Search NASA⌕ Search

SEARCH · Search NASA

Results for “Random forests”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 379 records · Page 21

Maya Forest Water Resources I: Using NASA Earth Observations to Map Forested Inundation in the Maya Forest

As climate change increases the severity and frequency of extreme weather events in the tropics, it is vital for the safety of local communities and the health of ecosystems to monitor seasonal inundation. Forested inundation affects the ability of forested wetlands to provide ecosystem services, such as flood mitigation, water filtration, carbon storage, and erosion mitigation. While ground-based monitoring has traditionally been used to map inundation extent, those methods are costly and time-intensive. The NASA DEVELOP team focused on seasonal inundation throughout 2008 in the Maya Forest, when changes in inundation were drastic. To monitor seasonal inundation, our team used in situ field data and Earth observations from Landsat 7 Enhanced Thematic Mapper (ETM+), Advanced Land Observing Satellite (ALOS) Phased Array type L-band Synthetic Aperture Radar (PALSAR) 1, Shuttle Radar Topography Mission (SRTM), and products from the Ice, Cloud, and Land Elevation Satellite (ICESat). The team applied a Random Forest algorithm to Landsat 7 imagery, generating an object-level land cover classification with an overall accuracy of 72.1% and forest class with 100% recall and 78% precision. The team applied L-band backscatter thresholds from existing literature to forest-masked ALOS imagery and refined the thresholds in an iterative process using field data and hydrology models to delineate seasonal inundation extent. These publicly available data products help end users from Belize’s Land Information Center (LIC) and Forest Department, Guatemala’s Center for Monitoring and Evaluation (CEMEC), and Mexico’s El Colegio de la Frontera Sur (ECOSUR) to inform land management and protect community infrastructure.

Madelyn Savan↗

Transcriptomics-based Machine Learning Analysis Predicts Space-Exposed Murine Livers

Limited sample sizes, high data dimensionality, and sensitivity to technical and biological variability of next generation sequencing (NGS), typically limits machine learning (ML) approaches in spaceflight studies that include radiation effects. However, pooling smaller studies while addressing intra- and inter-study variabilities allows for ML predictive modeling. Here, integration methods were applied to whole transcriptome shotgun sequencing (RNA-seq) data from six mouse liver GeneLab datasets (GLDS) (n ranging from 6 to 39 samples) from with a total of 81 spaceflight and ground-control samples to determine top features (i.e. genes) relevant to spaceflight including the effect of radiation exposure. RNASeq counts were normalized for each study, then merged and scaled across all datasets. Data dimensionality was reduced using a minimum redundancy maximum relevance (MRMR) methodology. Redundancy and relevance were computed using the Pearson correlation and F-statistic, respectively. The top 100 MRMR features were used to predict spaceflight vs. ground-control samples using Random Forest (RF), Support Vector Machine (SVM), and Linear Discriminant Analysis (LDA) classifiers with 5-fold cross validation (CV). Principal component analysis (PCA) on the complete feature set versus the MRMR features shows separation between spaceflight samples and ground controls (Figure 1A). The ML-based gene sets were compared against differential gene expression results obtained with DESeq2 from individual GLDS. Using all features or randomly sampled subsets at matching set sizes with MRMR, a maximum classifier accuracy of 69% was shown on the test set over 5 folds. For all classifiers, CV training using at least the top 30 MRMR genes show minimum 89% accuracy and 0.95 AUC value on the test set over 5 folds (Figure 1B). Baseline set analysis on differentially expressed genes (DEGs) identified using padj ≤ 0.05 show 295 DEGs that overlap at least two studies and 13 DEGs that overlap three studies (Figure 1C). Set analysis between the top 100 MRMR features and the DEGs showed 47 genes that overlap at least one study and 24 genes that overlap two studies. Over-representation analysis showed overlapping biological processes related to fatty acid and lipid metabolism which may indicate these processes in the response to spaceflight stressors. MRMR feature selection for the selected ML methods improve performance relative to a classifier built on all features or randomly sampled subsets. Permutation feature importance within the decorrelated MRMR features showed concordance in feature ranking between ML methods. A challenge of applying ML methods across heterogeneous NGS data is accounting for signal:noise. Here, signal validation across studies was shown by intersecting sets between top MRMR genes and DEGs from DESeq2 analysis. Non-intersecting sets introduce opportunity to explore genes relevant to differentiating space flight exposed groups and implementing ML methods across existing NGS datasets may overcome sample size limitations.

Machine Learning↗

Transcriptomics-based Machine Learning (ML) Analysis Predicts Space-Exposed Murine Livers

Limited sample sizes, high data dimensionality, and sensitivity to technical and biological variability of next generation sequencing (NGS), typically limits machine learning (ML) approaches in spaceflight studies that include radiation effects. However, pooling smaller studies while addressing intra- and inter-study variabilities allows for ML predictive modeling. Here, integration methods were applied to whole transcriptome shotgun sequencing (RNA-seq) data from six mouse liver GeneLab datasets (GLDS) (n ranging from 6 to 39 samples) from with a total of 81 spaceflight and ground-control samples to determine top features (i.e. genes) relevant to spaceflight including the effect of radiation exposure. RNASeq counts were normalized for each study, then merged and scaled across all datasets. Data dimensionality was reduced using a minimum redundancy maximum relevance (MRMR) methodology. Redundancy and relevance were computed using the Pearson correlation and F-statistic, respectively. The top 100 MRMR features were used to predict spaceflight vs. ground-control samples using Random Forest (RF), Support Vector Machine (SVM), and Linear Discriminant Analysis (LDA) classifiers with 5-fold cross validation (CV). Principal component analysis (PCA) on the complete feature set versus the MRMR features shows separation between spaceflight samples and ground controls (Figure 1A). The ML-based gene sets were compared against differential gene expression results obtained with DESeq2 from individual GLDS. Using all features or randomly sampled subsets at matching set sizes with MRMR, a maximum classifier accuracy of 69% on the test set over 5 folds. For all classifiers, CV training using at least the top 30 MRMR genes show minimum 89% accuracy and 0.95 AUC value on the test set over 5 folds (Figure 1B). Baseline set analysis on differentially expressed genes (DEGs) identified using padj ≤ 0.05 show 295 DEGs that overlap at least two studies and 13 DEGs that overlap three studies (Figure 1C). Set analysis between the top 100 MRMR features and the DEGs showed 47 genes that overlap at least one study and 24 genes that overlap two studies. Over-representation analysis showed overlapping biological processes related to fatty acid and lipid metabolism which may indicate these processes in the response to spaceflight stressors. MRMR feature selection for the selected ML methods improve performance relative to a classifier built on all features or randomly sampled subsets. Permutation feature importance within the decorrelated MRMR features showed concordance in feature ranking between ML methods. A challenge of applying ML methods across heterogeneous NGS data is accounting for signal:noise. Here, signal validation across studies was shown by intersecting sets between top MRMR genes and DEGs from DESeq2 analysis. Non-intersecting sets introduce opportunity to explore genes relevant to differentiating space flight exposed groups and implementing ML methods across existing NGS datasets may overcome sample size limitations.

Machine Learning↗

Transcriptomics-based Machine Learning Analysis Predicts Space-Exposed Murine Livers

Limited sample sizes, high data dimensionality, and sensitivity to technical and biological variability of next generation sequencing (NGS), typically limits machine learning (ML) approaches in spaceflight studies that include radiation effects. However, pooling smaller studies while addressing intra- and inter-study variabilities allows for ML predictive modeling. Here, integration methods were applied to whole transcriptome shotgun sequencing (RNA-seq) data from six mouse liver GeneLab datasets (GLDS) (n ranging from 6 to 39 samples) from with a total of 81 spaceflight and ground-control samples to determine top features (i.e. genes) relevant to spaceflight including the effect of radiation exposure. RNASeq counts were normalized for each study, then merged and scaled across all datasets. Data dimensionality was reduced using a minimum redundancy maximum relevance (MRMR) methodology. Redundancy and relevance were computed using the Pearson correlation and F-statistic, respectively. The top 100 MRMR features were used to predict spaceflight vs. ground-control samples using Random Forest (RF), Support Vector Machine (SVM), and Linear Discriminant Analysis (LDA) classifiers with 5-fold cross validation (CV). Principal component analysis (PCA) on the complete feature set versus the MRMR features shows separation between spaceflight samples and ground controls (Figure 1A). The ML-based gene sets were compared against differential gene expression results obtained with DESeq2 from individual GLDS. Using all features or randomly sampled subsets at matching set sizes with MRMR, a maximum classifier accuracy of 69% was shown on the test set over 5 folds. For all classifiers, CV training using at least the top 30 MRMR genes show minimum 89% accuracy and 0.95 AUC value on the test set over 5 folds (Figure 1B). Baseline set analysis on differentially expressed genes (DEGs) identified using padj ≤ 0.05 show 295 DEGs that overlap at least two studies and 13 DEGs that overlap three studies (Figure 1C). Set analysis between the top 100 MRMR features and the DEGs showed 47 genes that overlap at least one study and 24 genes that overlap two studies. Over-representation analysis showed overlapping biological processes related to fatty acid and lipid metabolism which may indicate these processes in the response to spaceflight stressors. MRMR feature selection for the selected ML methods improve performance relative to a classifier built on all features or randomly sampled subsets. Permutation feature importance within the decorrelated MRMR features showed concordance in feature ranking between ML methods. A challenge of applying ML methods across heterogeneous NGS data is accounting for signal:noise. Here, signal validation across studies was shown by intersecting sets between top MRMR genes and DEGs from DESeq2 analysis. Non-intersecting sets introduce opportunity to explore genes relevant to differentiating space flight exposed groups and implementing ML methods across existing NGS datasets may overcome sample size limitations.

Machine Learning↗

Gila Water Resources III - Modeling the Impacts of Post-fire Restoration Methods on Vegetation Recovery in the Gila National Forest

In recent years, wildfires in New Mexico’s Gila National Forest have become increasingly common and more severe. Wildfires can have powerful impacts on hydrology and soil stability, including erosion, flooding, and debris-flows that threaten lives and infrastructure downstream. Vegetation restoration treatments like seeding and mulching can mitigate these effects and facilitate ecosystem recovery. Understanding the effectiveness of various restoration methods is vital to planning a cost-effective and successful post-fire recovery strategy. The immediate response to a fire on US Forest Service land is coordinated by a Burned Area Emergency Response (BAER) team, a group responsible for mitigating immediate post-fire risks to human life, property, and critical natural and cultural resources. This study created a proof-of-concept methodology for a decision-support tool designed to help BAER teams identify the restoration treatments most likely to succeed in a given burned area. Leveraging random forest regression, Google Earth Engine, and Landsat 7 and 8 Earth observations, this study modeled vegetation recovery after the 2013 Silver Fire for seeded areas, seeded/mulched areas, and untreated areas. Treatment type and initial burn severity were the largest drivers of vegetation recovery across the landscape. Seeded/mulched areas showed higher recovery levels than untreated areas three months post-fire, but by four years post-fire, treated and untreated areas displayed similar recovery levels. To produce a robust predictive tool for the Gila National Forest, the model should be trained on many more fires and incorporate post-fire weather conditions into the process. Such a model will help partners ensure efficient resource use and plan effective post-fire restoration strategies.

DEVELOP Project Summary↗

Automatic learning-based beam angle selection for thoracic IMRT

Purpose: The treatment of thoracic cancer using external beam radiation requires an optimal selection of the radiation beam directions to ensure effective coverage of the target volume and to avoid unnecessary treatment of normal healthy tissues. Intensity modulated radiation therapy (IMRT) planning is a lengthy process, which requires the planner to iterate between choosing beam angles, specifying dose–volume objectives and executing IMRT optimization. In thorax treatment planning, where there are no class solutions for beam placement, beam angle selection is performed manually, based on the planner’s clinical experience. The purpose of this work is to propose and study a computationally efficient framework that utilizes machine learning to automatically select treatment beam angles. Such a framework may be helpful for reducing the overall planning workload. Methods: The authors introduce an automated beam selection method, based on learning the relationships between beam angles and anatomical features. Using a large set of clinically approved IMRT plans, a random forest regression algorithm is trained to map a multitude of anatomical features into an individual beam score. An optimization scheme is then built to select and adjust the beam angles, considering the learned interbeam dependencies. The validity and quality of the automatically selected beams evaluated using the manually selected beams from the corresponding clinical plans as the ground truth. Results: The analysis included 149 clinically approved thoracic IMRT plans. For a randomly selected test subset of 27 plans, IMRT plans were generated using automatically selected beams and compared to the clinical plans. The comparison of the predicted and the clinical beam angles demonstrated a good average correspondence between the two (angular distance 16.8° ± 10°, correlation 0.75 ± 0.2). The dose distributions of the semiautomatic and clinical plans were equivalent in terms of primary target volume coverage and organ at risk sparing and were superior over plans produced with fixed sets of common beam angles. The great majority of the automatic plans (93%) were approved as clinically acceptable by three radiation therapy specialists. Conclusions: The results demonstrated the feasibility of utilizing a learning-based approach for automatic selection of beam angles in thoracic IMRT planning. The proposed method may assist in reducing the manual planning workload, while sustaining plan quality.

60 APPLIED LIFE SCIENCES↗

Mapping Post-fire Conifer Regeneration using Snow-on Imagery

The 2000 Jasper Fire in the Black Hills of South Dakota was the largest wildfire to date in the region, burning over 83,000 acres of ponderosa pine forest. In collaboration with partners from the United States Forest Service (USFS) Black Hills Experimental Forest, USFS Rocky Mountain Research Station, and United States Geological Survey Geosciences and Environmental Change Science Center, we characterized post-fire forest regeneration within high severity burn patches. We accomplished this by implementing novel conifer detection techniques using a snow index mask to create a winter, snow-on image composite from Landsat 8 Operational Land Imager (OLI) and Sentinel-2 Multispectral Instrument (MSI) data. We utilized 2015 USFS stem maps of field-observed regeneration plots and ocularly sampled reforestation sites planted from 2001–2013. The field data and imagery were used to train a Random Forest (RF) model in Google Earth Engine. The RF model classified conifer regeneration density as low, medium, or high across the high-severity burn area with an overall accuracy of 81.3% for 2021. Approximately 45.9% of the high-severity burn area had low or no regeneration (0-40 trees per acre) 20 years post-fire. Given our partners' desire to find easily accessible low conifer regeneration zones, we identified 4,079 acres of priority planting sites that were within 1,500 feet of roads, had not been planted previously, and were larger than 50 acres. This method supports the use of snow-on imagery as a successful technique to identify conifer regeneration in a post-wildfire landscape.

Yeshey Seldon↗

Maya Forest Water Resources II: Mapping Inundation Below the Forest Canopy in the Maya Tri-National Forest

To monitor seasonal flooding within the tri-National Maya Forest the team completed the methodology started by the Summer 2021 term to analyze changes in inundation dynamic throughout 2017. The team analyzed inundation dynamics in Google Earth Engine (GEE) using Earth observation products from the Landsat 8 Operational Land Imager (OLI), Advanced Land Observing Satellite (ALOS) Phased Array type L-band Synthetic Aperture Radar (PALSAR) 2, and International Space Station (ISS) Global Ecosystem Dynamics Investigation LiDAR (GEDI). The team improved the landcover classification using the Random Forest algorithm in GEE by adding canopy height data derived from GEDI, elevation and slope data from Copernicus, and additional multi-spectral band ratios from Landsat 8. The pixel-based land cover classification produced an overall accuracy of 88%. Experiments measuring inundation extent using L-band SAR included comparing results with a priori knowledge, topography datasets, and auxiliary datasets. We iteratively tested and found threshold values for identifying forested inundation using the ratio for HH divided by HV. The resulting methodology and products helped end users from Belize’s Land Information Center (LIC) and Forest Department, Guatemala’s Center for Monitoring and Evaluation (CEMEC), and Mexico’s El Colegio de la Frontera Sur (ECOSUR) manage land and water resources and protect communities.

Stephanie Jiménez↗

Black Hills Wildfires: Mapping Post-fire Conifer Regeneration using Snow-on Imagery

The 2000 Jasper Fire in the Black Hills of South Dakota was the largest wildfire to date in the region, burning over 83,000 acres of ponderosa pine forest. In collaboration with partners from the United States Forest Service (USFS) Black Hills Experimental Forest, USFS Rocky Mountain Research Station, and United States Geological Survey Geosciences and Environmental Change Science Center, we characterized post-fire forest regeneration within high-severity burn patches. We accomplished this by implementing novel conifer detection techniques using a snow index mask to create a winter, snow-on image composite from Landsat 8 Operational Land Imager (OLI) and Sentinel-2 Multispectral Instrument (MSI) data. We utilized 2015 USFS stem maps of field-observed regeneration plots and ocularly sampled additional reforestation sites planted in 2001–2013. In Google Earth Engine (GEE), the field data and imagery were used to train a Random Forest (RF) model. The RF model classified 2021 conifer regeneration density as low, medium, or high across the high-severity burn area with an overall accuracy of 81.3%. Approximately 45.9% of the high-severity burn had low or no regeneration (0-40 trees per acre) 20 years post-fire. Given our partners' desire to find easily accessible low conifer regeneration zones, we identified 4,079 acres of priority planting sites that were within 1,500 feet of roads, had not been planted previously, and were larger than 50 acres. This method supports the use of snow-on imagery as a successful technique to identify conifer regeneration.

Casey Menick​↗

Black Hills Wildfires: Mapping Post-Fire Conifer Regeneration using Snow-On Imagery

The 2000 Jasper Fire in the Black Hills of South Dakota was the largest wildfire to date in the region, burning over 83,000 acres of ponderosa pine forest. In collaboration with partners from the United States Forest Service (USFS) Black Hills Experimental Forest, USFS Rocky Mountain Research Station, and United States Geological Survey Geosciences and Environmental Change Science Center, we characterized post-fire forest regeneration within high-severity burn patches. We accomplished this by implementing novel conifer detection techniques using a snow index mask to create a winter, snow-on image composite from Landsat 8 Operational Land Imager (OLI) and Sentinel-2 Multispectral Instrument (MSI) data. We utilized 2015 USFS stem maps of field-observed regeneration plots and ocularly sampled additional reforestation sites planted in 2001–2013. In Google Earth Engine (GEE), the field data and imagery were used to train a Random Forest (RF) model. The RF model classified 2021 conifer regeneration density as low, medium, or high across the high-severity burn area with an overall accuracy of 81.3%. Approximately 45.9% of the high-severity burn had low or no regeneration (0-40 trees per acre) 20 years post-fire. Given our partners' desire to find easily accessible low conifer regeneration zones, we identified 4,079 acres of priority planting sites that were within 1,500 feet of roads, had not been planted previously, and were larger than 50 acres. This method supports the use of snow-on imagery as a successful technique to identify conifer regeneration.

Casey Menick↗

Southern Rockies Western Slope Agriculture: Identifying Drivers of Rangeland Production for Drought Planning on the Western Slope of the Southern Rockies

Over the last decade, the southern Rocky Mountains of the United States have experienced increasingly severe and variable drought. Local ranchers and landowners have reported strain on their operations, citing decreasing forage production for their cattle and a need to adjust their business models, even considering abandoning their businesses altogether. The study identified Major Land Resource Area-48 (MLRA-48) and northwestern Colorado as the key region for analysis. NASA DEVELOP partnered with the BLM Colorado River Field Office, Colorado State University Extension, USDA Forest Service, and the National Drought Mitigation Center to address stakeholder concerns of the efficacy of existing remotely sensed rangeland production estimation platforms and explore possible early warning climatic indicators of drought. The study identified two key rangeland platforms, the Rangeland Production Monitoring Service (RPMS) and Rangeland Analysis Platform (RAP) and used in-situ data to statistically validate their efficacy. RAP outperformed RPMS in estimating in-situ biomass and was therefore used in our climate modeling. Our study performed a random forest analysis, sampling 1500 points across the study area, comparing monthly RAP biomass estimates to a variety of climatic variables, including mean precipitation, temperature, palmer drought severity index, snow water equivalent, wind speed and direction, and vapor pressure deficit. After analysis, our study determined that vapor pressure deficit is a key indicator in predicting forage production in MLRA-48. Our study recommends the use of RAP in estimating potential forage, with caution for its tendency to overestimate. Our climate analysis provided our partners with greater understanding of the influence of various climatic factors in determining forage production and allows them to assist landowners in planning for future drought.

Addie Gonzalez↗

Target Selection and Validation of DESI Quasars

The Dark Energy Spectroscopic Instrument (DESI) survey will measure large-scale structures using quasars as direct tracers of dark matter in the redshift range 0.9 < z < 2.1 and using Lyα forests in quasar spectra at z > 2.1. We present several methods to select candidate quasars for DESI, using input photometric imaging in three optical bands (g, r, z) from the DESI Legacy Imaging Surveys and two infrared bands (W1, W2) from the Wide-field Infrared Survey Explorer. These methods were extensively tested during the Survey Validation of DESI. In this paper, we report on the results obtained with the different methods and present the selection we optimized for the DESI main survey. The final quasar target selection is based on a random forest algorithm and selects quasars in the magnitude range of 16.5 < r < 23. Visual selection of ultra-deep observations indicates that the main selection consists of 71% quasars, 16% galaxies, 6% stars, and 7% inconclusive spectra. Using the spectra based on this selection, we build an automated quasar catalog that achieves a fraction of true QSOs higher than 99% for a nominal effective exposure time of ~1000 s. With a 310 deg -2 target density, the main selection allows DESI to select more than 200 deg -2 quasars (including 60 deg -2 quasars with z > 2.1), exceeding the project requirements by 20%. The redshift distribution of the selected quasars is in excellent agreement with quasar luminosity function predictions.

79 ASTRONOMY AND ASTROPHYSICS↗

Southern Rockies Western Slope Agriculture: Identifying Drivers of Rangeland Production for Drought Planning on the Western Slope of the Southern Rockies

Over the last decade, the southern Rocky Mountains of the United States experienced severe and variable drought. Local ranchers and landowners have reported strain on their operations, citing decreasing forage for their cattle and a need to adjust their business models. This study identified Major Land Resource Area-48 (MLRA-48) and northwestern Colorado as the key region for analysis. NASA DEVELOP partnered with the BLM Colorado River Field Office, Colorado State University Extension, USDA Forest Service, and the National Drought Mitigation Center to address concerns regarding the efficacy of remotely sensed rangeland production platforms and identify early warning climatic indicators of drought. The study identified two key platforms, The Rangeland Productivity Monitoring Service (RPMS) and Rangeland Analysis Platform (RAP), which use NASA Landsat 5 TM, Landsat 7 ETM+, Landsat 8 OLI, and Landsat 9 OLI-2 to estimate rangeland biomass. We regressed these with in situ biomass data to validate their efficacy and found that RAP was more effective than RPMS in estimating rangeland biomass, though it presents a tendency to overestimate. Our study performed a random forest analysis, comparing monthly RAP biomass estimates to a variety of climate variables, including mean precipitation, temperature, Palmer Drought Severity Index, snow water equivalent, snow persistence from Terra MODIS, wind speed and direction, and vapor pressure deficit. We determined that vapor pressure deficit and precipitation are key indicators in predicting forage production in MLRA-48. Our climate analysis provided our partners with greater understanding of the influence of various climate variables in determining rangeland production and allows them to assist land managers in drought mitigation.

remote sensing↗

California & Oregon Ecological Forecasting: Detecting and Forecasting Fog Occurrence, Frequency, and Change to Support Coast Redwood (Sequoia sempervirens) Habitat Assessments

Fog and low clouds play an important role in providing moisture to coastal ecosystems. Coast redwood (Sequoia sempervirens) forests are currently distributed along a narrow strip of coastline in California and Oregon and rely on the presence of marine fog for moisture availability during the dry season (June-October). Recent time series analyses presented an uncertain future of fog frequency; however, a decline in fog presence may have adverse effects on the coast redwood habitat. To support Save the Redwoods League, a non-profit organization dedicated to coast redwood forest management, the team analyzed hourly fog data from the Geospatial Operational Environmental Satellite 17 (GOES-17) Advanced Baseline Imager (ABI) and daily cloud cover data from the Moderate Resolution Imaging Spectroradiometer (MODIS) aboard the Terra satellite. To explore present day fog longevity, GOES-17 was utilized to map the number of fog hours per day for the 2019 and 2020 dry seasons. The MODIS cloud flag was used to map the presence or absence of daily fog, which was summarized to create a monthly fog frequency dataset and identify trends in fog presence between 2000-2020. Both datasets were used as inputs into the random forest machine learning algorithm to identify climatic drivers of fog presence and longevity over the landscape. The present-day models suggested that daily temperature difference is a driving force behind fog presence and longevity. Trends in fog presence from 2000-2020 indicated great interannual variability. Finally, fog presence was modeled under a 2080 climate projection to shed light on the future of fog presence under a projected warmer climate. Model results projected an overall decline in fog presence during the dry season in the 2080s. Decreased fog presence as a result of increased temperature difference under a warmer climate remains to be a topic of investigation as to the impact on future redwood habitat suitability.

DEVELOP Project Summary↗

California & Oregon Ecological Forecasting: Detecting and Forecasting Fog Occurrence, Frequency, and Change to Inform Coast Redwood (Sequoia sempervirens) Habitat Assessments

Fog and low clouds play an important role in providing moisture to coastal ecosystems. Coast redwood (Sequoia sempervirens) forests are currently distributed along a narrow strip of coastline in California and Oregon and rely on the presence of marine fog for moisture availability during the dry season (June-October). Recent time series analyses presented an uncertain future of fog frequency; however, a decline in fog presence may have adverse effects on the coast redwood habitat. To complement ongoing work by Save the Redwoods League, a non-profit organization dedicated to coast redwood forest management, the team analyzed hourly fog data from the Geospatial Operational Environmental Satellite 17 (GOES-17) Advanced Baseline Imager (ABI) and daily cloud cover data from the Moderate Resolution Imaging Spectroradiometer (MODIS) aboard the Terra satellite. To explore present day fog longevity, GOES-17 was utilized to map the number of fog hours per day for the 2019 and 2020 dry seasons. The MODIS cloud flag was used to map the presence or absence of daily fog, which was summarized to create a monthly fog frequency dataset and identify trends in fog presence between 2000-2020. Both datasets were used as inputs into the random forest machine learning algorithm to identify climatic drivers of fog presence and longevity over the landscape. The present-day models suggested that daily temperature difference is a driving force behind fog presence and longevity. Trends in fog presence from 2000-2020 indicated great interannual variability. Finally, fog presence was modeled under a 2080 climate projection to shed light on the future of fog presence under a projected warmer climate. Model results projected an overall decline in fog presence during the dry season in the 2080s. Decreased fog presence as a result of increased temperature difference under a warmer climate remains to be a topic of investigation as to the impact on future redwood habitat suitability.

DEVELOP Technical Paper↗

The M33 Synoptic Stellar Survey. II. Mira Variables

We present the discovery of 1847 Mira candidates in the Local Group galaxy M33 using a novel semi-parametric periodogram technique coupled with a random forest classifier. The algorithms were applied to ∼2.4 × 10{sup 5} I -band light curves previously obtained by the M33 Synoptic Stellar Survey. We derive preliminary period–luminosity relations at optical, near-infrared, and mid-infrared wavelengths and compare them to the corresponding relations in the Large Magellanic Cloud.

79 ASTRONOMY AND ASTROPHYSICS↗

Mapping tree height in complex terrain of northern China using ultra-high-resolution images

Tree height is a key parameter for estimating forest biomass and carbon sequestration. In recent years, notable progress has been made in mapping tree height using satellite imagery. However, existing tree height products show low accuracy in mountainous and complex terrains, and few studies typically addressed tree height estimations in mountain areas. This study examines the Mentougou district of Beijing, China, characterized by complex terrain and mountainous landscapes. We analyzed two methods for estimating tree height: one using only spectral features and another combining spectral features with topographic factors (elevation, slope, aspect). We used 3-m resolution PlanetScope 8-band multispectral imagery, with 710 field-measured individual tree heights averaged to obtain 471 pixel-level tree height values as ground-truth, to develop tree height prediction models using eXtreme Gradient Boosting (XGBoost), Random Forest (RF), and Gradient Boosting Machine (GBM) models. The results show that the XGBoost model consistently presented the highest accuracy for both methods evaluated. Specifically, the XGBoost model that combined spectral data with elevation and slope variables with an R² of 0.75 and an RMSE of 2.69 m. Using the XGBoost model, we generated the tree height map for the Mentougou area at 3 m resolution, showing tree heights ranging from 0.5 to 30.4 m, and the model’s prediction error standard deviations ranged from 2.50 to 4.71 m, indicating reliable performance across varied terrain. Additionally, we compared and evaluated the global tree height products, identifying limitations in the accuracy within complex terrains. This study demonstrates the potential for accurately predicting tree heights by combining high-resolution multispectral satellites with a terrain factor modeling approach.

Complex terrain↗

Modeling the topographic influence on aboveground biomass using a coupled model of hillslope hydrology and ecosystem dynamics

Abstract. Topographic heterogeneity and lateral subsurface flow at the hillslope scale of ≤1 km may have outsized impacts on tropical forest through their impacts on water available to plants under water-stressed conditions. However, vegetation dynamics and finer-scale hydrologic processes are not concurrently represented in Earth system models. In this study, we integrate the Energy Exascale Earth System Model (E3SM) land model (ELM) that includes the Functionally Assembled Terrestrial Ecosystem Simulator (FATES), with a three-dimensional hydrology model (ParFlow) to explicitly resolve hillslope topography and subsurface flow and perform numerical experiments to understand how hillslope-scale hydrologic processes modulate vegetation along water availability gradients at Barro Colorado Island (BCI), Panama. Our simulations show that groundwater table depth (WTD) can play a large role in governing aboveground biomass (AGB) when drought-induced tree mortality is triggered by hydraulic failure. Analyzing the simulations using random forest (RF) models, we find that the domain-wide simulated AGB and WTD can be well predicted by static topographic attributes, including surface elevation, slope, and convexity, and adding soil moisture or groundwater table depth as predictors further improves the RF models. Different model representations of mortality due to hydraulic failure can change the dominant topographic driver for the simulated AGB. Contrary to the simulations, the observed AGB in the well-drained 50 ha forest census plot within BCI cannot be well predicted by the RF models using topographic attributes and observed soil moisture as predictors, suggesting other factors such as nutrient status may have a larger influence on the observed AGB. The new coupled model may be useful for understanding the diverse impact of local heterogeneity by isolating the water availability and nutrient availability from the other external and internal factors in ecosystem modeling.

54 ENVIRONMENTAL SCIENCES↗