The magnetosphere, ionosphere and thermosphere (MIT) act as a coherently integrated system (geospace), driven in part by solar influences and characterized by variability and complexity. Among the most important and yet uncertain aspects of the geospace system is energy and momentum coupling between regions, which is, in part, accomplished by the transfer of charged particles from the magnetosphere to the ionosphere in a process known as particle precipitation, and in the opposite direction by ion outflow. Both processes are inherently multiscale and manifest the variabilities and complexities of the geospace system. Despite the importance of the transfer of particles, existing models are increasingly ill-equipped to provide the specification necessary for the growing demand for geospace now- and forecasts. Due to recent trends in the availability of data, we now face an exciting opportunity to progress particle transfer in geospace through the intersection of traditional approaches and state-of-the-art data-driven sciences. We reveal novel particle transfer models utilizing machine learning (ML), present results from the models, and provide an evaluation of their capabilities including comparisons with observations and the current ’state-of-the-art’ models (e.g., OVATION Prime for particle precipitation and the Gamera-Ionosphere Polar Wind Model for ion outflow). We detail the data wrangling required to utilize the available geospace observations to make progress on the long-standing challenge of particle transfer and place specific emphasis on the discovery possible when ML models are appropriate and robustly interrogated in the context of physical understanding. Our presentation helps illustrate the trends in the application of data science in space science.

Magmatic structures are well-preserved in a 201.5 Ma diabase sill (PA, USA, equivalent to the Palisades sill) formed as part of the Central Atlantic Magmatic Province during rifting of Pangea. The sill was emplaced at ~6 km depth and tilted ~20° NNW by post-magmatic fault movement. Detailed mush structures are exposed in a dimension stone quarry with walls cut parallel and perpendicular to the strike of the sill. Light gray, plagioclase-rich layers (PLR) a few mm thick contain up to 75% modal plag and are underlain by more pyroxene-rich layers with larger orthopyx antecrysts up to 1 mm length. PLR are sub-parallel to sill margins, have dm-m lateral dimensions, and spaced 0.33 m apart on average. Magma replenishments < 1m thick cross-cut plag-pyx layers at low angles and have basal load-cast-like structures. Since mafic replenishments have PLR at their tops and similar thickness to PLR spacing, we interpret all PLR as having formed by emplacement of small-volume magma pulses bearing ~30% larger pyx and smaller plag antecrysts. This model is similar to Petford and Mirhadizadeh (R Soc Open Sci, 2017) for the Basement sill, Antarctica. Upward migration of mafic melts in pipe-like channels (cm to dm wide) disrupted plag-pyx layers to form dm-scale graben-like and slump-like structures that resemble sediment liquefaction. Channelized flow late in sill development may have been enhanced by seismicity (Davis et al., JVGR, 2007). Diabase micro-structures are similar to published experimental results and numerical simulations of flow and shear-thinning in particle-rich slurries (e.g., Cimarelli et al., G3, 2011; Ishibashi, JVGR, 2009; Deubelbeiss et al., G3, 2011). These include layers such as the plag-pyx couplets and orientations of euhedral plag around pyx phenocrysts. Plag long-axis orientations and tiling indicators in the PLR have strike-parallel and strike-normal components in vertical and plan views consistent with flow alignment in the plane perpendicular to the stress gradient. Plag chemical zoning patterns, limited deformation, and long-axis orientations parallel to inclined layer margins also indicate magmatic flow rather than compaction. Mineral x-ray maps are used to derive initial crystal fraction and aspect ratios for modeling relative viscosity and explore compositional aspects of layer development.

Laboratory reanalysis of paleomagnetic samples from two sets of superposed ~ 4.5 m.y. old lava flows (N=64) collected on Kauai (Hawaii) in the 1960s, when combined with more recently published results from two additional flow sequences (N=59) on the island, yields a mean paleomagnetic pole that differs significantly from the spin axis. The new laboratory work was motivated by the fact the 60% of the published flow mean directions on the old collection derived from measurements of natural remanent magnetization without any demagnetization treatment. In our reanalysis, we subjected all specimens to stepwise alternating field (AF) demagnetization (13 steps; 90 mT peak AF) and modern line-fit analysis. As has been found in many restudies of this sort on young, well-behaved basaltic lavas, modern demagnetization technique produced little change in the flow-mean directions (an average change of just over 2° in this case.) To minimize the bias introduced when rapidly erupted flow sequences sample a single geomagnetic field direction, we thinned the combined data by grouping superposed flows with similar paleomagnetic directions. The procedure requires choosing a minimum angle between directions that will be considered independent samples of the ancient geomagnetic field. Previous studies have commonly used a 6° threshold. We found, however, that much larger values (10° to 16°) were needed to eliminate significant serial correlation (measured in several ways) in the thinned directional sequence. Our conclusion is that the 123 lava flows in this study provide only 33 independent estimate of the ancient field direction. The mean normal polarity paleomagnetic pole resulting from the analysis (Lon=076.8, Lat=82.2, N=13) is almost perfectly antiparallel (179°) to the mean reverse polarity pole (Lon=258.4, Lat=-83.2, N=20). The grand mean of 33 normal and reverse polarity data is Lon=077.3 Lat=82.8 k=31.8 α95=4.5°. The VGP dispersion (without correction for within-site dispersion, of order 0.5° for these very well-behaved paleomagnetic recorders) is 14.3° about the mean pole and 15.9° about the spin axis, comparable to that found in other analyses of Hawaiian paleomagnetic data.

The newest atmospheric climate model at GFDL, AM4, succeeded at significantly reducing the toa radiative flux biases as compared to CERES observations. Despite a relatively low top-of-atmosphere sensitivity to uniform warming of SSTs (Cess warming experiments), the corresponding coupled climate model, CM4, has high transient and equilibrium climate sensitivities. We will present a systematic picture of the modeled clouds across a hierarchy of model configurations which utilize this atmospheric model. This hierarchy includes the CFMIP Aquaplanet and AMIP experiments, fully coupled model experiments (using GFDL’s CM4 model) as well as additional AMIP-like experiments with particular SST patterns. This demonstrates the large range of sensitivities that are possible from a single atmospheric climate model. Looking at the global mean radiative feedbacks across the different model configurations as well as in the context of CMIP5 and CMIP6 models will allow us to assess to what extent the cloud feedbacks in the idealized experiments relate to the fully coupled experiments and to observed clouds.

Monitoring of in-situ, stress-induced, seismic velocity change provides an increasingly important contribution to the study of the earthquake nucleation process. Continuous Active-Source Seismic Monitoring (CASSM) with borehole sources and sensors has proven to be a very effective tool to monitor seismic velocity and to identify its temporal variations at depth. Since June 2017, we have been operating a crosswell CASSM field experiment at the San Andreas Fault Observatory at Depth (SAFOD) where a previous CASSM experiment identified the two seismic velocity reductions approximately 10 and 2 hours before micro-earthquakes. The ultimate goal of our experiment is to continuously monitor tectonic stress for the San Andreas Fault near seismogenic depth. Our active-source experiment makes use of two boreholes drilled at the SAFOD project site. A piezoelectric source and a three-component accelerometer have been installed in the SAFOD pilot and main holes, respectively, at about 1 km depth. A seismic pulse is generated by the piezoelectric source four times per second, and waveforms are recorded with a 48 kHz sample rate, with recordings summed for 1 to 10 minutes to capture seismic velocity changes at a high-temporal resolution. Since deployment in June 2017, and as of July, 2019, local seismicity has not been above our current threshold of detection. However, we have identified a velocity reduction at the SAFOD site (0.5 microsecond change in crosswell travel time, measured in a coda window) possibly induced by dynamic stress changes from the distant 6 July 2019 M 7.1 Ridgecrest earthquake, California. We will characterize and report the co-seismic change and post-seismic recovery process for this remotely triggered velocity change. We will also report on the overall status of this unique CASSM experiment.

Wetlands are the largest environmental sources of methane, and interannual changes in wetland methane fluxes explain most of the variability in the global flux. Despite their importance, global wetland maps, a key component of methane models, are inaccurate for at least three reasons: (1) Their temporal variability is poorly suited for static maps; (2) Optical remote sensing cannot penetrate foliage, making water hard to identify; and (3) satellites cannot resolve their fine-scale features. Furthermore, small, unmapped water bodies may emit methane disproportionately to their size due their shallow depths inhibiting bacterial oxidation from the water column and their large perimeter: volume ratios, which introduce the potential for organic matter input and plant-mediated fluxes from shorelines. However, in boreal regions, there is conflicting evidence on the effects of water body size on methane and carbon dioxide fluxes. Here, we measure methane emissions in lakes and wetlands in an Arctic-Boreal delta and compare to open water and vegetated area with the goal of improving methane emission estimates in this region. We expect small, shallow, and vegetated wetlands to produce more methane than those bordering deeper lakes. To test this hypothesis, we map wetlands in the Peace-Athabasca Delta, a 5,000 km2 inland delta in northern Alberta, Canada containing abundant open and vegetated wetlands. We use airborne remote sensing from three sources: (1) High-resolution (<5 cm pixel) unmanned aerial vehicle (UAV) imagery, (2) Coincident L-band synthetic aperture radar (SAR) from NASA’s UAVSAR airborne imaging system, and (3) 2017 AirSWOT Ka-band interferometric SAR with color-infrared imagery. With a wavelength of 23.8 cm, UAVSAR L-band returns are ideal for mapping vegetated wetlands due to double-bounce backscatter between vegetation and the water surface. Combining two field campaigns of flux chamber gas sampling from over twenty lakes, walked shoreline surveys, and over 70 thousand UAV photos, we present a collection of wetland maps and a methodology for efficiently mapping them from UAV. We then upscale methane and carbon dioxide emissions to the scale of the delta and compare to existing estimates. These results will help improve greenhouse gas emission estimates for boreal zone wetlands.

Nitrous oxide (N2O) is a potent greenhouse gas and stratospheric ozone-depleting substance. More than half of anthropogenic N2O emissions result from agricultural activities. A broad objective of this on-farm research in eastern Maryland was to investigate whether drainage water management, which reduces nitrate export, would increase greenhouse gas emissions, but here we focus upon comparing chamber and tower measurements of N2O fluxes from a single field. Chamber methods usually suffer from poor spatial and temporal resolution. Automating chambers using in situ fast response analyzers improves temporal but not spatial resolution. Tower-based micrometeorological methods improve both temporal and spatial resolution, but require a high-frequency, high-sensitivity laser instrument. We compared auto-chamber and micrometeorological gradient methods for N2O flux measurement during a period early in the 2019 corn-growing season. A 3 m tall tower was deployed to allow for near-continuous gradient flux measurements using an Aerodyne Quantum Cascade Laser. Four Eosense closed dynamic automated chambers (eocAC) and a multiplexer (eosMX) were installed near the tower and connected to a Picarro G2308 gas analyzer. Both methods captured strong pulses of N2O fluxes after rainfall and fertilization events, demonstrating these major drivers of large emissions. Fluxes from the two methods were linearly correlated (R2 = 0.54), but the slope (1.29 ± 0.08) and y-intercept (48.3 ± 19.2) indicate that the chambers generally estimated higher fluxes. Aggregating over the measurement period, the automated chamber estimate was 2.5 kg N2O-N/ha in 19 days, whereas the tower-based gradient estimate was 1.3 kg N2O-N/ha in 19 days. The tower footprint includes some area (4%) covered by ditches and could extend beyond the field at times, but this is unlikely the only explanation. The small number of chambers may have sampled an area of above average flux, or there could be unknown measurement bias or interpolation error in one or both methods. To our knowledge, this is the first such methodological comparison of N2O fluxes since these sensitive, fast response instruments have become available, and our results demonstrate that additional work is needed to gain more confidence in reported fluxes by either method.

The cycling of carbon within the earth system is intrinsically linked with major nutrients, notably nitrogen and phosphorus, due to the tendency of these elements to limit the productivity of terrestrial ecosystems. To understand the response of the carbon cycle to global change pressures, models must integrate Carbon-Nitrogen-Phosphorus cycles. Whilst such models exist, to-date these have focused on natural and semi-natural ecosystems. Agriculture results in significant modification to natural biogeochemical cycling, and currently represents approximately 37% of land-use. With the projected increase in global food demand over the 21st century, this area is expected to increase. It is therefore critical to understand and simulate biogeochemical cycling in both natural and agricultural systems, and the transition between these, to estimate ecosystem response to environmental change. In this study we present an integrated C-N-P model including both natural and agricultural temperate ecosystems. The N14CP model has been developed to include representation of both arable and grassland systems, with the inclusion of agricultural management practices such as fertilizer application, crop removal, grazing and yield estimation. The model has been tested both spatially and temporally using a range of long-term experimental sites across Northern-Europe, and applied at both local and national scales. We use the model to assess impacts of land-use change and management on long-term nutrient cycling, and discuss the implications of this for sustainable agriculture and ecosystem functioning.

Phosphorus is a critical nutrient in sustaining food production. In agricultural systems, application of P fertilizers has significantly increased since the green revolution to become common practice globally, contributing to increased productivity. However, excess use of P fertilizer does not only pose a cost to farmers, but costs for society in the form of water quality problems and environmental degradation. Furthermore, rock phosphates from which these fertilizers are derived are a finite resource, which brings into question the long-term sustainability of this resource and the food production it supports. Soils play a critical role in hosting the P cycle, and organic forms of P (monoesters, diesters) often represent a significant portion of soil P stocks, that are so often overlooked. The mineralization of organic P by phosphatase enzymes is recognized as a key mechanism for converting organic to inorganic forms, which can then be potentially used for P uptake by plants. However, quantification of their contribution still remains a significant challenge. In order to sustainably meet growing food production demands over the next century and reduce the impacts of P fertilizers on waterways, there is a need to understand the extent that soil organic P is available, or can be made available for plants. Here, we present recent findings from a soil-plant biogeochemical model of integrated carbon-nitrogen-phosphorus cycling in agricultural environments. Comparison of observational yield data taken from various long-term experimental sites with model simulations indicate a gap in current scientific understanding of P sources. Whilst yields in the experiments are maintained under low P addition conditions, the model indicates yield declines due to exhaustion of available P resources. We use the model to explore the missing links: potential P inputs, processes and pathways. Finally, we discuss the need for additional empirical evidence to support understanding of organic P cycling, and development of models to include these processes to inform future land management and ensure long-term food security and sustainable water resources.

In recent years there has been growing interest in Nature-Based Solutions (NBS) as a way of reducing flood risk. The premise is that large numbers of small-scale features distributed across the landscape can provide the same level of protection as large-scale traditional flood defences. The in-channel Leaky Barrier (LB) is an example NBS feature that has been widely implemented. LBs cause flow to back up and temporarily move onto flood plains during high rainfall. However, there is considerable resistance to their use at the scales required to impact significantly on flood risk due to a lack of quantitative data on their effectiveness. Notably, their hydraulics is poorly understood. This motivated the research reported here. Physical modelling of simple LBs was performed in a flume to improve understanding of fundamental behaviour and provide data for mathematical models. The features consist of one or more horizontal sheets spanning the channel with a gap underneath allowing water to pass unimpeded in low flows. For intermediate depths the feature acts as a sluice gate, while for high depths water passes under, over and, in the case of more than one sheet, through the feature. Experiments were carried out in steady-state and flood-wave conditions. A finite volume model of the flume and features was developed using a 1D Godunov-type scheme. Riemann solvers are used to find mass and momentum fluxes between cells. The LB is treated as an internal boundary condition using a combined weir and sluice gate equation. Good agreement with experimental results was obtained for steady-state configurations. Validation is limited for the flood-wave experiments, but general behaviour was captured by the numerical model for these cases. This approach could fill a key evidence gap by answering questions about the optimal leakiness of LBs, the limits to their usefulness, and how combining them may or may not cause synchronisation problems when the effect of multiple features is aggregated.

Taylor Glacier, located in the McMurdo Dry Valleys of Antarctica has piqued curiosity since the first observations in 1903. Episodic release of iron-rich brine at or near the glacier terminus rapidly oxidizes, forming a visually striking red stain on the ice and glacier forefield called ‘Blood Falls’. The triggering mechanism behind these releases is unknown. The recent history of brine releases have been well documented since the 1993-94 summer season. To better understand the frequency and extent of brine releases over a longer time period we compile a detailed history of observations of the Taylor terminus from photographs, journals, field reports, oral histories, and published papers prior to the onset of more frequent monitoring in the 1990s. We developed a confidence assessment framework for our interpretation of the presence/absence of brine icing deposits. Results show that of the 30 summer seasons between 1903-1904 and 1993-1994 with interpretable observations, 21 seasons (70%) show evidence of brine flow events, and 9 seasons show no evidence of brine flow. At least two of these brine flow events are newly reported by our study. Concurrent observations of the glacier terminus over the same period showed a localized advance and collapse of a small portion of the southern terminus. We demonstrate a framework to fuse multiple data types and qualitatively assess the confidence level of our interpretations that could be applied to similar investigations of environmental history. We encourage other researchers to explore and contribute to the growing collection of open access historical archives.

As high-resolution geospatiotemporal data sets from observatory networks, remote sensing platforms, and computational Earth systems increase in abundance, fidelity, and richness, machine learning approaches that can fully utilize increasingly powerful parallel computing resources are becoming essential for analysis and exploration of such data sets. We explore one such approach, applying a state-of-the-art distributed memory parallel implementation of Support Vector Machine (SVM) classification to large remote-sensing data sets. We have used MODIS 8-day surface reflectance (MOD09A1) and land surface temperature (MOD11A2) for classifying wildfires over Alaska and California. Monitoring Trends in Burn Severity (MTBS) burn perimeter data was used to set boundaries of burned and unburned areas for our two-class problem. MTBS covers years from 1984-2019, recording only fires over 1000 acres or greater in the western United States. We seek to find a parallel computing solution (using the PermonSVM solver, described below) to accurately classify wildfires and find smaller unrecorded wildfires. An initial assessment for wildfire classification over interior Alaska shows that PermonSVM has an accuracy of 96% and over 5000 false positives (i.e., fires unrecorded in MTBS). Next steps include mapping larger regions over Alaska and California and understanding the tradeoffs of scalability and accuracy. The parallel tool we employ is PermonSVM, which is built on top of the widely-used open source toolkit PETSc, the Portable, Extensible Toolkit for Scientific Computation. Recent developments in PETSc have focused on supporting cutting-edge GPU-based high-performance computing (HPC) architectures, and these can be easily leveraged in PermonSVM by using appropriate GPU-enabled matrix and vector types in PETSc. We achieve significant GPU speedup for the SVM calculations on the Summit supercomputer at Oak Ridge National Laboratory – currently one of the best available “at scale” proxies for upcoming exascale-class supercomputers – and are actively working to further improve computational efficiency on Summit as well as on prototype exascale node architectures.

Melt inclusion analysis used to constrain and compare parental melt compositions of the nakhlite and chassignite martian meteorites.

High pressures, along with thermal processes and irradiation, have a measurable effect on carbonaceous compounds which are found throughout the solar system. High pressure environments, such as those generated by impacts, occur frequently during the evolution of the solar system, and the effects of pressure on the carbonaceous materials present can influence the subsequent chemistry of the body. In situ high pressure synchrotron source Fourier Transform Infrared (FTIR) spectroscopy coupled with computational models has been used to directly study the effects of pressure on carbonaceous materials. Our work using these techniques investigates the specific case of structural sugars, where ribose and deoxyribose have differing responses to high pressures. These particular carbonaceous materials play a key role in the prebiotic chemistry of the early Earth as constituents of the bioinformational molecules ribonucleic acid (RNA) and deoxyribonucleic acid (DNA) respectively. Ribose is substantially less stable than deoxyribose at pressures exceeding 14 GPa and shows less recovery on decompression. Our results imply that the modest impacts experienced throughout the solar system could substantially alter the carbonaceous payload of many bodies, with consequences for any prebiotic chemistry.

This study was conducted to assess precipitation particle properties, including ice-nucleating particle (INP) concentration (L^-1 Air), in West Texas, where the semi-arid climate prevails and typically <40 inches of rainfall coincides per year. Further, the West Texas region is dominated by deep convective clouds, where INPs play a crucial role in hailstorm and thunderstorm processes (e.g., Li et al., 2017; Rosenfeld et al., 2008). In this study, we looked into major precipitation events observed throughout the year in 2018 and 2019 in the Texas Panhandle area. More specifically, to characterize immersion freezing efficiency (T > -25 degree C) of our precipitation samples, we used a cold-stage instrument called West Texas Cryogenic Refrigerator Applied to freezing Test (WT-CRAFT) system (Hiranuma et al., 2019). Additionally, a disdrometer is used to look into the relationship between INP concentration, intensity and size of precipitation particles. An indigenously developed Internet of Things (IoT) air quality sensors were also used to compare ambient air quality (i.e., particulate matter concentrations) and meteorological conditions to the measured INP concentrations. Overall, the study’s preliminary results show a reasonable correlation between INP concentration and precipitation properties (i.e., intensity). We also find a high ice nucleation efficiency at higher temperatures (i.e., T > -15 degree C), which can be attributed to the biological INPs from local agricultural sources. The results also suggest that INPs play an important role in the precipitation particle size. These findings may be important in artificially varying the severity of the precipitation by varying the INP concentration in the West Texas region.

From extracting nutrients to releasing energy, biological metabolism plays an integral role in determining evolutionary patterns of organisms through geologic time. A previous study depicted a positive relationship between metabolic rate and extinction probability for Mollusca within the Neogene period. We hypothesized that this relationship extends to other metazoan phyla during the Cenozoic Era. Using specific respiration rates measured from living organisms and body size data for fossil taxa, we estimated metabolic rates of animals across different phyla: Arthropoda, Brachiopoda, Echinodermata, and Mollusca. This analysis was performed at the class level by using the classes with the most data available to represent each phylum: Malacostraca, Ostracoda, Cirripedia, Rhynchonellata, Echinoidea, Bivalvia, Cephalopoda, and Gastropoda. We then used logistic regression to estimate the relationship between the calculated metabolic rates and extinction probability during each epoch of the Cenozoic Era. Results indicate that while each individual phylum has a different extinction probability across each epoch, the regression coefficients for the combination of all studied phyla illustrate no relationship since there is not enough evidence to reject the null hypothesis of no relationship between metabolic rate and extinction probability. Although this means that there is no significant correlation for most of the phyla, there are some exceptions where metabolism does affect extinction probability. During the Oligocene epoch, animals within the Mollusca phylum portray a clear negative correlation between metabolic rate and extinction probability. A negative relationship is also observed for Echinoderms during the Eocene epoch. Despite the crucial role that metabolism plays in species survival, our results indicate that more information is needed regarding specific environmental conditions in order to accurately predict the factors that ultimately affect species survival across marine animals within the Cenozoic Era.

Exploration of SST-Streamflow connection unravels the large scale climate influences that have a potential role in modulating local hydrological components. Most studies exploring this relationship only focus on seasonal or annual scales however, various atmospheric and oceanic phenomena occur at different timescales, which need to be considered. This study investigates the influence of sea surface temperature (SST) on German streamflow, divided into Alpine, Atlantic and Continental streamflow regions, at timescales ranging from sub-seasonal to decadal by integrating wavelet transform and complex network techniques. Wavelet transform is used to decompose the time series into multiple frequency signals, and the spatial connections are identified based on these decomposed signals for the 99 percentile correlation coefficient value by applying network theory. The degree centrality metric is used to evaluate the characteristics of the spatially embedded networks. Our results re-establish known SST regions that have a potential connection with the various streamflow regions of Germany. Spatial patterns that resemble the North Atlantic SST tripole-like pattern is predominant for Alpine streamflow regions at lower timescale. Equatorial Atlantic Mode regions observed for Atlantic streamflow at inter-annual timescale and Vb weather system connected regions in the Mediterranean Sea have appeared for all the streamflow regions of Germany. Besides, continental streamflow regions exhibited combined characteristics of the Alpine and Atlantic streamflow spatial patterns. In addition to the above regions, we also identify the scale specific patterns in the Pacific, Indian and Southern Ocean regions at different timescales ranging from seasonal to decadal scale.

An autonomous mobile ground-control point (AMGCP) was redesigned and refined for improved collaborative operation with an unmanned aerial vehicle (UAV) to enable calibration of image mosaics from multispectral (MS) and thermal cameras. The AMGCP has built-in reflectance panels and electronically controlled thermal panels that provide high and low reflectance and temperature references that can be used for calibration of reflectance measurements in MS images and temperature measurements in thermal-infrared images. The AMGCP also has an onboard temperature sensor that enables image-based temperature measurements to be compared to ambient temperature so that canopy temperature depression (CTD) can be calculated. The collaborative robotic system consists of the AMGCP and a UAV that have real-time kinematic (RTK) geographic positioning system (GPS) receivers onboard so their precise position can be determined in real time. The system also includes wireless communication capability between the AMGCP and UAV so they can transmit their position and other data to each other during a mission, in which the AMGCP positions itself at multiple locations under the flight path of the UAV, providing multiple instances of reflectance and temperature references in image mosaics collected by the UAV. Testing has shown that reflectance measurements can be calibrated to less than 1% reflectance error, and canopy temperatures of crop plants can be calibrated to within 1.0 C, enabling consistently accurate measurements to be made efficiently and without human intervention in various fields and regions and at different times and dates. This system is also suited to accurate measurement of CTD to facilitate genetic selection relative to various stresses and resilience characteristics like drought tolerance.