Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Sangram Ganguly has a total of 91 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Deep Neural Network Cloud-Type Classification (DeepCTC) Model and Its Application in Evaluating PERSIANN-CCS

    by Vesta Afzali Gorooh, Subodh Kalia, Phu Nguyen, Kuo-Lin Hsu, Soroosh Sorooshian, Sangram Ganguly, Ramakrishna R. Nemani

    Satellite remote sensing plays a pivotal role in characterizing hydrometeorological components including cloud types and their associated precipitation. The Cloud Profiling Radar (CPR) on the Polar Orbiting CloudSat satellite has provided a unique dataset to characterize cloud types. However, data from this nadir-looking radar offers limited capability for estimating precipitation because of the narrow satellite swath coverage and low temporal frequency. We use these high-quality observations to build a Deep Neural Network Cloud-Type Classification (DeepCTC) model to estimate cloud types from multispectral data from the Advanced Baseline Imager (ABI) onboard the GOES-16 platform. The DeepCTC model is trained and tested using coincident data from both CloudSat and ABI over the CONUS region. Evaluations of DeepCTC indicate that the model performs well for a variety of cloud types including Altostratus, Altocumulus, Cumulus, Nimbostratus, Deep Convective and High clouds. However, capturing low-level clouds remains a challenge for the model. Results from simulated GOES-16 ABI imageries of the Hurricane Harvey event show a large-scale perspective of the rapid and consistent cloud-type monitoring is possible using the DeepCTC model. Additionally, assessments using half-hourly Multi-Radar/Multi-Sensor (MRMS) precipitation rate data (for Hurricane Harvey as a case study) show the ability of DeepCTC in identifying rainy clouds, including Deep Convective and Nimbostratus and their precipitation potential. We also use DeepCTC to evaluate the performance of the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Cloud Classification System (PERSIANN-CCS) product over different cloud types with respect to MRMS referenced at a half-hourly time scale for July 2018. Our analysis suggests that DeepCTC provides supplementary insights into the variability of cloud types to diagnose the weakness and strength of near real-time GEO-based precipitation retrievals. With additional training and testing, we believe DeepCTC has the potential to augment the widely used PERSIANN-CCS algorithm for estimating precipitation.


  • Open Access

    Multi-Scale Association between Vegetation Growth and Climate in India: A Wavelet Analysis Approach

    by Dawn Emil Sebastian, Sangram Ganguly, Jagdish Krishnaswamy, Kate Duffy, Ramakrishna Nemani, Subimal Ghosh

    Monsoon climate over India has high degree of spatio-temporal heterogeneity characterized by the existence of multi-climatic zones along with strong intra-seasonal, seasonal, and inter-annual variability. Vegetation growth of Indian forests relates to this climate variability, though the dependence structure over space and time is yet to be explored. Here, we present a comprehensive analysis of this association with quality-controlled satellite-based remote sensing dataset of vegetation greenness and radiation along with station based gridded precipitation datasets. A spatio-temporal time-frequency analysis using wavelets is performed to understand the relative association of vegetation growth with precipitation and radiation at different time scales. The inter-annual variation of forest greenness over the Tropical India are observed to be correlated with the seasonal monsoon precipitation. However, at inter and intra-seasonal scales, vegetation has a strong association with radiation in regions of high precipitation like the Western Ghats, Eastern Himalayas, and Northeast hills. Forests in Western Himalayas were found to be correlated more on the winter precipitation from western disturbances than the south west monsoon precipitation. Our results provide new and useful region-specific information for dynamic vegetation modelling in the Indian monsoon region that may further be used in understanding global vegetation-land-atmosphere interactions.


  • Open Access

    Conditional Generative Adversarial Networks (cGANs) for Near Real-Time Precipitation Estimation from Multispectral GOES-16 Satellite Imageries—PERSIANN-cGAN

    by Negin Hayatbini, Bailey Kong, Kuo-Lin Hsu, Phu Nguyen, Soroosh Sorooshian, Graeme Stephens, Charless Fowlkes, Ramakrishna Nemani, Sangram Ganguly

    In this paper, we present a state-of-the-art precipitation estimation framework which leverages advances in satellite remote sensing as well as Deep Learning (DL). The framework takes advantage of the improvements in spatial, spectral and temporal resolutions of the Advanced Baseline Imager (ABI) onboard the GOES-16 platform along with elevation information to improve the precipitation estimates. The procedure begins by first deriving a Rain/No Rain (R/NR) binary mask through classification of the pixels and then applying regression to estimate the amount of rainfall for rainy pixels. A Fully Convolutional Network is used as a regressor to predict precipitation estimates. The network is trained using the non-saturating conditional Generative Adversarial Network (cGAN) and Mean Squared Error (MSE) loss terms to generate results that better learn the complex distribution of precipitation in the observed data. Common verification metrics such as Probability Of Detection (POD), False Alarm Ratio (FAR), Critical Success Index (CSI), Bias, Correlation and MSE are used to evaluate the accuracy of both R/NR classification and real-valued precipitation estimates. Statistics and visualizations of the evaluation measures show improvements in the precipitation retrieval accuracy in the proposed framework compared to the baseline models trained using conventional MSE loss terms. This framework is proposed as an augmentation for PERSIANN-CCS (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Network- Cloud Classification System) algorithm for estimating global precipitation.


  • Open Access

    Constraints to Vegetation Growth Reduced by Region-Specific Changes in Seasonal Climate

    by Hirofumi Hashimoto, Ramakrishna R. Nemani, Govindasamy Bala, Long Cao, Andrew R. Michaelis, Sangram Ganguly, Weile Wang, Cristina Milesi, Ryan Eastman, Tsengdar Lee, et al.

    We qualitatively and quantitatively assessed the factors related to vegetation growth using Earth system models and corroborated the results with historical climate observations. The Earth system models showed a systematic greening by the late 21st century, including increases of up to 100% in Gross Primary Production (GPP) and 60% in Leaf Area Index (LAI). A subset of models revealed that the radiative effects of CO2 largely control changes in climate, but that the CO2 fertilization effect dominates the greening. The ensemble of Earth system model experiments revealed that the feedback of surface temperature contributed to 17% of GPP increase in temperature-limited regions, and radiation increase accounted for a 7% increase of GPP in radiation-limited areas. These effects are corroborated by historical observations. For example, observations confirm that cloud cover has decreased over most land areas in the last three decades, consistent with a CO2-induced reduction in transpiration. Our results suggest that vegetation may thrive in the starkly different climate expected over the coming decades, but only if plants harvest the sort of hypothesized physiological benefits of higher CO2 depicted by current Earth system models.


  • Open Access

    Quantifying Uncertainty in Discrete-Continuous and Skewed Data with Bayesian Deep Learning

    by Thomas Vandal, Evan Kodra, Jennifer Dy, Sangram Ganguly, Ramakrishna Nemani, Auroop R. Ganguly

    Deep Learning (DL) methods have been transforming computer vision with innovative adaptations to other domains including climate change. For DL to pervade Science and Engineering (S&E) applications where risk management is a core component, well-characterized uncertainty estimates must accompany predictions. However, S&E observations and model-simulations often follow heavily skewed distributions and are not well modeled with DL approaches, since they usually optimize a Gaussian, or Euclidean, likelihood loss. Recent developments in Bayesian Deep Learning (BDL), which attempts to capture uncertainties from noisy observations, aleatoric, and from unknown model parameters, epistemic, provide us a foundation. Here we present a discrete-continuous BDL model with Gaussian and lognormal likelihoods for uncertainty quantification (UQ). We demonstrate the approach by developing UQ estimates on `DeepSD', a super-resolution based DL model for Statistical Downscaling (SD) in climate applied to precipitation, which follows an extremely skewed distribution. We find that the discrete-continuous models outperform a basic Gaussian distribution in terms of predictive accuracy and uncertainty calibration. Furthermore, we find that the lognormal distribution, which can handle skewed distributions, produces quality uncertainty estimates at the extremes. Such results may be important across S&E, as well as other domains such as finance and economics, where extremes are often of significant interest. Furthermore, to our knowledge, this is the first UQ model in SD where both aleatoric and epistemic uncertainties are characterized.

  • Open Access

    Exploring Subpixel Learning Algorithms for Estimating Global Land Cover Fractions from Satellite Data Using High Performance Computing

    by Uttam Kumar, Sangram Ganguly, Ramakrishna R. Nemani, Kumar S Raja, Cristina Milesi, Ruchita Sinha, Andrew Michaelis, Petr Votava, Hirofumi Hashimoto, Shuang Li, et al.

    Land cover (LC) refers to the physical and biological cover present over the Earth’s surface in terms of the natural environment such as vegetation, water, bare soil, etc. Most LC features occur at finer spatial scales compared to the resolution of primary remote sensing satellites. Therefore, observed data are a mixture of spectral signatures of two or more LC features resulting in mixed pixels. One solution to the mixed pixel problem is the use of subpixel learning algorithms to disintegrate the pixel spectrum into its constituent spectra. Despite the popularity and existing research conducted on the topic, the most appropriate approach is still under debate. As an attempt to address this question, we compared the performance of several subpixel learning algorithms based on least squares, sparse regression, signal–subspace and geometrical methods. Analysis of the results obtained through computer-simulated and Landsat data indicated that fully constrained least squares (FCLS) outperformed the other techniques. Further, FCLS was used to unmix global Web-Enabled Landsat Data to obtain abundances of substrate (S), vegetation (V) and dark object (D) classes. Due to the sheer nature of data and computational needs, we leveraged the NASA Earth Exchange (NEX) high-performance computing architecture to optimize and scale our algorithm for large-scale processing. Subsequently, the S-V-D abundance maps were characterized into four classes, namely forest, farmland, water and urban areas (in conjunction with nighttime lights data) over California, USA using a random forest classifier. Validation of these LC maps with the National Land Cover Database 2011 products and North American Forest Dynamics static forest map shows a 6% improvement in unmixing-based classification relative to per-pixel classification. As such, abundance maps continue to offer a useful alternative to high-spatial-resolution classified maps for forest inventory analysis, multi-class mapping, multi-temporal trend analysis, etc.


  • Unmixing Algorithms

    by Uttam Kumar, Cristina Milesi, S. Kumar Raja, Ramakrishna Nemani, Sangram Ganguly, Weile Wang, Petr Votava, Andrew Michaelis, Saikat Basu

  • Deep Learning for Very High-Resolution Imagery Classification

    by Sangram Ganguly, Saikat Basu, Ramakrishna Nemani, Supratik Mukhopadhyay, Andrew Michaelis, Petr Votava, Cristina Milesi, Uttam Kumar

    This chapter demonstrates the use of a scalable machine learning algorithm using airborne imagery data acquired by the National Agriculture Imagery Program (NAIP) for the Continental United States (CONUS) at an optimal spatial resolution of 1 m. It focuses on leveraging the power of deep belief networks (DBNs) for the representation and classification of aerial imagery data acquired at very high spatial resolutions of the order of 1 m. The chapter provides a detailed description of the unsupervised segmentation phase using the SRM algorithm. It enumerates the key components of the feature extraction step and the learning phase based on DBN. The chapter illustrates the details of the CRF algorithm and the online update procedure for the training data. It provides an overview of the NASA Earth Exchange high-performance computing (NEX HPC) architecture. The chapter contains results and comparative studies with National Land Cover Database (NLCD), and light detection and ranging (LiDAR). It also provides details of the AWS infrastructure.


  • Open Access

    DeepSD: Generating High Resolution Climate Change Projections through Single Image Super-Resolution

    by Thomas Vandal, Evan Kodra, Sangram Ganguly, Andrew Michaelis, Ramakrishna Nemani, Auroop R Ganguly

    The impacts of climate change are felt by most critical systems, such as infrastructure, ecological systems, and power-plants. However, contemporary Earth System Models (ESM) are run at spatial resolutions too coarse for assessing effects this localized. Local scale projections can be obtained using statistical downscaling, a technique which uses historical climate observations to learn a low-resolution to high-resolution mapping. Depending on statistical modeling choices, downscaled projections have been shown to vary significantly terms of accuracy and reliability. The spatio-temporal nature of the climate system motivates the adaptation of super-resolution image processing techniques to statistical downscaling. In our work, we present DeepSD, a generalized stacked super resolution convolutional neural network (SRCNN) framework for statistical downscaling of climate variables. DeepSD augments SRCNN with multi-scale input channels to maximize predictability in statistical downscaling. We provide a comparison with Bias Correction Spatial Disaggregation as well as three Automated-Statistical Downscaling approaches in downscaling daily precipitation from 1 degree (~100km) to 1/8 degrees (~12.5km) over the Continental United States. Furthermore, a framework using the NASA Earth Exchange (NEX) platform is discussed for downscaling more than 20 ESM models with multiple emission scenarios.


  • Learning Sparse Feature Representations Using Probabilistic Quadtrees and Deep Belief Nets

    by Saikat Basu, Manohar Karki, Sangram Ganguly, Robert DiBiano, Supratik Mukhopadhyay, Shreekant Gayaka, Rajgopal Kannan, Ramakrishna Nemani