 chromosome-length scaffolds, including putative Z and W sex chromosomes. We annotated the genome with NCBI's EGAPx pipeline, integrating database and novel transcript alignment with Hidden Markov Model-based gene predictions, yielding 17 191 genes with a BUSCO score of 98.1%. RepeatMasker identified that 26.68% (424.97 Mb) of the genome consists of repetitive elements. We also assembled a mitochondrial genome for P. behrii (15 391 bp) containing 2 rRNAs, 22 unique transfer RNAs, and 13 protein-coding genes. We also reviewed 514 high-quality butterfly genomes available from the National Center for Biotechnology Information (NCBI). Parnassius species were observed to have the largest genomes, with P. behrii being the largest. This assembly provides a foundational resource for whole-genome research on P. behrii and the broader P. phoebus complex, enabling analyses of evolutionary differentiation, local adaptation, inbreeding, gene flow, speciation, and conservation practices.

  PublisherOA PDFDOI

• Biodiversity of the World

  2024-10-30 · 1 citations

  book-chapter1st authorCorresponding

  This chapter reviews recent advances in remote sensing that can be used to study biodiversity from space. In particular, this chapter examines ways to measure, model, and monitor biodiversity patterns and processes using spaceborne imagery. First, we examine advances currently being used to measure biodiversity from space. Second, we examine advances in modeling patterns of species and biodiversity. Third, we examine monitoring applications of remote sensing for the conservation of biodiversity. Finally, we identify spaceborne sensors that can be used to study biodiversity from space.

  PublisherDOI

• Assessing mangrove cover change in Madagascar (1972–2019): Widespread mangrove deforestation is slowing down

  Global Ecology and Conservation · 2024-06-01 · 5 citations

  articleOpen access

  Between the 1970s and early 2000s, Madagascar witnessed significant mangrove extent and condition declines due to deforestation and degradation. Data on mangrove extent prior to the early 1990s is scarce, contributing to uncertainties in long-term trend analyses. Recent estimates (since 2010) have reported widely varying mangrove extents from 2,000 km² to 3,400 km², indicating large uncertainties in current trends. Leveraging five decades of Landsat satellite data, this study offers a methodologically consistent approach to classify Madagascar's mangrove extent, providing a comprehensive dataset from 1972, 1989, 1999, 2009, to 2019. Our findings reveal an overall decrease of 8% from 2,935 km² in 1972 to 2,699 km² in 2019, with loss rates decelerating from 0.4% per year in the initial period to 0.2% per year post-1999. This decline primarily reflects anthropogenic pressures, notably in the northern regions. Conversely, the last decade has witnessed a 5% increase in mangrove coverage, primarily in the central and southern regions, thanks to concerted conservation and reforestation efforts, highlighting a positive shift in mangrove management and protection strategies. This study's long-term perspective is crucial for understanding the dynamics of mangrove coverage and guiding effective intervention strategies in Madagascar.

  PublisherDOI

• Remote sensing approaches to identify trees to species-level in the urban forest: A review

  Progress in Physical Geography Earth and Environment · 2024-05-07 · 8 citations

  reviewSenior author

  Most urban tree inventories depend on resource-intensive, field-based assessments, which are unevenly distributed in space and time. Recently, these inventories have been conducted using field inventories combined with airborne multispectral, hyperspectral, LiDAR, and spaceborne multispectral remote sensing. Significant advances have been made in urban tree GIS databases and remote sensing methods, which include delineating individual tree crowns, extracting tree species metrics, and employing classification techniques. Generally, remote sensing methods distinguish individual urban trees using either pixel-based or object-based methods, while image classification procedures are typically divided into parametric (e.g., regression-based classification, Bayesian, and principal component analysis) and non-parametric approaches such as machine learning (e.g., random forests support vector machines) and deep learning (e.g., convolutional neural networks). Our synthesis of the current state of science suggests sensors with the highest spatial (m), spectral (bands), and temporal (repeat time) resolutions result in the most accurate tree species identification. Combining airborne LiDAR/hyperspectral or airborne LiDAR/spaceborne high-resolution multispectral sensors yields the highest accuracy for the most diverse urban forests. An object-based non-parametric approach, like a fully convolutional neural network, scores higher in accuracy assessments than pixel-based parametric approaches. Future studies can leverage global/regional GIS field inventory databases to expand the scope of studies within and across multiple cities, utilizing LiDAR and spaceborne sensors.

  PublisherDOI

• Exploring and integrating differences in niche characteristics across regional and global scales to better understand plant invasions in Hawaiʻi

  Biological Invasions · 2024-03-23 · 2 citations

  articleSenior author
  PublisherDOI

• Monitoring native, non-native, and restored tropical dry forest with Landsat: A case study from the Hawaiian Islands

  Ecological Informatics · 2024-09-12 · 3 citations

  articleOpen accessSenior author

  Tropical dry forests are highly threatened at a global scale. Long-term monitoring of remaining stands is needed to assess forest health, efficacy of management practices, and potential impacts of climate change. Using a multi-seasonal Landsat time series, we examined Normalized Difference Vegetation Index (NDVI) patterns in native dry forest, non-native vegetation types, and dry forest restoration sites from 1999 to 2022 in the Hawaiian Islands. We calculated trends in median NDVI and robust coefficient of variation of NDVI for dry and wet seasons, and used Breaks for Additive Seasonal and Trend analysis to detect trend departures. To assess the impact of regional drying trends, NDVI trends were compared to the seasonal long-term precipitation anomaly and cumulative precipitation anomaly. We found that native dry forest was less green than non-native forest, particularly during the dry season, and that median NDVI increased in both native and non-native dry forests over the study period despite negative precipitation anomaly trends. This result differs from coarser-scale studies in Hawaii, but is supported by trends in other dry forest regions. Greening was also observed in restoration study sites, especially larger sites where native species establishment and recruitment has been reported. Non-native grassland NDVI exhibited a strong positive link to precipitation anomalies, suggesting that drier climate scenarios may exacerbate the invasive grass-wildfire cycle that threatens native dry forest. These results demonstrate that Landsat time series may be used to detect seasonal variation in dry forest plots and to support restoration site monitoring in a highly fragmented ecosystem. • Native Hawaiian dry forest vegetation has become greener in last two decades. • Increases in the vegetation index occurred despite regional drying trends. • Landsat-based results differ from previous studies at coarser spatial scales. • Greenness trends and breakpoints can facilitate monitoring of restoration sites.

  PublisherDOI

• Drought-vulnerable vegetation increases exposure of disadvantaged populations to heatwaves under global warming: A case study from Los Angeles

  Sustainable Cities and Society · 2023-02-26 · 35 citations

  article
  PublisherDOI

• Author response for "Citizen science can complement professional invasive plant surveys and improve estimates of suitable habitat"

  2023-05-23

  peer-reviewSenior author
  PublisherDOI

• High resolution lidar data shed light on inter‐island translocation of endangered bird species in the Hawaiian Islands

  Ecological Applications · 2023-05-22 · 10 citations

  articleOpen access

  Abstract Translocation, often a management solution reserved for at‐risk species, is a highly time‐sensitive intervention in the face of a rapidly changing climate. The definition of abiotic and biotic habitat requirements is essential to the selection of appropriate release sites in novel environments. However, field‐based approaches to gathering this information are often too time intensive, especially in areas of complex topography where common, coarse‐scale climate models lack essential details. We apply a fine‐scale remote sensing‐based approach to study the 'akikiki ( Oreomystis bairdi ) and 'akeke'e ( Loxops caeruleirostris ), Hawaiian honeycreepers endemic to Kaua'i that are experiencing large‐scale population declines due to warming‐induced spread of invasive disease. We use habitat suitability modeling based on fine‐scale light detection and ranging (lidar)‐derived habitat structure metrics to refine coarse climate ranges for these species in candidate translocation areas on Maui. We found that canopy density was consistently the most important variable in defining habitat suitability for the two Kaua'i species. Our models also corroborated known habitat preferences and behavioral information for these species that are essential for informing translocation. We estimated a nesting habitat that will persist under future climate conditions on east Maui of 23.43 km 2 for 'akikiki, compared to the current Kaua'i range of 13.09 km 2 . In contrast, the novel nesting range for 'akeke'e in east Maui was smaller than its current range on Kaua'i (26.29 vs. 38.48 km 2 , respectively). We were also able to assess detailed novel competitive interactions at a fine scale using models of three endemic Maui species of conservation concern: 'ākohekohe ( Palmeria dolei ), Maui 'alauahio ( Paroreomyza montana ), and kiwikiu ( Pseudonestor xanthophrys ). Weighted overlap areas between the species from both islands were moderate (&lt;12 km 2 ), and correlations between Maui and Kaua'i bird habitat were generally low, indicating limited potential for competition. Results indicate that translocation to east Maui could be a viable option for 'akikiki but would be more uncertain for 'akeke'e. Our novel multifaceted approach allows for the timely analysis of both climate and vegetation structure at informative scales for the effective selection of appropriate translocation sites for at‐risk species.

  PublisherOA PDFDOI

Frequent coauthors

• Stephanie Pau

  University of California, Berkeley

  15 shared

• Chunyu Dong

  Sun Yat-sen University

  12 shared

• Geoffrey A. Fricker

  California Polytechnic State University

  11 shared

• Gunnar Keppel

  University of South Australia

  11 shared

• Sassan Saatchi

  Jet Propulsion Laboratory

  11 shared

• Glen M. MacDonald

  University of California, Los Angeles

  10 shared

• Philippe Birnbaum

  9 shared

• Diane E. Pataki

  Arizona State University

  9 shared

Labs

• Thomas Gillespie's LabPI

Education

• Ph.D., Environmental Science and Engineering

  University of California, Los Angeles

  2008

• M.S., Environmental Science and Engineering

  University of California, Los Angeles

  2004

• B.S., Environmental Science

  University of California, Los Angeles

  2002