Joseph McGlinchy

Building materials and their spatial distribution play a significant role in determining the outcomes of human-caused and natural disasters in urban and peri-urban areas. However, building-level data on building and roofing materials are scarce. Here, we explore the feasibility and performance of a Convolutional Neural Network (CNN) model using spectrally transformed high-resolution multispectral imagery to map roofprints (i.e., classifying and delineating roofing materials at the building footprint-level) in Washington, District of Columbia (D.C.) and Denver, CO, United States. To generate consistent training data, we integrate geospatial vector data of individual building footprints with real estate industry-derived building-level roofing material data to create labeled image data from Planet SuperDove imagery. We compare the CNN classifier to a pixel-based machine learning (ML) model to demonstrate the capability of our roofprints mapping approach. With F1-scores ranging from 0.56 to 0.95 for the most common roof material classes, the CNN model outperformed the pixel-based ML classifier by 15% and 17% in Washington, D.C., and Denver, respectively. Results demonstrate within-domain robustness for the studied metropolitan areas, which are characterized by differing building densities, roof morphologies, and material patterns. While cross-region transferability was not evaluated, our findings provide a controlled comparison of pixel-based and context-aware approaches for rooftop material mapping and highlight the importance of hierarchical representations that integrate spectral information with roof texture, edge characteristics, spatial arrangement, and neighborhood context for improving classification performance. Accurately mapping building materials has the potential to advance urban planning and environmental policies, including assessments of heat exposure, energy demand, as well as hazard risk and community resilience.

The transition from soil-mantled to bedrock-dominated landscapes is important to hydrology, ecology, and landscape evolution. Where classical theory predicts that this transition abruptly occurs once erosion rates exceed soil production ones, actual landscapes show a much more gradual transition. The Rampart Range, CO, is an unglaciated, forested mountain range with uniform lithology that spans this transition, providing an opportunity to interrogate how topography, soil cover, and forest structure coevolve. Using drone imagery and airborne lidar, we show the utility of object-based methods that segment the lidar topography into individual outcrops. Lower elevation hillslopes are rockier (~14% rock) than their higher elevation counterparts (~2% rock) due to late Cenozoic incision of a regional, high elevation, low relief surface. After accounting for the erosional control on rockiness, we show that bedrock exposure also depends on topographic aspect. Equator-facing hillslopes are rockier (~20% rock at low elevations; ~3% rock at high elevations) than their more heavily forested, pole-facing counterparts (~11% rock at low elevations; ~1% rock at high elevations). To simulate these patterns, hillslope evolution models need “humped” soil production functions that can generate rocky outcrops within a soil mantle. When combined with depth-dependent soil creep, such models readily produce large gradients in the height and frequency of bedrock outcrops in response to aspect-dependent weathering. Given the correspondence between higher forest biomass (e.g., increased tree density, canopy height) with lower bedrock exposure on pole-facing hillslopes, we argue that forest dynamics may be an underappreciated regulator of hillslope morphology via root zone weathering.

We report an innovative method of extracting water distribution network (WDN) historical repair location data from images of paper repair data maps, to provide usable geo‐referenced digitally formatted data. For most water utilities, repair location data typically fall into two eras: pre‐ and post‐GIS, approximately corresponding to pre‐ and post‐2000. Automated conversion of pre‐2000 paper maps to a geo‐referenced digital format provides additional data to clarify trends in pipe repair causative factors, such as material defects, corrosive or creeping soils, and traffic. We applied the methodology to more than 3,000 maps of the Los Angeles Department of Water and Power WDN, thereby extending the record of repairs backward from 2000 to 1975, almost doubling the number of repair records. The methodology's value, when using the resulting data for analysis, lies in the following: (a) large volumes of hard copy data can now be acquired in an automated manner, saving significant time and effort, (b) specific repair locations are accurately captured, resulting in (c) more reliable, confident, analyses, and results, (d) ongoing problem areas, due to corrosive or creeping soils for example, can be more specifically understood.

Macroecology research seeks to understand ecological phenomena with causes and consequences that accumulate, interact, and emerge across scales spanning several orders of magnitude. Broad‐extent, fine‐grain information (i.e., high spatial resolution data over large areas) is needed to adequately capture these cross‐scale phenomena, but these data have historically been costly to acquire and process. Unoccupied aerial systems (UAS or drones carrying a sensor payload) and the National Ecological Observatory Network (NEON) make the broad‐extent, fine‐grain observational domain more accessible to researchers by lowering costs and reducing the need for highly specialized equipment. Integration of these tools can further democratize macroecological research, as their strengths and weaknesses are complementary. However, using these tools for macroecology can be challenging because mental models are lacking, thus requiring large up‐front investments in time, energy, and creativity to become proficient. This challenge inspired a working group of UAS‐using academic ecologists, NEON professionals, imaging scientists, remote sensing specialists, and aeronautical engineers at the 2019 NEON Science Summit in Boulder, Colorado, to synthesize current knowledge on how to use UAS with NEON in a mental model for an intended audience of ecologists new to these tools. Specifically, we provide (1) a collection of core principles for collecting high‐quality UAS data for NEON integration and (2) a case study illustrating a sample workflow for processing UAS data into meaningful ecological information and integrating it with NEON data collected on the ground—with the Terrestrial Observation System—and remotely—from the Airborne Observation Platform. With this mental model, we advance the democratization of macroecology by making a key observational domain—the broad‐extent, fine‐grain domain—more accessible via NEON/UAS integration.

It is a critical time to reflect on the National Ecological Observatory Network (NEON) science to date as well as envision what research can be done right now with NEON (and other) data and what training is needed to enable a diverse user community. NEON became fully operational in May 2019 and has pivoted from planning and construction to operation and maintenance. In this overview, the history of and foundational thinking around NEON are discussed. A framework of open science is described with a discussion of how NEON can be situated as part of a larger data constellation—across existing networks and different suites of ecological measurements and sensors. Next, a synthesis of early NEON science, based on >100 existing publications, funded proposal efforts, and emergent science at the very first NEON Science Summit (hosted by Earth Lab at the University of Colorado Boulder in October 2019) is provided. Key questions that the ecology community will address with NEON data in the next 10 yr are outlined, from understanding drivers of biodiversity across spatial and temporal scales to defining complex feedback mechanisms in human–environmental systems. Last, the essential elements needed to engage and support a diverse and inclusive NEON user community are highlighted: training resources and tools that are openly available, funding for broad community engagement initiatives, and a mechanism to share and advertise those opportunities. NEON users require both the skills to work with NEON data and the ecological or environmental science domain knowledge to understand and interpret them. This paper synthesizes early directions in the community’s use of NEON data, and opportunities for the next 10 yr of NEON operations in emergent science themes, open science best practices, education and training, and community building.

Harnessing the fire data revolution, i.e., the abundance of information from satellites, government records, social media, and human health sources, now requires complex and challenging data integration approaches. Defining fire events is key to that effort. In order to understand the spatial and temporal characteristics of fire, or the classic fire regime concept, we need to critically define fire events from remote sensing data. Events, fundamentally a geographic concept with delineated spatial and temporal boundaries around a specific phenomenon that is homogenous in some property, are key to understanding fire regimes and more importantly how they are changing. Here, we describe Fire Events Delineation (FIRED), an event-delineation algorithm, that has been used to derive fire events (N = 51,871) from the MODIS MCD64 burned area product for the coterminous US (CONUS) from January 2001 to May 2019. The optimized spatial and temporal parameters to cluster burned area pixels into events were an 11-day window and a 5-pixel (2315 m) distance, when optimized against 13,741 wildfire perimeters in the CONUS from the Monitoring Trends in Burn Severity record. The linear relationship between the size of individual FIRED and Monitoring Trends in Burn Severity (MTBS) events for the CONUS was strong (R² = 0.92 for all events). Importantly, this algorithm is open-source and flexible, allowing the end user to modify the spatio-temporal threshold or even the underlying algorithm approach as they see fit. We expect the optimized criteria to vary across regions, based on regional distributions of fire event size and rate of spread. We describe the derived metrics provided in a new national database and how they can be used to better understand US fire regimes. The open, flexible FIRED algorithm could be utilized to derive events in any satellite product. We hope that this open science effort will help catalyze a community-driven, data-integration effort (termed OneFire) to build a more complete picture of fire.

Vessel-based sonar systems that focus on the water column provide valuable information on the distribution of underwater marine organisms, but such data are expensive to collect and limited in their spatiotemporal coverage. Satellite data, however, are widely available across large regions and provide information on surface ocean conditions. If satellite data can be linked to subsurface sonar measurements, it may be possible to predict marine life over broader spatial regions with higher frequency using satellite observations. Here, we use random forest models to evaluate the potential for predicting a sonar-derived proxy for subsurface biomass as a function of satellite imagery in the California Current Ecosystem. We find that satellite data may be useful for prediction under some circumstances, but across a range of sonar frequencies and depths, overall predictive performance was low. Performance in spatial interpolation tasks exceeded performance in spatial and temporal extrapolation, suggesting that this approach is not yet reliable for forecasting or spatial extrapolation. We conclude with some potential limitations and extensions of this work.

EarthPy makes commonly performed spatial data exploration tasks easier for scientist by building upon functions in the widely used packages: Rasterio and GeoPandas. EarthPy is designed for users who are new to Python and spatial data with a focus on scientific data. When a user is working with spatial data for research, there are a suite of data exploration activities that are often performed including: