Browsing by Subject "Remote sensing"

This dissertation investigates the limnology of the southwestern Peruvian Amazon centered on the Madre de Dios department by examining first the geomorphology and then the ecology and biogeochemistry of the region's fluvial systems.

Madre de Dios, Peru is world renowned for its prolific biodiversity and its location within the Andes biodiversity hotspot. It is also a site of study regarding the development of the Fitzcarrald Arch and that feature's geomorphological importance as the drainage center for the headwaters of the Madeira River - the Amazon's largest tributary and as well as its role as a physical divider of genetic evolution in the Amazon. Though each of these has been studied by a variety of prominent researchers, the ability to investigate all the aspects of this unique region is hampered by the lack of a regional geomorphological map. This study aims to fill that gap by using remote sensing techniques on digital elevation models, satellite imagery and soil, geology and geoecological maps already in publication to create a geomorphological map. The resulting map contains ten distinct landform types that exemplify the dominance of fluvial processes in shaping this landscape. The river terraces of the Madre de Dios River are delineated in their entirety as well as the various dissected relief units and previously undefined units. The demarcation of the boundaries of these geomorphic units will provide invaluable assistance to the selection of field sites by future researchers as well as insights into the origin of the high biodiversity indices of this region and aid in planning for biodiversity conservation.

Secondly this study examines 25 tropical floodplain lakes along 300 km of the Manu River within the Manu National Park in the Madre de Dios department. Alternative stable state and regime shifts in shallow lakes typically have been examined in lakes in temperate and boreal regions and within anthropogenically disturbed basins but have rarely been studied in tropical or in undisturbed regions. In contrast this study focuses on a tropical region of virtually no human disturbance and evaluates the effects of hydrological variability on ecosystem structure and dynamics. Using satellite imagery a 23 yr timeline of ecological regime shifts in Amazon oxbow lakes or "cochas" is reconstructed. The study shows that almost 25% of the river's floodplain lakes experience periodic abrupt vegetative changes with an average 3.4% existing in an alternative stable state in any given year. State changes typically occur from a stable phytoplankton-dominated state to a short lived, <3 yr, floating macrophytic state and often occur independent of regional flooding. We theorize that multiple dynamics, both internal and external, drive vegetative regime shifts in the Manu but insufficient data yet exists in this remote region to identify the key processes.

To complete the investigation of tropical limnology the third study compares and contrasts the nutrient-productivity ration of floodplain and non-floodplain lakes globally and regionally. For over 70 years a strong positive relationship between sestonic chlorophyll-a (Chl-a) and total phosphorus (TP) has been established with phosphorus generally viewed as the most limiting factor to productivity. Most of these studies, however, have focused on northern, temperate regions where the lakes are typically postglacial, isolated and fed by small streams. Relatively little work has been done on floodplain lakes which are semi or permanently connected to the river. This study examines the relationship between nutrients and productivity in floodplain lakes globally through an extensive literature synthesis. Values for total phosphorus, total nitrogen and chlorophyll-a were collected for 523 floodplain lakes, represented by 288 data points while 551 data points were collected for 5444 non-floodplain lakes. Analysis revealed that globally, floodplain lakes do not show any significant difference in the total phosphorus/chlorophyll-a relationship from that found in non-floodplain lakes but significant differences are seen between tropical and temperate lakes. We propose that the term `floodplain' lake should serve as purely a geographical descriptor and that it is lacking as an ecological indicator. Instead factors such as precipitation seasonality, hydrological connectivity and regional flooding regimes are better indicators of high or low productivity in floodplain lakes.

Pinniped species undergo a life history, unique among marine mammals, that includes discrete periods of occupancy on land or ice within a predominantly marine existence. This makes many pinniped species valuable sentinels of marine ecosystem health and models of marine mammal physiology and behavior. Pinniped research has often progressed hand-in-hand with advances at the technological frontiers of wildlife biology, and drones represent a leap forward in the long-established field of aerial photography, heralding opportunities for data collection and integration at new scales of biological importance. The following chapters employ and evaluate recent and emerging methods of wildlife surveillance that are uniquely enabled and facilitated by drone methods, in applied research and management campaigns with near-polar pinniped species. These methods represent advancements in abundance estimation and distribution modeling of pinniped populations that are dynamically shifting amid climate change, fishing pressure, and recovery from historical depletion.Conventional methods of counting animals from aerial imagery—typically visual interpretation by human analysts—can be time-consuming and limits the practical use of this data type. Deep learning methods of computer vision can ease this burden when applied to drone imagery, but are not yet characterized for practical and generalized use. To this end, I used a common implementation of deep learning for object detection in imagery to train and test models on a variety of datasets describing breeding populations of gray seals (Halichoerus grypus) in the northwest Atlantic Ocean (Chapter 2). I compare standardized performance metrics of models trained and tested on different combinations of datasets, demonstrating that model performance varies depending on both training and testing data choices. We find that models require careful validation to estimate error rates, and that they can be effectively deployed to aid, but not replace, conventional human visual interpretation of novel datasets for gray seal detection, location, age-classification and abundance estimation. Spatial analysis and species distribution modeling can use fine-scale drone-derived data to describe local species–habitat relationships at the scale of individual animals. I applied structure-from-motion methods to a survey of three pinniped species, pacific harbor seals (Phoca vitulina richardii), northern fur seals (Callorhinus ursinus), and Steller sea lions (Eumetopias jubatus), in adjacent non-breeding haul-outs to compare occupancy and habitat selection (Chapter 3). I describe and compare fitted occupancy models of pacific harbor seals and northern fur seals, finding that conspecific attraction is a key driver of habitat selection for each species, and that each species exhibits distinct topographic preferences. These findings illustrate both opportunities and limitations of spatial analysis at the scale of individual pinnipeds. Ease of deployment and rapid data collection make drones a powerful tool for monitoring populations of interest over time, while animal locations, revealed in high-resolution imagery, and contextual habitat products can reveal spatial relationships that persist beyond local contexts. I designed and carried out a campaign of drone surveillance over coastal habitats near Palmer Station, Antarctica, in the austral summer of 2020 to assess the seasonal abundance and habitat use of Antarctic fur seals (Arctocephalus gazella) in the Palmer Archipelago and adjacent regions (Chapter 4). I modeled abundance as a function of date, with and without additional terms to capture variance by site, and used these models to estimate peak abundance near Palmer Station in the 2020 summer season. These findings leverage the spatial and temporal advantages of drone methods to estimate species phenology, distribution and abundance. Together, these chapters describe emerging applications of drone technology that can advance pinniped research and management into new scales of analytical efficiency and ecological interpretation. These studies describe methods that have been proven in concept, but not yet standardized for practical deployment, and their findings reveal new ecological insights, opportunities for methodological advancement, and current limitations of drone methods for the study of pinnipeds in high-latitude environments.

Exposure to higher PM2.5 can lead to increased risks of mortality; however, the spatial concentrations of PM2.5 are not well characterized, even in megacities, due to the sparseness of regulatory air quality monitoring (AQM) stations. This motivates novel low-cost methods to estimate ground-level PM2.5 at a fine spatial resolution so that PM2.5 exposure in epidemiological research can be better quantified and local PM2.5 hotspots at a community-level can be automatically identified. Wireless low-cost particulate matter sensor network (WLPMSN) is among these novel low-cost methods that transform air quality monitoring by providing PM information at finer spatial and temporal resolutions; however, large-scale WLPMSN calibration and maintenance remain a challenge because the manual labor involved in initial calibration by collocation and routine recalibration is intensive, the transferability of the calibration models determined from initial collocation to new deployment sites is questionable, as calibration factors typically vary with urban heterogeneity of operating conditions and aerosol optical properties, and the stability of low-cost sensors can drift or degrade over time. This work presents a simultaneous Gaussian Process regression (GPR) and simple linear regression pipeline to calibrate and monitor dense WLPMSNs on the fly by leveraging all available reference monitors across an area without resorting to pre-deployment collocation calibration. We evaluated our method for Delhi, where the PM2.5 measurements of all 22 regulatory reference and 10 low-cost nodes were available for 59 days from January 1, 2018 to March 31, 2018 (PM2.5 averaged 138 ± 31 μg m-3 among 22 reference stations), using a leave-one-out cross-validation (CV) over the 22 reference nodes. We showed that our approach can achieve an overall 30 % prediction error (RMSE: 33 μg m-3) at a 24 h scale and is robust as underscored by the small variability in the GPR model parameters and in the model-produced calibration factors for the low-cost nodes among the 22-fold CV. Of the 22 reference stations, high-quality predictions were observed for those stations whose PM2.5 means were close to the Delhi-wide mean (i.e., 138 ± 31 μg m-3) and relatively poor predictions for those nodes whose means differed substantially from the Delhi-wide mean (particularly on the lower end). We also observed washed-out local variability in PM2.5 across the 10 low-cost sites after calibration using our approach, which stands in marked contrast to the true wide variability across the reference sites. These observations revealed that our proposed technique (and more generally the geostatistical technique) requires high spatial homogeneity in the pollutant concentrations to be fully effective. We further demonstrated that our algorithm performance is insensitive to training window size as the mean prediction error rate and the standard error of the mean (SEM) for the 22 reference stations remained consistent at ~30 % and ~3–4 % when an increment of 2 days’ data were included in the model training. The markedly low requirement of our algorithm for training data enables the models to always be nearly most updated in the field, thus realizing the algorithm’s full potential for dynamically surveilling large-scale WLPMSNs by detecting malfunctioning low-cost nodes and tracking the drift with little latency. Our algorithm presented similarly stable 26–34 % mean prediction errors and ~3–7 % SEMs over the sampling period when pre-trained on the current week’s data and predicting 1 week ahead, therefore suitable for online calibration. Simulations conducted using our algorithm suggest that in addition to dynamic calibration, the algorithm can also be adapted for automated monitoring of large-scale WLPMSNs. In these simulations, the algorithm was able to differentiate malfunctioning low-cost nodes (due to either hardware failure or under heavy influence of local sources) within a network by identifying aberrant model-generated calibration factors (i.e., slopes close to zero and intercepts close to the Delhi-wide mean of true PM2.5). The algorithm was also able to track the drift of low-cost nodes accurately within 4 % error for all the simulation scenarios. The simulation results showed that ~20 reference stations are optimum for our solution in Delhi and confirmed that low-cost nodes can extend the spatial precision of a network by decreasing the extent of pure interpolation among only reference stations. Our solution has substantial implications in reducing the amount of manual labor for the calibration and surveillance of extensive WLPMSNs, improving the spatial comprehensiveness of PM evaluation, and enhancing the accuracy of WLPMSNs. Satellite-based ground-level PM2.5 modeling is another such low-cost method. Satellite-retrieved aerosol products are in particular widely used to estimate the spatial distribution of ground-level PM2.5. However, these aerosol products can be subject to large uncertainties due to many approximations and assumptions made in multiple stages of their retrieval algorithms. Therefore, estimating ground-level PM2.5 directly from satellites (e.g., satellite images) by skipping the intermediate step of aerosol retrieval can potentially yield lower errors because it avoids retrieval error propagating into PM2.5 estimation and is desirable compared to current ground-level PM2.5 retrieval methods. Additionally, the spatial resolutions of estimated PM2.5 are usually constrained by those of the aerosol products and are currently largely at a comparatively coarse 1 km or greater resolution. Such coarse spatial resolutions are unable to support scientific studies that thrive on highly spatially-resolved PM2.5. These limitations have motivated us to devise a computer vision algorithm for estimating ground-level PM2.5 at a high spatiotemporal resolution by directly processing the global-coverage, daily, near real-time updated, 3 m/pixel resolution, three-band micro-satellite imagery of spatial coverages significantly smaller than 1 × 1 km (e.g., 200 × 200 m) available from Planet Labs. In this study, we employ a deep convolutional neural network (CNN) to process the imagery by extracting image features that characterize the day-to-day dynamic changes in the built environment and more importantly the image colors related to aerosol loading, and a random forest (RF) regressor to estimate PM2.5 based on the extracted image features along with meteorological conditions. We conducted the experiment on 35 AQM stations in Beijing over a period of ~3 years from 2017 to 2019. We trained our CNN-RF model on 10,400 available daily images of the AQM stations labeled with the corresponding ground-truth PM2.5 and evaluated the model performance on 2622 holdout images. Our model estimates ground-level PM2.5 accurately at a 200 m spatial resolution with a mean absolute error (MAE) as low as 10.1 μg m-3 (equivalent to 23.7% error) and Pearson and Spearman r scores up to 0.91 and 0.90, respectively. Our trained CNN from Beijing is then applied to Shanghai, a similar urban area. By quickly retraining only RF but not CNN on the new Shanghai imagery dataset, our model estimates Shanghai 10 AQM stations’ PM2.5 accurately with a MAE and both Pearson and Spearman r scores of 7.7 μg m-3 (18.6% error) and 0.85, respectively. The finest 200 m spatial resolution of ground-level PM2.5 estimates from our model in this study is higher than the vast majority of existing state-of-the-art satellite-based PM2.5 retrieval methods. And our 200 m model’s estimation performance is also at the high end of these state-of-the-art methods. Our results highlight the potential of augmenting existing spatial predictors of PM2.5 with high-resolution satellite imagery to enhance the spatial resolution of PM2.5 estimates for a wide range of applications, including pollutant emission hotspot determination, PM2.5 exposure assessment, and fusion of satellite remote sensing and low-cost air quality sensor network information. We later, however, found out that this CNN-RF sequential model, despite effectively capturing spatial variations, yields higher average PM2.5 prediction errors than its RF part alone using only meteorological conditions, most likely the result of CNN-RF sequential model being unable to fully use the information in satellite images in the presence of meteorological conditions. To break this bottleneck in PM2.5 prediction performance, we reformulated the previous CNN-RF sequential model into a RF-CNN joint model that adopts a residual learning ideology that forces the CNN part to most effectively exploit the information in satellite images that is only “orthogonal” to meteorology. The RF-CNN joint model achieved low normalized root mean square error for PM2.5 of within ~31% and normalized mean absolute error of within ~19% on the holdout samples in both Delhi and Beijing, better than the performances of both the CNN-RF sequential model and the RF part alone using only meteorological conditions. To date, few studies have used their simulated ambient PM2.5 to detect hotspots. Furthermore, even the hotspots studied in these very limited works are all “global” hotspots that have the absolute highest PM2.5 levels in the whole study region. Little is known about “local” hotspots that have the highest PM2.5 only relative to their neighbors at fine-scale community levels, even though the disparities in outdoor PM2.5 exposures and their associated risks of mortality between populations in local hotspots and coolspots within the same communities can be rather large. These limitations motivated us to concatenate a local contrast normalization (LCN) algorithm at the end of the RF-CNN joint model to automatically reveal local PM2.5 hotspots from the estimated PM2.5 maps. The RF-CNN-LCN pipeline reasonably predicts urban PM2.5 local hotspots and coolspots by capturing both the main intra-urban spatial trends in PM2.5 and the local variations in PM2.5 with urban landscape, with local hotspots relating to compact urban spatial structures while coolspots being open areas and green spaces. Based on 20 sampled representative neighborhoods in Delhi, our pipeline revealed that on average a significant 9.2 ± 4.0 μg m-3 long-term PM2.5 exposure difference existed between the local hotspots and coolspots within the same community, with Indian Gandhi International Airport area having the steepest increase of 20.3 μg m-3 from the coolest spot (the residential area immediately outside the airport) to the hottest spot (airport runway). This work provides a possible means of automatically identifying local PM2.5 hotspots at 300 m in heavily polluted megacities. It highlights the potential existence of substantial health inequalities in long-term outdoor PM2.5 exposures within even the same local neighborhoods between local hotspots and coolspots. Apart from posing serious health risks, deposition of dust and anthropogenic particulate matter (PM) on solar photovoltaics (PVs), known as soiling, can diminish solar energy production appreciably. As of 2018, the global cumulative PV capacity crossed 500 GW, of which at least 3–4% was estimated to be lost due to soiling, equivalent to ~4–6 billion USD revenue losses. In the context of a projected ~16-fold increase of global solar capacity to 8.5 TW by 2050, soiling will play an increasingly more important part in estimating and forecasting the performance and economics of solar PV installations. However, reliable soiling information is currently lacking because the existing soiling monitoring systems are expensive. This work presents a low-cost remote sensing algorithm that estimates utility-scale solar farms’ daily solar energy loss due to PV soiling by directly processing the daily (near real-time updated), 3 m/pixel resolution, and global coverage micro-satellite surface reflectance (SR) analytic product from the commercial satellite company Planet. We demonstrate that our approach can estimate daily soiling loss for a solar farm in Pune, India over three years that on average caused ~5.4% reduction in solar energy production. We further estimated that around 437 MWh solar energy was lost in total over the 3 years, equivalent to ~11799 USD, at this solar farm. Our approach’s average soiling estimation matches perfectly with the ~5.3% soiling loss reported by a previous published model for this solar farm site. Compared to other state-of-the-art PV soiling modeling approaches, the proposed unsupervised approach has the benefit of estimating PV soiling at a precisely solar farm level (as in contrast to coarse regional modeling for only large spatial grids in which a solar farm resides) and at an unprecedently high temporal resolution (i.e., 1 day) without resorting to solar farms’ proprietary solar energy generation data or knowledge about the specific components of deposited PM or these species’ dry deposition flux and other physical properties. Our approach allows solar farm owners to keep close track of the intensity of soiling at their sites and perform panel cleaning operations more strategically rather than based on a fixed schedule.

In high elevated regions, ground-based precipitation measurements are scarce or non-existing. Remote sensing estimates are prone to underestimation due to a lack of sensitivity to light precipitation and ground-clutter contamination of radar signals by steep terrain. The latter creates a blind zone extending up a few kilometers above the ground for the vertical profiling radars onboard the Global Precipitation Measurement mission (GPM) core satellite that serves as a reference standard to unify precipitation measurements from research and operational satellites. Ground radars which are not affected by the blind zone, show an increase in effective reflectivity factor (ERF) from the melting layer below the isotherm of zero towards the surface. The blind zone is critical as rainfall characteristics at the ground results from the vertical evolution of the rain drop-size distributions (RDSDs). Explicit modeling of the rainfall microphysics from the top of the blind zone, if located below the zero-degree isotherm, towards the surface can improve rainfall estimation at the ground, as demonstrated in the Southern Appalachian Mountains. However, if the top of the blind zone is located above the zero-degree isotherm, ice and mix-phase hydrometeors coexist at the top boundary condition precluding explicit modeling because the microphysics of these hydrometeors is yet an active research topic. This work has three parts. The first part is devoted to characterizing precipitating systems and the blind zone in the Central Andes of Peru. Twenty years of TRMM and GPM radar measurements reveal that stratiform precipitation is the most frequent type in the Central Andes. Long-lasting precipitating systems (LDPSs) having a stratiform structure with embedded convection determine the interannual variability of the diurnal cycle of precipitation. Moisture availability at high elevations in the Central Andes is scarce. Moisture sources that sustain LDPSs are located in the adjacent eastern foothills of the Andes (Western Amazon basin) that result from enhanced moisture convergence and low and mid-levels due to the interplay of the South American low-level jet and cold air incursions. In the Central Andes of Peru, the top of the GPM blind zone is 46 % of the time above the zero-degree isotherm located at 5000 masl missing liquid precipitation, and a comparison with ground disdrometer data shows that GPM DPR does not capture the variability of the rain drop size distribution. These findings show that rainfall in the Central Andes depends on the dynamics of atmospheric circulation and microphysics at multiple spatial and temporal scales. These scales cannot be modeled explicitly by one single model nor validated with current observations. However, progress in that direction can be made by using different models to simulate different weather regimes to elucidate the role of different microphysics and dynamic processes. Observations needed to validate the models are lacking in the Andes but available in the Southern Appalachian Mountains (SAM). So, the modeling part of this work is focused on the SAM, and the knowledge gained by modeling the SAM region can be used to improve the representation of dynamical and microphysical processes in numerical models to be used in the Andes.Changes in the cloud drop size distribution ultimately capture the interplay of microphysics, dynamics, and thermodynamics in clouds. The chain of microphysical mechanisms from aerosol activation to cloud drop size distribution (CDSD) evolution during cloud development to the raindrop size distribution (RDSD) dynamics spans a scale range of four orders of magnitude from fractions of micrometers to millimeters. It can be modeled explicitly in cloud parcel and rain-shaft models (for liquid hydrometeors), and it is described using different parameterizations with varying degrees of complexity in Numerical Weather Prediction (NWP) models. However, in the Weather Research and Forecast (WRF) model, parameterized microphysics often fail to capture the diurnal cycle and spatial distribution of precipitation, rainfall intensity, and duration depending on the weather regime, regional topography, and regional aerosol characteristics. Besides, the representation of the dynamical processes affecting the aerosol activation and the CDSDs evolution during cloud formation and of the cloud-droplet-raindrop continuum in precipitating clouds is lacking in NWP models. In the second part of this work, a two-moment bulk microphysics scheme in WRF is modified to add an aerosol activation spectrum from in-situ measurement and compared to the default activation spectrum to characterize the impact of aerosol activation in the onset of precipitation in different weather conditions. WRF simulations show that using the in-situ aerosol activation spectrum yields higher cloud droplet number concentrations (CDNC) than the default WRF aerosol activation spectrum, with smaller cloud droplets and delayed onset of rainfall under weak synoptic forcing conditions. For large-scale systems with strong and sustained moisture convergence at low levels (frontal and tropical systems), mechanically forced rainfall efficiency is enhanced despite high CDNC, there is no delay in the onset of precipitation, and the impact of ACPI on the spatial and temporal variability of rainfall is negligible (significant) at onset (hours later) consistent with rainfall observations. The simulated cloud vertical structure from CDNC indicates that convective development is more intense in the inner SAM region than in the adjacent plains. In the inner region, valley-ridge circulations organize the spatial patterns of cloudiness under weak synoptic forcing conditions. The formation of early afternoon low-level clouds over the ridges in the summertime reflects the high sensitivity of cloud mixing ratios and cloud droplet concentrations to aerosol activation properties. In the third part of this work, the dynamical and microphysical processes not resolved by WRF are modeled using a large eddy simulation coupled with a spectrum bin microphysics that permits aerosol replenishment to characterize the heterogeneity of cloud microstructure associated with cloud circulations and entrainment in non-precipitating cumulus clouds. Furthermore, the sensitivity of cloud variables to different vertical profiles of aerosol loading is tested and validated with aircraft observations. The coupled model simulated a convective case under weak synoptic conditions, and the spatial variability of the LWC, CDNC, and CDSD was characterized. Simulations show large differences between the region around the updraft and regions of entrainment located at cloud edges and preferred locations. The interplay of environmental wind and cloud circulations explains these locations. More and larger droplets towards the location of the updraft and towards but before the cloud's top result in large LWC in the upper half of the cloud. The opposite in regions of entrainment, having small droplets in minor concentrations resulting in low LWC. Different initial vertical profiles of aerosol concentrations result in significantly different values of CDNC; as larger the aerosol loading, the larger the CDNC. The updraft transport aerosols and moisture up to the cloud altitude from near the ground elevation. In the updraft, a sharp CDSD with a small standard deviation has small values of relative dispersion. In regions of entrainment, the CDSD has smaller droplets in minor concentrations due to the evaporation of droplets. Its shape resembles the distribution of interstitial aerosols with a large standard deviation resulting in large values of relative dispersion and smaller values of LWC. Aircraft measurements show agreement with the simulated CDNC and relative dispersion for the simulation with an initial vertical profile of aerosol concentration that decays exponentially with height and has a scale height of 500m. The significant impact of aerosol loading in the CDNC affects the cloud optical thickness (COT). The COT in the cloud mature stage for a scale height of 2000m is approximately 1.75 times the COT for a scale height of 500m.

Wetlands are invaluable ecosystems that provide critical ecological services and contribute to global climate regulation as large carbon sinks. The Southeastern (SE) United States (US), especially its coastal regions, is rich in wetlands. Compared to their northern counterparts, wetlands in the SE US are more vulnerable to climate change, facing threats from global warming, changes in rainfall patterns, and sea-level rise. These climate change-driven disturbances can profoundly alter the hydrological processes within these wetlands, resulting in wetland degradation and a substantial reduction in their carbon storage capacity. As a result, masses of greenhouse gases like carbon dioxide (CO2) and methane (CH4) are likely to be released into the atmosphere, potentially shifting wetlands from net carbon sinks to carbon sources and further exacerbating global warming. Despite the importance of wetlands, there remains a lack of good understanding regarding the spatiotemporal patterns of wetland degradation and carbon fluxes, as well as the underlying factors and mechanisms driving these changes, especially on regional scales such as the SE US.

This dissertation aims to advance our understanding of climate change-driven wetland degradation and carbon emissions in the Southeastern United States by answering two key questions: 1) How has climate change impacted wetlands on a regional scale? and 2) What is the feedback of wetlands to the climate through their carbon emissions? To address these questions, I established a framework for monitoring regional wetland degradation, investigated the primary regulators and processes of wetland degradation, developed high-resolution and long-term wetland carbon flux datasets, and identified key environmental factors and mechanisms controlling the spatiotemporal patterns of wetland carbon fluxes in the SE US.

Specifically, by analyzing fine-scale, long-term remotely sensed Normalized Difference Vegetation Index (NDVI) data from Landsat, a new framework was developed to detect the spatial and temporal patterns of wetland degradation. This framework was applied at the Alligator River National Wildlife Refuge in coastal North Carolina, uncovering spatiotemporal patterns of coastal wetland degradation between 1995 and 2019. Most degradation occurred within two kilometers of the shoreline over the past five years (2015−2019), primarily due to accelerated sea-level rise. To further identify the key hydrological factors driving different types of coastal wetland degradation detected by the framework, random forest classification models were employed. The analysis underscored the varying importance of specific hydrological drivers depending on the wetland type, with woody wetlands being vulnerable to saltwater intrusion and emergent herbaceous wetlands to inundation and droughts. Distances to canals played a key role in determining the status of woody wetlands after degradation.

Additionally, to investigate how wetland carbon fluxes evolved with climate change, we integrated FLUXNET/AmeriFlux data, machine learning methods, and the process-based biogeochemical model, Forest-DNDC, to pinpoint the crucial factors and processes influencing wetland carbon fluxes across the SE US. Variable importance analysis revealed that temperature and water table levels collectively regulate methane emissions from subtropical freshwater wetlands, different from high-latitude peatlands where CH4 emissions are primarily sensitive to temperature, and tropical wetlands, where CH4 emissions are predominantly sensitive to water table levels. Moreover, we constructed the first-ever high-spatial-resolution (~1 km × 1 km) and long-term (1982-2010) monthly gridded regional wetland CH4 flux product for the SE US, estimating annual methane emissions from subtropical freshwater wetlands in the region at 5.07 ± 0.12 Tg CH4 yr-1. We also detected a significant increasing trend in annual wetland CH4 emissions, with an approximate increase of 0.006 Tg CH4 per year. Our ongoing study is exploring critical environmental variables and mechanisms governing wetland CO2 fluxes while also developing a regional-scale, long-term, and high-spatial-resolution CO2 flux data product for the SE US subtropical wetlands.

In conclusion, this dissertation offers valuable insights into the spatiotemporal patterns and primary drivers and mechanisms of wetland degradation and carbon fluxes in the Southeastern United States. Frameworks and methodologies developed in this dissertation—including a remote sensing-hydrologic model integrated scheme, a machine learning-hydrological model coupled method, and the upscaling of site-level wetland carbon fluxes using eddy covariance-based flux tower measurements, remote sensing data, and machine learning—are readily applicable to wetland regions worldwide. The knowledge gained and the datasets developed in this dissertation not only enrich our comprehension of wetlands’ role in climate feedback mechanisms but also inform strategic wetland conservation efforts. As wetlands continue to face the threats of climate change, the findings from this dissertation are essential for guiding wetland management efforts and for leveraging wetlands as potent nature-based solutions to mitigate climate change.

Recent recognition models for remote sensing data (e.g., infrared cameras) are based upon machine learning models such as deep neural networks (DNNs) and typically require large quantities of labeled training data. However, many applications in remote sensing suffer from limited quantities of training data. To address this problem, we explore style transfer methods to leverage preexisting large and diverse datasets in more data-abundant sensing modalities (e.g., color imagery) so that they can be used to train recognition models on data-scarce target tasks. We first explore the potential efficacy of style transfer in the context of Buried Threat Detection using ground penetrating radar data. Based upon this work we found that simple pre-processing of downward-looking GPR makes it suitable to train machine learning models that are effective at recognizing threats in hand-held GPR. We then explore cross modal style transfer (CMST) for color-to-infrared stylization. We evaluate six contemporary CMST methods on four publicly-available IR datasets, the first comparison of its kind. Our analysis reveals that existing data-driven methods are either too simplistic or introduce significant artifacts into the imagery. To overcome these limitations, we propose meta-learning style transfer (MLST), which learns a stylization by composing and tuning well-behaved analytic functions. We find that MLST leads to more complex stylizations without introducing significant image artifacts and achieves the best overall performance on our benchmark datasets.

Climate change is driving vegetation community shifts in coastal regions of the world, where low topographic relief makes ecosystems particularly vulnerable to sea level rise, salinization, storm surge, and other effects of global climate change. Salinization has clear effects on vegetation, as few plant species can survive in brackish water, and these shifts in vegetation lead to declines in biomass carbon stocks, as well as significant changes in habitat structure and biodiversity. The rate and extent of these impacts on other wetland ecosystem properties and function is far less certain. This dissertation investigates the ecosystem consequences of saltwater intrusion in coastal wetlands, from shifting vegetation at the landscape scale, to soil biogeochemistry and wetland carbon cycling.Coastal plant communities globally are highly vulnerable to future sea-level rise and storm damage, but the extent to which these habitats are affected by the various environmental perturbations associated with chronic salinization remains unclear. In 2016, a series of vegetation plots across the Albemarle-Pamlico Peninsula that had been surveyed 7-13 years earlier were revisited in order to measure changes in tree basal area and community composition over time. I found reduced tree basal area in plots at lower elevations and with higher current soil salt content, while these factors explained only a small fraction of the measured changes in tree community composition. While tree basal area increased in the majority of plots, I measured declines in basal area in multiple sites with high soil salt content or low elevation. This decadal comparison provides convincing evidence that increases in soil salinity and saturation can explain recent changes in tree biomass, and potential shifts in community composition in low-elevation sites along the North Carolina coast. In Chapter 3, I quantified land and land cover change in the Alligator River National Wildlife Refuge (ARNWR), North Carolina’s largest coastal wildlife preserve, from 1985 to 2019 using classification algorithms applied to a long-term record of satellite imagery. Despite ARNWR’s protected status, and in the absence of any active forest management, 32 % (31,600 hectares) of the refuge area has changed land cover classification during the study period. A total of 1151 hectares of land was lost to the sea and ~19,300 hectares of coastal forest habitat were converted to shrubland or marsh habitat. As much as 11 % of all forested cover in the refuge transitioned to ghost forest, a unique land cover class that is characterized by standing dead trees and fallen tree trunks. This is the first attempt to map and quantity coastal ghost forests using remote sensing. These unprecedented rates of deforestation and land cover change due to climate change may become the status quo for coastal regions worldwide, with implications for wetland function, wildlife habitat and global carbon cycling. Salinization of freshwater wetlands is a symptom of climate change induced sea level rise. The ecosystem consequences of increasing salinity are poorly constrained and highly variable within prior observational and experimental studies. Chapter 4 presents the results of the first attempt to conduct a salinization experiment in a coastal forested wetland. Over four years, marine salts were applied to experimental plots several times annually with the goal of raising soil salinity to brackish levels while soil porewater in control plots remained fresh. Each year I measured aboveground and belowground vegetation biomass along with soil carbon stocks and fluxes. Despite adding more than 1.5 kg of salt per m2 to our experimental plots over four years, the ecosystem responses to salt treatments were subtle and varied over the multi-year experiment. In the final year of the experiment, soil respiration was suppressed, and bulk and aromatic soil carbon became less soluble as a result of salt treatments. The more stable carbon pools—soil organic carbon and vegetation associated carbon—remained unaffected by the salt treatment. This experiment demonstrates substantial ecosystem resistance to low dose salinity manipulations. The inconsistent soil carbon responses to experimental salinization I observed in the field led me to question how differences in soil pH and base saturation might alter the impacts of salinity of soil microbial activity. To test this, I performed a salt addition experiment on two series of wetland soils with independently manipulated salt concentrations and solution pH to tease apart the effect of these seawater components on soil carbon cycling (Chapter 5). Microbial respiration and dissolved organic carbon solubility were depressed by marine salts in both soils, while pH manipulation alone had no effect. Salinity treatments had a far greater effect on soil pH than did our intentional pH manipulation and there was a strong interaction between salt treatments and soil type that affected the magnitude of soil carbon responses. Site soils varied significantly in pH and base saturation, suggesting that the interaction between salinity and edaphic factors is mediating soil carbon processes. The degree of salinization and the effective pH shift following seawater exposure may vary widely based on initial soil conditions and may explain much of the variation in reports of salt effects on soil carbon dynamics. I suggest that these edaphic factors may help explain the heretofore inconsistent reports of carbon cycle responses to experimental salinization reported in the literature to date.

Rising temperatures with increased drought pose three challenges for management of future biodiversity. First, are the species expected to be vulnerable concentrated in specific regions and habitats? Second, are the impacts of drought and warming varying across regions? Third, could recent advances in remote sensing techniques help us in monitoring the impacts in real-time? This dissertation is an effort to address the above questions in the three chapters.

First, I used foliar chemistry as a proxy for drought vulnerability. I used soil and moisture gradients to quantify habitat variation that could be critical for alleviating drought. I used a large dataset of forest plots covering the eastern united states to understand how community weighted mean foliar nitrogen and phosphorus vary across climate and soil gradients. I exploited trends in these variables between species, traits, and habitats to evaluate sensitivity. Critical to our approach is the capacity to jointly model trait responses. Our data showed that nutrient demanding species strongly respond to environmental gradients. I identified a wide range of sites across low to high latitudes threatened by drought. The sensitivity of species to high temperatures is largely explained by soil variations. Drought vulnerability of nutrient and moisture demanding species could be amplified depending on local soil and moisture gradients. Although local soil moisture may dampen drought-induced stress for species with large leaves and high water use, nutrient demanding species remain vulnerable in wet regions during droughts. Phosphorus demanding species adapted to dry sites are drought resilient compared to communities in wet sites. This research is consistent with the studies that supports declining nutrient demanding species with increasing temperature and decreasing moisture. I also detected strong soil effects on shaping community weighted traits across a large geographical and environmental range. Our data showed that soil effects on controlling foliar traits strongly vary across different climates. The findings are critical for conservations and maintaining the biodiversity.

Next, I used space-borne remotely sensed vegetation indices to monitor the process of leaf development across climate gradients and ecoregions in the southeastern United States. A hierarchical state-space Bayesian model was developed to quantify how air temperature, drought severity, and canopy thermal stress contribute to changes in leaf opening from mountainous to coastal regions. I synthesized daily field climate data with daily remotely sensed vegetation indices and canopy surface temperature during spring green-up season. The study was focused on observation of leaf phenology at 59 sites in the southeast United States between 2001 to 2012. Our results suggest strong interaction effects between ecosystem properties and climate variables across ecoregions. The findings showed that despite the much faster spring green-up in the mountains, coastal forests express a larger sensitivity to inter-annual anomaly in temperature than mountain sites. In spite of the decreasing trend in sensitivity to warming with temperature in all regions, there is an ecosystem interaction: Deciduous-dominated forests are less sensitive to warming than are those with few deciduous trees, possibly due to the presence of developed leaves in evergreen species throughout the season. The findings revealed mountainous forests are more susceptible to intensifying drought and moisture deficit, while coastal areas are relatively resilient. I found that increasing canopy thermal stress, defined as canopy-air temperature difference, slows the leaf-development following a dry year, accelerates it after a wet year.

Finally, I demonstrate how space-borne canopy “thermal stress”, i.e. surface-air temperature difference, could be used as a surrogate for drought-induced stress to estimate forest transpiration. Using physics-based relationships that accommodates uncertainties, I showed how changes in canopy water flux may be reflected in surface energy balance and in remotely-sensed thermal stress. Validating with field measurements of canopy transpiration in the southeastern US, I quantified sensitivity of transpiration to thermal stress in a range of atmospheric and climate conditions. I found that a 1 mm change in daily transpiration may cause 3 to 4 °C of thermal stress, depending on site conditions. The cooling effect is large when solar radiation is high or wind speed is low. The effect has the highest control on water-use during warm and dry seasons, when monitoring drought is essential. I applied our model to available satellite and metrological data to detect patterns of drought. Using only air and surface temperatures, I predicted anomaly in water-use across the contiguous United States over the past 15 years, and then compared with anomaly in soil water content and conventional drought indices. Our simple model showed a reliable accuracy in compare to the state-of-the-art general circulation models. The technique can be used in varying time-scales to monitor surface water-use and drought in large scales.

Coastal ecosystems provide a wide range of services including protecting the mainland from the destructive effects of storms, nutrient cycling, water filtration, nurseries for fish and crustaceans, and carbon sequestration. These zones are threatened by human impacts and climate change through more frequent intense storms and sea level rise with a projected increase of up to 16 mm/yr for the last two decades of the 21st century. However, it is not fully understood what mechanisms control the formation and degradation of these landforms, and determine their resilience to environmental change. In this work, I highlight the role of various physical characteristics and environmental parameters that contribute to the formation and stability of coastal environments.

First, I develop and use remote sensing analyses to quantitatively characterize coastal dune eco-topographic patterns by simultaneously identifying the spatial distribution of topographic elevation and vegetation biomass in order to understand the coupled dynamics of vegetation and coastal dunes. LiDAR-derived leaf area index and hyperspectral-derived normalized difference vegetation index patterns yield vegetation distributions at the whole-system scale which are in agreement with each other and with field observations. LiDAR-derived concurrent quantifications of biomass and topography show that plants more favorably develop on the landward side of the foredune crest and that the foredune crestline marks the position of an ecotone, which is interpreted as the result of a sheltering effect sharply changing local environmental conditions. The findings reveal that the position of the foredune crestline is a chief ecomorphodynamic feature resulting from the two-way interaction between vegetation and topography.

Next, to shed light on the vertical depositional dynamics of salt marshes in response to sea level rise, I investigate the hypothesis that competing effects between biomass production and aeration/decomposition determine an approximately spatially constant contribution of soil organic matter (SOM) to total accretion. I use concurrent observations of SOM and decomposition rates from marshes in North Carolina. The results are coherent with the notion that SOM does not significantly vary in space and suggest that this may be the result of an at least partial compensation of opposing trends in biomass productivity and decomposed organic matter. The analyses show that deeper soil layers are characterized by lower decomposition rates and higher stabilization factors than shallower layers, likely because of differences in inundation duration. However, overall, decomposition processes are sufficiently rapid that the labile material in the fresh biomass is completely decomposed before it can be buried and stabilized. The findings point to the importance of the fraction of initially refractory material and of the stabilization processes in determining the final distribution of SOM within the soil column.

Finally, I develop a process-based model to evaluate the relative role of watershed, estuarine, and oceanic controls on salt marsh depositional/erosional dynamics and define how these factors interact to determine salt marsh resilience to environmental change at the estuary scale. The results show that under some circumstances, vertical depositional dynamics can lead to transitions between salt marsh and tidal flat equilibrium states that occur much more rapidly than marsh/tidal flat boundary erosion or accretion could. Additionally, the analyses reveal that river inputs affect the existence and extent of marsh/tidal flat equilibria by both modulating exchanges with the ocean (by partially “filling” the basin) and by providing suspended sediment.

Forward Looking Infrared (FLIR) cameras have recently been studied as a sensing modality for use in landmine detection systems. FLIR-based detection systems benefit from larger standoff distances and faster rates of advance than other sensing modalities, but they also present significant challenges for detection algorithm design. FLIR video typically yields multiple looks at each object in the scene, each from a different camera perspective. As a result each object in the scene appears in multiple video frames, and each time at a different shape and size. This presents questions about how best to utilize such information. Evidence in the literature suggests such multi-look information can be exploited to improve detection performance but, to date, there has been no controlled investigation of multi-look information in detection. Any results are further confounded because no precise definition exists for what constitutes multi-look information. This thesis addresses these problems by developing a precise mathematical definition of "a look", and how to quantify the multi-look content of video data. Controlled experiments are conducted to assess the impact of multi-look information on FLIR detection using several popular detection algorithms. Based on these results two novel video processing techniques are presented, the plan-view framework and the FLRX algorithm, to better exploit multi-look information. The results show that multi-look information can have a positive or negative impact on detection performance depending on how it is used. The results also show that the novel algorithms presented here are effective techniques for analyzing video and exploiting any multi-look information to improve detection performance.

During my PhD journey, I have lived at the intersection of a previous military career, leadership as an elected official, and a student passionate about robotics and protecting our beautiful coastal ecosystem. As a non-traditional student, Duke University has presented me with experiences I could not have imagined. With the Duke Marine Robotics and Remote Sensing (MaRRS) Lab drones, I have had the opportunity to survey the mass nesting of thousands of olive ridley sea turtles in Costa Rica, hundreds of gray seals in Massachusetts, endangered right whales off the coast of Florida, dozens of World War I shipwrecks in Maryland, Etruscan and Roman archeological sites in Italy, and hundreds of seals in the Bering Sea. And there have been many more multi-domain surveys of our glorious coastal ecosystem in Carteret County. There have been more than our fair share of challenges during this time frame to include preparing and responding to Hurricanes Florence, Dorian, and Isaias, plus the COVID-19 pandemic. These events took a toll on many fronts, but also presented leadership opportunities. With our drones, we have been able to survey before and after storms, and we’re watching barrier islands move at centimeter scale. The increasing effects of climate change are very personal for those of us living in eastern North Carolina, but in the MaRRS Lab we are well postured with our robotics to air-, sea-, and ground-truth these effects. Perhaps most importantly, the knowledge I gained during my PhD program informed my policy positions during my tenure as the Mayor of the Town of Beaufort, NC. I am very proud of the progress that we made to include a massive clean-up of our waterways following Hurricane Florence, a Harbor Management Ordinance to better manage our waterways, expanded municipal jurisdiction to further manage our ecosystem, unprecedented repairs of infrastructure that were neglected for decades and have negatively affected our water quality, investment in the community, and a five-year budgeting plan to provide greater stability for Beaufort.

This dissertation is a summation of some of the work performed during my Duke PhD experience. In Chapter 1, I describe the evolution of autonomous drones, define distinct generations of this technology, and articulate the negative impacts of a regulatory system that is stifling critical research. For Chapter 2, I discuss the lexicon, taxonomy, and ontology of small autonomous drones, the critical importance of situational awareness, and a framework of considerations and best practices for those interested in pursuing autonomous mobile robots to enhance their research. With Chapters 1 and 2 as a foundation, I next highlight my expansion to the marine domain for water quality research with autonomous surface vessels (Chapter 3) and multi-disciplinary archeological drone surveys in Vulci, Italy (Chapter 4). Finally in Chapter 5, I address scientific research that informed policy successes during my time as a mayor and PhD student. What a great journey!