With last year’s commitment to all in-state sales of new passenger cars and trucks being zero-emission by 2035 (California Executive Order N-79-20), California is leading the charge for transportation electrification in the United States. Despite being at the forefront of climate change management and mitigation, California has some of the worst air quality in the nation. While primarily motivated by a desire to reduce carbon dioxide emissions and reliance on fossil fuels, transportation electrification will also have a significant impact on local air quality. The goal of this study is to quantify and qualify this impact in the context of urban ozone production. From robust studies of the weekend ozone effect, we know that reductions in vehicle emissions on weekend days can actually increase urban ozone concentrations. By examining data from eight ground monitoring stations in California’s South Coast Air Basin (SoCAB) over a period of 40 years, we show that this region is a volatile organic compound (VOC)-limited system in which the weekend ozone effect is a clear trend. Additionally, these data reveal that despite a significant decline in average annual nitrogen oxides (NOx) emissions, mean ozone levels have changed very little. With this in mind, the question looking forward becomes: how will local atmospheric chemistry and air quality evolve as transportation electrification accelerates? To investigate this question, VOC/NOx ratios are modeled for varying rates of light and heavy-duty vehicle electrification in order to gauge how urban ozone production will be affected. While it is clear that vehicle electrification will ultimately improve air quality and help mitigate climate change, this study provides a unique perspective into the less understood transient impacts of electrification.

The end of the California vineyard growing season is marked with a celebration of the harvest; however, the event may instead be cause for environmental concern. During the growth phase of wine grapes, a variety of organic compounds form and are stored under the skin of the grape. During the harvest and subsequent crush, these organic compounds are released from the skin and volatilize in the atmosphere. In early summer, mean background ethanol (EtOH) levels hover around 0.5-3 ppbv. Interestingly, during the mid-summer flights of the Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) campaign, average EtOH concentrations skyrocketed to 12-15 ppbv. In the late summer, EtOH concentrations return to still-enhanced levels of 5 ppbv. While the FIREX-AQ data were collected too early to capture the atmospheric impact of the crush, Central Valley EtOH concentrations are still much higher than earlier months of the year. With the help of EPA ground monitoring stations, general trends in ozone formation related to EtOH emissions will be investigated to understand the environmental impact of increased summer temperatures. These patterns indicate some potential sources of atmospheric ozone and shine light on season changes in Central Valley’s air quality. Finally, these data can then be applied to better understand the health impact for communities in the Central Valley.

The San Joaquin Valley (SJV) is home to Tulare County — a leader in the dairy industry and home to the largest population of cows in California. In recent studies, dairy farms have been identified as a significant source of volatile organic compounds (VOCs), which can degrade air quality and have negative public health effects. One such pollutant is dimethyl sulfide (DMS), a VOC typically emitted by phytoplankton and usually associated with marine sources. Recent findings suggest that there may be a number of terrestrial sources — including dairy farms. Dairy cows have been shown to emit DMS during rumination as a way of releasing excess sulfur. This study aims to determine if airborne data from the NASA Student Airborne Research Program 2009-2019 flights over the SJV can detect enhanced DMS mixing ratios originating from dairy farms. In the northern SJV, mixing ratios of 190 pptv were observed, a significant enhancement for an inland location. NOAA HYSPLIT backward trajectory models and correlations with marine chemical tracers revealed that this enhanced DMS was a result of transport from the coast. In the southern SJV near Tulare County, the average DMS mixing ratio was calculated to be 14 pptv, with a maximum of 49 pptv. The topography of the area and calculations of DMS lifetime suggest that while marine DMS can still reach the southern SJV, there is a lesser marine influence because much will be lost during transport. Subsequently, we suggest that DMS mixing ratios in Tulare County observed greater than ~10 pptv and correlated with enhanced methane, were influenced by a local dairy source.

Methyl nitrate (MeONO2), an important gas in the production of ozone (O3), is distributed unevenly at the sea surface across the globe. MeONO2 is one of the main sources of NOx in the remote atmosphere and is responsible for production of O3, which protects Earth from UV radiation. Accordingly, understanding the production, sources, and sinks of MeONO2 is important under the context of human health. This study evaluates the potential causes for enhanced concentrations of MeONO2, which are localized to the tropical and southern Pacific and Southern oceans. In this study, wind speed was evaluated as a potential influence for this trend in MeONO2 distribution.  Global wind speed at the sea surface from satellite data was sourced from the Remote Sensing Systems (RSS) Monthly 1-degree Merged Wind Climatology netCDF dataset V7R01. The wind speed measurements included in this dataset were collected by satellite microwave radiometers SSM/I on satellites F08 through F15, SSMIS 16, SSMIS 17, AMRS-E, AMSR-2, and WindSat. The wind speeds over the oceans globally, as recorded in this dataset, were analyzed in this study in conjunction with MeONO2 concentrations measured by the UCI whole air sampling (WAS) instrument during NASA Atmospheric Tomography mission (ATom) flights 1-4. Results indicate that wind speed may be an influencer on global distributions of MeONO2, however further research is needed to determine whether this trend is due to increased chemical production of MeONO2, or whether the elevated amounts of MeONO2in these regions are only due to increased ocean-atmosphere mixing.

Every summer, fires produce a thick haze in the air. The kind of haze that turns the sun red and the sky brown. This haze is seen across the United States, including the south. The Mississippi valley in particular is marked by a multitude of small fires, typically crop burns. While the individual fires aren’t large in magnitude, the quantity of them produces a ubiquitous haze comparable to large fires seen in the west. Fire tracing compounds — such as ethyne and furan — act as an indication of smoke age by comparing ratios of these short-lived gases against a longer-lived gas such as carbon monoxide. These ratios will be compared for fresh smoke from active fires, haze in the boundary layer, and smoke in the lower free troposphere, allowing for a determination on the pervasiveness of smoke. While the aging process and chemical changes of smoke have been studied extensively, there has been comparatively little research on the lingering effects. The presence of aged smoke will indicate a high pervasiveness and a long-lived impact on air quality. The presence of new smoke will imply that the high quantity of fires continuously replenishes the haze, and that the negative impact of individual fires on air quality are relatively short lived. Understanding the lasting effects of wildfire smoke grows ever more important as the change in earth’s climate leads to more large smoke events. If smoke haze is becoming the new normal, it is vital that a strong understanding of this haze is developed.

Methyl chloride (CH3Cl) is the most abundant source of natural organic chlorine in the atmosphere and a significant stratospheric ozone depleter. As a tracer of biomass burning, slightly enhanced concentrations of methyl chloride are expected in fire plumes, but FIREX-AQ (Fire Influence on Regional to Global Environments Experiment – Air Quality) data revealed significant enhancements in the southeastern United States. Methyl chloride has an average background concentration of 0.6 ppb, but concentrations as great as 9.9 ppb were observed over select fires in the southeast, about a factor of 10 greater than what I observed in the northwestern US. Upon further examination, the major sources of methyl chloride were not forest fires, but prescribed agricultural burns. The crops burned in this region were corn, soybeans, and rice, and each crop fire emitted excess methyl chloride. Agricultural burns are expected to have similar emissions of methyl chloride to forest fires, so this phenomenon suggests there are more factors at work than just biomass burning. Fertilizer, the composition of crops, and pesticides are discussed as potential sources of excess methyl chloride. Additional research is needed to better understand the source of enhanced methyl chloride above prescribed burns and quantify the impact of these emissions on stratospheric ozone depletion.

Methyl chloroform (CH3CCl3) was a commonly used solvent and degreasing agent prior to its classification as an ozone depleting substance, and subsequent prohibition by the Montreal Protocol in 1987. With an atmospheric lifetime of five to six years, temporal measurements of methyl chloroform illustrate the success of these restrictions as background values fell from approximately 130 ppt in the early 1990s to 1.5 ppt in 2020. The COVID-19 lockdowns brought about an ease in EPA restrictions, as well as a new opportunity for the NASA Student Airborne Research Program to extend whole air sampling across the United States. The results from this widespread sampling discovered enhanced concentrations of methyl chloroform near various highly populated coastal areas with the highest observed in Ithaca, NY at 16.4 ppt. Given the status of methyl chloroform as an ozone depleting substance, its use requires permission from the EPA implying its use should be easy to track. An investigation into past uses of methyl chloroform was conducted to determine if any plausible sources for these enhanced concentrations still currently exist. To confirm the validity of these sources, multiple NOAA HYSPLIT trajectories were run on the dates of enhanced observations. There are promising associations between the location of potential emission sources and enhanced methyl chloroform measurements. In order to further support these results additional whole air sampling must be conducted near sites of interest.

California wildfires result in the loss of life and destruction of property, and have a significant impact on the environment. As wildfires are becoming more prevalent and severe over the years, there is an increased need to better understand them, and measure their severity. In this project, we studied the Thomas fire and sought to find a measure of fire severity, stemming from the relationship between the type of pre-fire fuels and burned products, comparing this to more standard measures of severity, such as the difference Normalized Burn Ratio (dNBR). We used data from the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) acquired on December 21, 2017, during the fire, over Santa Barbara and Ventura counties. A spectral library was built consisting of four classes: green vegetation (GV), non-photosynthetic vegetation (NPV), soil/rock, and ash, then optimized using Iterative Endmember Selection (IES). The image was then classified into GV, NPV, soil/rock, and ash, using Multiple Endmember Spectral Mixture Analysis (MESMA). We also used data from the European Space Agency’s Sentinel-2 satellites, taken before and after the fire. With these images, burn severity was calculated using the dNBR. Regions of varying burn severity were classified according to the dNBR value. The Sentinel-2 images were then classified using MESMA with the optimized spectral library, and evaluated to determine the relationship between the pre-fire fuel type and fire severity. AVIRIS-derived maps of burned products were compared to dNBR to evaluate how closely they modeled fire severity.

Wildfires will continue to be an issue due to sustained extreme drought conditions and meteorological factors, specifically in the western United States. This affects downstream air quality and public health. Based on the diurnal effects of solar radiation, moisture, turbulent eddies, behavior of the boundary layer, etc., fire behavior should have diurnal changes as well. Little research has been done in terms of cross-comparing sensors and the raw radiance values they collect, so this research seeks to look into the differences that each instrument has in depicting fire events. Three instruments that gather radiance data at varying temporal and spatial resolutions were compared to quantify how wildfires behave through time and space: the Moderate Resolution Imaging Spectroradiometer (MODIS) from the Aqua and Terra satellites, the Visible Infrared Imaging Radiometer Suite (VIIRS) from the Suomi NPP satellite, and the Advanced Baseline Imager (ABI) from the GOES-16 satellite. The Williams Flat Fire, from 2019 outside of Keller, Washington, was targeted because it was sampled by all three sensors, burned a large area (44,000 acres), and had a long duration (24 days in August). Radiance values collected by each sensor were compared within the subset area. Longwave infrared wavelengths are optimal for looking at active fires within this range, so MODIS Band 7 (2.1 μm), VIIRS I4 Band (3.55 μm), and ABI Band 7 (3.9 μm) were chosen to cut down on cloud contamination. MODIS maximum radiance values were collected once per day and graphed to show changes in the radiance with time. With more ABI data readily available, the highest temporal resolution was utilized to show how the fire evolved over a 24-hour period and its movement. Finally, VIIRS has the highest spatial resolution to quantify the amount of pixels on fire at 375 meters. Mapping the maximum radiance value with a specific pixel displays a pattern of the fire’s movement through time. Results indicate that a diurnal cycle can be seen from the use of high temporal resolution.

Wildfires have become a large topic of discussion as they have been utilized as evidence to support the validity of anthropogenic climate variability. The topic of my current research project to analyze two previously-extinguished California wildfires, Kincade (2019) and El Dorado (2020), and how the different types of vegetation that these fires burn affect the composition of the atmosphere in and around the fire. The overarching question of my project is to determine which types of currently-present native California vegetation will emit the highest aerosol pollution if burned. Methods include utilizing level 2 remote sensing imagery, via SENTINEL-2 and NAIP, and creating regions of interest (ROI) within the ENVI and ENVI Classic software packages. Once the regions of interest are drawn, the image and ROI files are put through a random forest machine learning algorithm to map dominant fuel types. Upon receiving a high percentage of accuracy, FIREX-AQ data will be observed to correlate collected in-situ aerosol data to different locations within the area of Kincade and El Dorado burn scars. The project will result in the determination of the most destructive vegetation, when burned, to the overall health of the central and southern California atmosphere. In conclusion, the long-term end goal of this project is to provide new information to end-users (fire departments, government officials, etc.) concerning how best to prevent the spread of wildfires within this region.

The Thomas Fires and Subsequent Montecito Debris Flows were both historic events that caused unprecedented damage and loss of life. Unfortunately, these were not isolated incidents, but instead part of a larger global pattern of increased occurrence and severity of natural disasters due to climate change. Although actions are being taken by many countries to prevent further climate change effects, anthropogenic activity has already cause irreversible damage in many ways. Thus, it is more important than ever to be able to predict and assess risk in order to safely live in natural-disaster prone areas like the state of California. This project aims to contribute to this need by modeling debris flow risk in post-wildfire conditions, based on factors such as slope, geology, vegetation cover, and fire severity. The model was created in GIS with least-cost-path raster analysis—rasters were either found through public sources or manually created through the use of spectral libraries and AVIRIS imagery. The output model shows that areas of risk can be identified through this form of analysis, however, future work could be done to make a more robust, complex, and accurate model. Specifically, tools like machine learning could be an important asset in optimizing the modelling process.

Plant productivity is the evaluation of a plant’s ability to produce biomass, uptake carbon, photosynthesize, and its overall greenness. In this study, twenty-four Sentinel-2 images from March 2019-February 2021 of the broad southern California region were used. The region of study focused on the Oceanside city area and the, mostly, undeveloped area to the north. Furthermore, using the GIS-based vegetation maps provided by the San Diego Association of Governments (SANDAG), a detailed map of the vegetation types was created for the region of study. Types of vegetation identified were grasslands, woodlands, chaparrals, etc. Four measurements of plant productivity were calculated and used to determine the productivity of the region. Normalized Difference Vegetation Index (NDVI) was used to find an average measure of greenness for each vegetation type. Light Use Efficiency (LUE), estimated using Photochemical Reflectance Index (PRI), was used to measure the ability of vegetation types to convert absorbed energy into biomass. Using Absorbed Photosynthetically Active Radiation (APAR) and the LUE measurements, values of carbon uptake were generated. Using LUE with PRI values, plant photosynthesis could be estimated. The measurements calculated were used to evaluate and compare a vegetation type’s productivity on the yearly and seasonal scale.

Over the past decades, California has spent the majority of its time in a state of severe drought. The droughts were periodically mitigated by large-scale precipitation events brought about by El Niño Southern Oscillation (ENSO). Recent studies suggest that these ENSO precipitation events could help restore arid ecosystems. With Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) and Landsat imagery, this research utilizes remote sensing techniques to understand the effects of ENSO on vegetation communities of ecological or conservation importance in California. The AVIRIS imagery were used to build a spectral library of various plant functional types (PFTs) which can classify the spectral signatures found in Landsat. In three regions of high conservation priority, changes in the senescence of PFTs were monitored through Landsat images taken in March, May, and July of 1998, 2005, and 2016. These three years correspond to high ENSO intensity, low to no ENSO intensity, and anomalous ENSO intensity (when precipitation patterns were not as predicted), respectively. The insights provided on the specific interactions of El Niño and California vegetation communities can be leveraged to assist in future conservation efforts.

As global temperatures continue to increase, the consequences of drought are becoming more prevalent and are affecting every aspect of the environment, including its natural vegetation. Invasive species are naturally more resilient to droughts due to their genetic variability, but they are also harmful to the environment. To better understand and quantify the relationship between plant functional type and drought, two native species – quercus agrifolia (live oak) and adenostoma fasciculatum (chamise) – and two invasive species – brassica nigra (black mustard) and eucalyptus globus (blue gum eucalyptus) – were observed during drought and non-drought years. Based on average rainfall compared to the thirty-year average, 2018 was the drought year and 2019 was the non-drought year. The area of interest is the Santa Barbara and Goleta regions stretching from the coast to the Santa Ynez mountain range. Airborne Visible InfraRed Imaging Spectrometry – Next Generation (AVIRIS – NG) data taken on 17 April 2021 were used to locate each species in the region. One-meter high-resolution data taken by the National Agricultural Imagery Program (NAIP) in May of 2020 and the imagery available on Google Earth Pro were both used to increase accuracy when mapping the species on the AVIRIS-NG data. The greenness in the mapped regions was then measured on forty-nine Sentinel-2 images spaced approximately fifteen days apart from 28 December 2017 to 28 December 2019. The drought impact was determined by greenness levels throughout the season and in the peak of the growth season. Higher levels of greenness in a species during drought would suggest greater drought resilience. It is expected that the invasive species will be slightly less impacted by drought, meaning it will have higher levels of greenness. If research were to be conducted over longer periods of wetness and dryness, this slight impact could be amplified.

Iron is a key micronutrient for phytoplankton growth in the surface ocean. Studies have shown that volcanic eruptions create significant amounts of bioavailable iron in the surface ocean. Low Nutrient-Low Chlorophyll (LNLC) oceanic regions have an insufficient supply of macronutrients and are iron-limited. In LNLC regions, also referred to as oligotrophic regions, phytoplankton biomass sustains low concentrations year-round. Previous research has been done on the effects of volcanic ash on Marine Primary Production (MPP) in high nutrient regions, specifically, the 2008 Kasatochi eruption leading to the largest phytoplankton blooms observed in the subarctic North Pacific which had a Volcanic Explosive Index (VEI) of four. Similar research has also been done on the effects of MPP on higher explosive volcanoes of VEI ≥ 4. However, there is a lack of research on the impacts of smaller explosions in nutrient-depleted regions. Using MODIS-AQUA satellite data at both 1 and 4 km spatial resolution, we analyzed the effects of a smaller eruption (VEI 2) in an LNLC region by studying the late 2012 North Pagan Volcanic Eruption. The eruption occurred October 30th through approximately December 11th, 2012. Between the week prior to the eruption and one day after the eruption began, we observed a 0.0413-0.0809 average increase in bioavailable iron using a bio-optical proxy bbp/NFLH. Following the eruption, the average chlorophyll-a concentrations fluctuated from 0.0502 to 0.0748 mg/m3 during the months of December through early February. In this project, we created a time series prior to, during, and after the eruption in 2012 as well as a synchronized time series of 2008 for adequate comparisons. During major points in the eruption, we generated a HYSPLIT model for visualization of airborne ash particles depositing into the ocean surface. The results show compelling evidence that small-scale volcanic eruptions can create phytoplankton blooms in LNLC regions.

Within the last six years, several organizations surrounding San Francisco Bay have proposed restoration projects aimed at improving the area’s water quality. The goal of this project is to quantify the effects of these restorative efforts by focusing on the relationship between vegetation presence, specifically within riparian buffer zones (RBZs), and the observed suspended particulate matter (SPM) within the Bay. RBZs were defined as a 30m buffer zone surrounding the streams and located within the planning watersheds adjacent to the Bay. Using Sentinel-2 satellite imagery, the average normalized difference vegetation index (NDVI) within the RBZs was computed from September 2015 through June 2021. Additionally, the average SPM was calculated using the Nechad algorithm via ACOLITE on a synonymous timescale. Furthermore, precipitation data was obtained via NOAA National Center for Environmental Information and used to help constrain SPM anomalies due to high river discharge. The NDVI and SPM values were then plotted against each other and on a time-series over the past 6 years to visualize how the relationship has changed over time. The Kendall rank correlation coefficient was determined to better understand the statistical significance of this relationship while considering the seasonality of the data. Preliminary results suggest that the correlation between the NDVI within the RBZs and SPM is positively significant in the northern part of the bay, and likely due to precipitation patterns. Further analysis was done examining NDVI values within individual planning watersheds and the SPM within a 1km buffer on the adjoining coast. A linear regression equation was then used to relate different watersheds and how the NDVI and SPM relationship varies. This more narrowed approach may provide information on how the presence of plants visibly improves sediment retention detected via satellite.

As the earth’s climate continues to fluctuate, marine heatwaves are becoming more common, impacting many facets of marine ecosystems. The rising sea surface temperatures associated with these heatwaves have resulted in changes in nutrient availability and species composition. In recent years, studies have shown dramatic decreases in the availability of previously abundant marine species off the coast of California. These observed shifts have the capability to disrupt marine food webs and other marine life processes. This study aims to estimate the trends in concentration and composition of phytoplankton within the Santa Barbara Channel through a marine heatwave that occurred from 2014-2017. MODIS-Aqua satellite data was used to create time series analyses from 2005-2021 in order to track the chlorophyll levels and sea surface temperatures within the channel. Additionally, map projections from 2013, 2015, 2018, and 2020 were rendered from the MODIS-Aqua data averaged monthly over the area of the channel at 1 km resolution to better visualize the distribution of the chlorophyll. The combination of time series and map projection data allowed for trends to be revealed, indicating a drop in chlorophyll levels during the heatwave and a gradual recovery after the heatwave. Furthermore, in situ data gathered from water samples taken from Stearns Wharf in Santa Barbara suggested a shift in the types of phytoplankton present in the post-heatwave recovery within the channel. Results of this study show an increase in phytoplankton species such as Cochlodinium, which cause red tides and harmful algal blooms. This suggests that these toxic and dangerous species could be outcompeting the phytoplankton that existed within the channel prior to the heat wave.

Atmospheric rivers are long columns of condensed water vapor carried from over the ocean. Strong atmospheric rivers carry as much water as 15 Mississippi rivers and can be extremely disruptive. When these atmospheric rivers come in contact with the United States West Coast they generate a series of storms. On average 30-50% of the West Coast annual precipitation comes from just a handful of atmospheric river events, thus the significant events are enormously impactful on the water balance of the coast. This study examines the impact of atmospheric river events affecting the coastal water before and after significant atmospheric river events, with regards to temperature, turbidity, and algal blooms. The study covers four recent atmospheric river events from 2019-2021, focusing on an event that lasted from February 4th, 2020 to February 10th, 2020. The potent atmospheric river during this time resulted in widespread rainfall of 3-6”, with some areas receiving up to 16” of rain. This led to significant flooding impacts throughout the State. Analysis was conducted using MODIS-Aqua data for sea surface temperature, chlorophyll, and KD 490, which is the diffuse attenuation coefficient at 490 nm and one indicator of the turbidity of the water column. This study aims to examine the data before and after an atmospheric river event to identify effects. Preliminary results indicate a pattern of a decreased coastal sea surface temperature after an atmospheric river event, as well as an increase in chlorophyll and KD 490. This examination is applicable to aid in the preparation for the coastal effects of these rain events as well as contribute to forecasting during or after an atmospheric river.

Global warming and increased carbon emissions have created an unprecedented need for clean production of electricity. One solution has been the implementation of offshore wind farms to take advantage of stronger, more reliable air currents found over large bodies of water. However, this technology has only recently been implemented with the first offshore wind farm being constructed only 30 years ago. Likewise, not all aspects of offshore wind farms and their environmental impacts are entirely understood. This study uses the Modis (Moderate Resolution Imaging Spectroradiometer) instrument aboard NASA satellite Aqua to analyze the impacts of three offshore wind farms on localized sea surface temperature (SST). The wind farms being observed are the Block Island Wind Farm off the coast of Rhode Island, Kentish Flats Wind Farm in the Thames Estuary of the North Sea, and the London Array also in the Thames Estuary of the North Sea. This study was carried out using two methods. The first method uses the software SNAP (Sentinel Application Program) to display temperature spatially to search for increases or decreases in temperature created when the turbines are generating electricity. This was carried out by creating regions of interest around the wind farms and used images taken before and after turbine installations. The second method uses a climatology approach to understand temporal patterns in SST which may have been impacted by the presence of the wind farms. This has been done by plotting time series graphs of SST across the lifespan of Modis Aqua and averaging SST before installation from years prior to installation and after operations began. Together, this study is intended to further the understanding of how our pursuit of offshore wind energy has impacted the environment so that more informed decisions can be made in the future.

Remote sensing datasets have applications across all fields of science. But what good is a bad dataset? Atmospheric correction is important for removing the effects of gases, aerosols, and water vapor that filter the top-of-atmosphere radiance data received in imaging spectroscopy. This can impact a wide range of applications of remote sensing such as analysis of wildfires, surface temperature, weather forecasting, and so much more. For the airborne sensor, AVIRIS, data are currently atmospherically corrected using the radiative transfer-based ATmospheric REMoval (ATREM) program, but several other methods and programs exist, such as the Cloud-Shadow approach. Though large datasets require quick and efficient corrections, individual imagery analysis may need more thorough, in-depth corrections to provide accurate data. In these cases, the Cloud-Shadow method may provide better quality atmospheric corrections when applied to a single image. Image criteria is simply a discernible object and its shadow, which can be a useful approach when atmospheric conditions are not readily available. While this is a more individual approach, performing the correction one image at a time, could be the key to improved atmospheric correction in imagery from airborne sensors such as AVIRIS. Satellite imagery is often continuous and produces hundreds of images a day. As airborne sensors are only flown at certain times, they produce fewer images, but in higher quality, allowing the implementation of a more manual, intensive approach. Analysis was done with imagery from an AVIRIS flight on 16 August 2018 of the Bay Area, CA, compared to the JPL-produced L2 reflectance data. Quality control comparisons were also done with satellite imagery from Sentinel 3 and MODIS-Aqua, as well as in-situ data from local AERONET sensors, within the timeframe of 2 days of the flight to verify atmospheric conditions and the accuracy of the atmospheric corrections.

Colored Dissolved Organic Matter (CDOM) is an important environmental marker in ocean water. The concentration (absorption) and spectral slope provide information about the source of the water. Consisting of organic dissolved material from various sources, it strongly absorbs light at blue and ultraviolet wavelengths, thereby influencing the energy available for phytoplankton, as well as overlapping with chlorophyll absorption spectra. This study models CDOM spectra based on satellite measurements of downwelling irradiance and water-leaving radiance. Samples measured in a laboratory setting are used to construct linear models that map ratios of radiance and irradiance values to CDOM absorption values. By fitting these extrapolated values to known mathematical models of CDOM spectra, an algorithm is developed for finding the CDOM spectral slope and reference absorption based on available remote-sensing quantities alone. Error bounds for the algorithm are determined by evaluating the root mean squared distance between the modeled and the reference samples. Using a second dataset linking spectral slope and absorption values to different CDOM types, a classification method is established to categorize a given unknown value from the remote sensing algorithm. The categorization method is then applied to Sentinel-3 and Landsat imagery, resulting in geospatial data extrapolating CDOM type from radiance values for each pixel of the image. The model results are compared to ground-truth data collected at sites located within the satellite image.

Power plants have long been a subject of interest in studying the effects of anthropogenic activity on Earth’s atmosphere and human health. Specifically, power plant plumes often exhibit elevated concentrations of various gases (e.g., SO2, NO, NO2, CO, CO2) as well as particles directly emitted (primary aerosol). These gases can also proceed to form secondary aerosols, including ultrafine particles (here defined as particles with a diameter of less than 100 nm), which have severe health effects for humans as they are able to penetrate deep into the respiratory system. Data from the NASA KORUS-AQ campaign were used to identify and characterize six power plant plumes in South Korea during May and June 2016. Plumes were identified by increased concentrations of various gases (SO2, NO, NO2) and then analyzed for their aerosol mass concentrations of certain species (sulfate [SO42- ], nitrate [NO3-], and ammonium [NH4+]) as well as number concentrations of ultrafine particles. The aforementioned aerosol concentrations within the plumes were compared to background concentrations observed over the West Sea to understand the health effects associated with living near each of these power plants. Mass concentrations of SO42-, NO3-, and NH4+, as well as number concentrations of ultrafine particles, were enhanced above background levels in all plumes. On average, two plumes (Plume #4 and #5) had the highest concentrations of aerosol precursor gases. However, Plume #4 had significantly higher conversion to aerosol and was identified as being the most dangerous to local inhabitants due to its greatly enhanced particle mass and number concentrations.

Over sixteen million people live in the Po Valley of Italy, a region with some of Europe’s worst air quality. Previous studies found a seasonal cycle with higher concentrations of ground-level air pollution during the winter, but none have analyzed aerosol vertical profiles across the valley. The vertical extent of this pollution is of importance as the Po Valley is surrounded by mountains to the north, south, and west. If the pollution is not able to rise above the elevation of the surrounding mountains, it may become trapped within the valley, putting its inhabitants at high risk. Understanding which meteorological conditions trap polluted air near the surface can inform pollution mitigation policies. Vertical structure of aerosols in the Po Valley on certain nights from 2017 through 2018 were observed using profiles of aerosol backscatter coefficients, an indicator of cloud presence and aerosol particle concentrations, from the CALIOP (Cloud-Aerosol Lidar with Orthogonal Polarization) lidar on the NASA/CNES CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation) satellite. A case study was conducted looking at two nights of comparable temperature, one in the boreal winter (19 February 2017; average nightly temperature [ANT] of 5°C) and the other in the spring (26 March 2018; ANT of 5°C), as well as two nights during the summer (14 August 2017 and 17 August 2018; ANTs of 20 and 21 °C, respectively). Both nights studied in the winter and spring had high concentrations of aerosol at the surface (PM2.5 concentrations of 159 and 70 µg m-3, respectively), with a cleaner atmosphere above and a well-defined aerosol layer extending ~2 km and ~4 km above the surface, respectively. Boundary layer heights and temperature inversions could not explain the presence of an aerosol layer during colder months. However, there was a correlation between the aerosol layer and higher relative humidity. In comparison, the summer nights had a more well-mixed aerosol distribution and lower surface PM2.5 concentrations (25 and 42 µg m-3, respectively). Thus, a seasonal trend in aerosol vertical distribution was present, showing the importance of understanding which conditions form well-defined aerosol layers.

Clouds have an important role in Earth’s radiation budget and climate change as effective reflectors of incoming light. Aerosols emitted from both natural and anthropogenic sources can affect cloud properties such as brightness and precipitation efficiency, a process known as the aerosol indirect effect. The aerosol indirect effect is currently the largest source of uncertainty in determining the impact of anthropogenic activity on radiative forcing. According to the Twomey Effect, an increase in aerosol concentration from pollution causes an increase in cloud droplet number concentration and, therefore, a decrease in cloud droplet size for clouds with the same liquid water content. This increase in droplet number concentration causes clouds to become more reflective of incoming solar radiation which, on large scales, results in a stronger cooling of the atmosphere. This project focuses on understanding the effects of aerosols on cloud microphysical and radiative properties in the Philippines, an area of the world particularly susceptible to climate change and pollution from both local and global sources. Data from the NASA Cloud, Aerosol, and Monsoon Processes – Philippines Experiment (CAMP2Ex) were used to obtain cloud droplet size distributions for clouds over the Sulu Sea affected by the Borneo smoke plume (observed during the ninth research flight – RF9) as well as clouds over the South China Sea affected by urban pollution (observed during RF17). Carbon monoxide (CO) concentrations were used to differentiate between polluted and clean clouds, and cloud droplet size was examined by calculating the cloud droplet effective radius (reff) for each cloud pass. During RF9, high levels of pollution (CO > 0.24 ppm) over the southern region of the Sulu Sea correlated with smaller cloud droplet effective radii (reff < 8µm) compared to cleaner clouds with similar water content and at a similar altitude. The same general trend was observed for RF17. Further work must be done to correlate this observation with cloud albedo and thus quantify the effect of pollution on cloud radiative forcing in the study regions.

Black carbon (BC), or soot, is one of the few aerosol types that absorbs light and, therefore, has a net warming effect on the atmosphere. BC is primarily released via incomplete combustion, which is observed during many anthropogenic processes as well as during biomass burning events, such as wildfires. With the increasing occurrence of extreme heat and drought accompanying climate change, wildfires are predicted to increase in breadth and severity. Thus, there is an increasing urgency to understand how the projected increase in BC emissions will affect the atmospheric radiation budget in and around fire plumes. FIREX-AQ 2019 (Fire Influence on Regional to Global Environments and Air Quality 2019), a NASA and NOAA mission, was a six-week investigation dedicated to retrieving wildfire smoke data across the western and southern regions of the United States. BC and other necessary data were retrieved in and around multiple fires, including the Tucker Fire in northern California on July 29, 2019. Total instantaneous energy absorption rates of BC and the resulting ambient air heating rates were determined for each of the nine airborne transects through the Tucker Fire smoke plume. In this study, total instantaneous energy absorption rate reflects the rate at which BC absorbed radiation with wavelengths between 400 and 640 nm, while the ambient air heating rate describes the rate at which the ambient air was heated due to the absorption of radiation by the BC particles. BC mass concentrations within the smoke plume ranged from 350 – 3049 ng m-3, which was considerably higher than the average background value observed outside the plume (1.4 ± 4.9 ng m-3). Total instantaneous energy absorption rates and ambient air heating rates within the plume were assumed to only depend on BC mass concentrations and ranged from 9.89×10-5 to 8.62×10-4 W m-3 and 7.05 x10-3 to            6.14 x10-2 K day-1, respectively, which were both at least 250 times higher than the average background levels. This study assumed that all BC particles were uncoated, which is often not the case in smoke plumes where various organics and other species can accumulate on the surface of BC particles over time. These coatings can thus change the optical and heating properties of the BC. Further studies looking into the evolution of BC particles and their coatings with time are necessary in order to fully understand the magnitude of warming associated with BC particle energy absorption.

The North Atlantic plankton bloom is a particularly active bloom located to the east of St. Pierre and Miquelon in Canada and extends into the Labrador Sea towards Greenland. The plankton emit varying amounts of organic components such as dimethyl sulfide (DMS) throughout the year, with the peak of emissions observed in spring and summer and the lowest emissions observed in the winter. DMS is converted to sulfate aerosols in the atmosphere, which can lead to increased aerosol number concentrations in this relatively pristine environment. However, the influence of increased DMS concentrations on cloud condensation nuclei (CCN) activity in this region is not yet fully understood. This study utilizes airborne data from the NASA North Atlantic Aerosols and Marine Ecosystems Study (NAAMES) campaign as well as satellite retrievals of chlorophyll a concentrations in the northwest Atlantic Ocean during the winter of 2015 (low bloom activity associated with low DMS emissions) and spring/summer of 2016 (high bloom activity associated with increased DMS emissions). Relationships between DMS concentrations and CCN concentrations within the boundary layer (< 1 km) did not show a strong correlation. This finding is important as it indicates that increased DMS emissions in the spring and summer have little effect on local CCN concentrations in the region.  The scientific community must look at other factors to predict changes in CCN. Concentrations of CCN directly impact climate through their ability to influence cloud radiative properties, and, thus, future work is needed to understand which aerosols and events influence their concentrations in this location.

Volcanic eruptions release high concentrations of sulfur dioxide (SO2), which reacts with water to form sulfuric acid aerosols. Sulfuric acid aerosols reflect shortwave solar radiation, significantly reducing the amount of solar radiation reaching Earth’s surface. Studying local radiative forcing resulting from volcanic eruptions can reduce the uncertainty in estimating the cooling effects of natural sulfur dioxide emissions. The 2018 Kīlauea flank eruption on the lower East Rift Zone (ERZ) and the resulting collapse of the floor of the volcanic crater released enhanced levels of SO2 and coarse material into a relatively pristine environment. Analyzing the optical and radiative properties derived from two AErosol RObotic NETwork (AERONET) sites (Naalehu and Mauna Loa; 49 km and 34 km from Kilauea, respectively) and Interagency Monitoring of Protected Visual Environments (IMPROVE) data of aerosol composition and abundances showed the aerosols at Naalehu site to be primarily dominated by sulfates and coarse mass while there was a weak correlation between ammonium sulfate and increases in aerosol optical depth (AOD) at the Mauna Loa site. Abundance of volcanic aerosols at Naalehu are supported by Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) modeling and satellite imagery of the plume. The estimated radiative forcing effect of sulfates based on AERONET inversions at Naalehu were -1 W/m2 per μg/m3 of SO2 mass concentration at Kīlauea. However, only 29 of these inversions were performed during the two months of measurements at Naalehu due to constraints on when they can be performed.  By fitting the power law wavelength dependence of single scatter albedo and backscatter fraction, we can recalculate the radiative forcing using AOD measurements to obtain higher temporal resolution radiative forcing estimates. With higher resolution data, the radiative forcing effect of volcanic sulfates can be better constrained

This study uses airborne data from the NASA FIREX-AQ campaign to investigate a pyro cumulonimbus (pyrocb) cloud that formed near the Williams Flats Fire on 8 August 2019. A pyrocb is a cloud that is influenced by wildfire aerosol emissions and rarely investigated due to the lack of data. Microphysical cloud properties (cloud droplet effective radius [reff] and liquid water content [LWC]) for the pyrocb were compared to “clean” clouds (i.e., clouds that were not influenced by smoke). Results shows that for the clean cloud, reff values (0.38 – 23.98 µm) were larger than for the pyrocb (0.38 – 18.46 µm). Comparisons of clouds with the same LWC values showed smaller reff values for the pyrocb than in the clean cloud. This is in agreement with the Twomey Effect, which suggests that more polluted clouds produce smaller droplet sizes which cause more visible light from the Sun’s solar radiation to be reflected back to space; thus, essentially increasing the albedo of clouds.