Seyed Hamed Alemohammad

Water, Energy, and Carbon with Artificial Neural Networks (WECANN): a statistically based estimate of global surface turbulent fluxes and gross primary productivity using solar-induced fluorescence

A new global estimate of surface turbulent fluxes, latent heat flux (LE) and sensible heat flux (H), and gross primary production (GPP) is developed using a machine learning approach informed by novel remotely sensed Solar-Induced Fluorescence (SIF) and other radiative and meteorological variables. This is the first study to jointly retrieve LE, H and GPP using SIF observations. The approach uses an artificial neural network (ANN) with a target dataset generated from three independent data sources, weighted based on triple collocation (TC) algorithm. The new retrieval, named Water, Energy, and Carbon with Artificial Neural Networks (WECANN), provides estimates of LE, H and GPP from 2007 to 2015 at 1° × 1° spatial resolution and on monthly time resolution. The quality of ANN training is assessed using the target data, and the WECANN retrievals are evaluated using eddy covariance tower estimates from FLUXNET network across various climates and conditions. When compared to eddy covariance estimates, WECANN typically outperforms other products, particularly for sensible and latent heat fluxes. Analysing WECANN retrievals across three extreme drought and heatwave events demonstrates the capability of the retrievals in capturing the extent of these events. Uncertainty estimates of the retrievals are analysed and the inter-annual variability in average global and regional fluxes show the impact of distinct climatic events - such as the 2015 El Niño - on surface turbulent fluxes and GPP.

Tall Amazonian forests are less sensitive to precipitation variability

Climate change is altering the dynamics, structure and function of the Amazon, a biome deeply connected to the Earth’s carbon cycle. Climate factors that control the spatial and temporal variations in forest photosynthesis have been well studied, but the influence of forest height and age on this controlling effect has rarely been considered. Here, we present remote sensing observations of solar-induced fluorescence (a proxy for photosynthesis), precipitation, vapour-pressure deficit and canopy height, together with estimates of forest age and aboveground biomass. We show that photosynthesis in tall Amazonian forests, that is, forests above 30 m, is three times less sensitive to precipitation variability than in shorter (less than 20 m) forests. Taller Amazonian forests are also found to be older, have more biomass and deeper rooting systems1, which enable them to access deeper soil moisture and make them more resilient to drought. We suggest that forest height and age are an important control of photosynthesis in response to interannual precipitation fluctuations. Although older and taller trees show less sensitivity to precipitation variations, they are more susceptible to fluctuations in vapour-pressure deficit. Our findings illuminate the response of Amazonian forests to water stress, droughts and climate change.Tall trees are less sensitive to variation in precipitation than short trees, according to analyses of photosynthetic sensitivity to drought in tall and short Amazon forests. The results demonstrate higher resilience of tall trees to drought.

Quantifying Precipitation Uncertainty for Land Data Assimilation Applications

AbstractEnsemble-based data assimilation techniques are often applied to land surface models in order to estimate components of terrestrial water and energy balance. Precipitation forcing uncertainty is the principal source of spread among the ensembles that is required for utilizing information in observations to correct model priors. Precipitation fields may have both position and magnitude errors. However, current uncertainty characterizations of precipitation forcing in land data assimilation systems often do no more than applying multiplicative errors to precipitation fields. In this paper, an ensemble-based Bayesian method for characterization of uncertainties associated with precipitation retrievals from spaceborne instruments is introduced. This method is used to produce stochastic replicates of precipitation fields that are conditioned on precipitation observations. Unlike previous studies, the error likelihood is derived using an archive of historical measurements. The ensemble replicates are ge...

Evaluation of Long-Term SSM/I-Based Precipitation Records over Land

Therecordofglobalprecipitation mappingusingSpecialSensorMicrowaveImager(SSM/I)measurements now extends over two decades. Similar measurements, albeit with different retrieval algorithms, are to be used in the Global Precipitation Measurement (GPM) mission as part of a constellation to map global precipitation with a more frequent data refresh rate. Remotely sensed precipitation retrievals are prone to both magnitude (precipitation intensity) and phase (position) errors. In this study, the ground-based radar precipitation product from the Next Generation Weather Radar stage-IV (NEXRAD-IV) product is used to evaluate a new metric of error in the long-term SSM/I-based precipitation records. The new metric quantifies the proximity of two multidimensional datasets. Evaluation of the metric across the years shows marked seasonality and precipitation intensity dependence. Drifts and changes in the instrument suite are also evident. Additionally, the precipitation retrieval errors conditional on an estimate of background surface soil moisture are estimated. The dynamic soil moisture can produce temporal variability in surface emissivity, which is a source of error in retrievals. Proper filtering has been applied in the analysis to differentiate between the detection error and the retrieval error. The identification of the different types of errors and their dependence on season, intensity, instrument, and surface conditions provide guidance to the development of improved retrieval algorithms for use in GPM constellation-based precipitation data products.

Estimating surface soil moisture from SMAP observations using a Neural Network technique

A Neural Network (NN) algorithm was developed to estimate global surface soil moisture for April 2015 to March 2017 with a 2-3 day repeat frequency using passive microwave observations from the Soil Moisture Active Passive (SMAP) satellite, surface soil temperatures from the NASA Goddard Earth Observing System Model version 5 (GEOS-5) land modeling system, and Moderate Resolution Imaging Spectroradiometer-based vegetation water content. The NN was trained on GEOS-5 soil moisture target data, making the NN estimates consistent with the GEOS-5 climatology, such that they may ultimately be assimilated into this model without further bias correction. Evaluated against in situ soil moisture measurements, the average unbiased root mean square error (ubRMSE), correlation and anomaly correlation of the NN retrievals were 0.037 m3m-3, 0.70 and 0.66, respectively, against SMAP core validation site measurements and 0.026 m3m-3, 0.58 and 0.48, respectively, against International Soil Moisture Network (ISMN) measurements. At the core validation sites, the NN retrievals have a significantly higher skill than the GEOS-5 model estimates and a slightly lower correlation skill than the SMAP Level-2 Passive (L2P) product. The feasibility of the NN method was reflected by a lower ubRMSE compared to the L2P retrievals as well as a higher skill when ancillary parameters in physically-based retrievals were uncertain. Against ISMN measurements, the skill of the two retrieval products was more comparable. A triple collocation analysis against Advanced Microwave Scanning Radiometer 2 (AMSR2) and Advanced Scatterometer (ASCAT) soil moisture retrievals showed that the NN and L2P retrieval errors have a similar spatial distribution, but the NN retrieval errors are generally lower in densely vegetated regions and transition zones.

Reconstructed Solar‐Induced Fluorescence: A Machine Learning Vegetation Product Based on MODIS Surface Reflectance to Reproduce GOME‐2 Solar‐Induced Fluorescence

Abstract Solar‐induced fluorescence (SIF) observations from space have resulted in major advancements in estimating gross primary productivity (GPP). However, current SIF observations remain spatially coarse, infrequent, and noisy. Here we develop a machine learning approach using surface reflectances from Moderate Resolution Imaging Spectroradiometer (MODIS) channels to reproduce SIF normalized by clear sky surface irradiance from the Global Ozone Monitoring Experiment‐2 (GOME‐2). The resulting product is a proxy for ecosystem photosynthetically active radiation absorbed by chlorophyll (fAPARCh). Multiplying this new product with a MODIS estimate of photosynthetically active radiation provides a new MODIS‐only reconstruction of SIF called Reconstructed SIF (RSIF). RSIF exhibits much higher seasonal and interannual correlation than the original SIF when compared with eddy covariance estimates of GPP and two reference global GPP products, especially in dry and cold regions. RSIF also reproduces intense productivity regions such as the U.S. Corn Belt contrary to typical vegetation indices and similarly to SIF.

A global spatially Continuous Solar Induced Fluorescence (CSIF) dataset using neural networks

In this work the authors produce three datasets of CSIF by filling spatial and temporal gaps of SIF 5 soundings by OCO2 using MODIS surface reflectances and machine learning. The resulting datasets in 0.05deg and 4-day resolution represent gap-filled instantaneous SIF under cloud-free conditions, cloud-free SIF integrated to a daily value and daily SIF under all-sky conditions. To illustrate the advantages and the usefulness of these high-resolution datasets they compare to another downscaled fluorescence product (RSIF, based on GOME2 and different MODIS reflectance datasets), GOME2 SIF, 10 EC GPP and OCO2-SIF itself based on drought occurrences.

Statistical retrieval of surface and root zone soil moisture using synergy of multi-frequency remotely-sensed observations

Plants photosynthetic activity and transpiration are constrained by the amount of water available to them through roots (i.e. root zone soil moisture) as well as nutrient and atmospheric conditions. Therefore, to better understand the response of plants to different stress conditions, knowledge of root zone soil moisture is essential. However, current global satellites dedicated to soil moisture monitoring are limited to L-band frequencies that have a low (< 5cm) penetration depth. In this study, we implement a new root zone soil moisture retrieval algorithm that takes advantage of multi-frequency microwave observations to infer root soil moisture from L-band measurements and inspired by plant hydraulics. The algorithm is a statistical retrieval that uses a set of target data to train an artificial neural network. Results of applying the retrieval algorithm to one year of observations along with future validation measures is presented.

Characterizing vegetation and soil parameters across different biomes using polarimetric P-band SAR measurements

This study presents a quantitative analysis of vegetation and soil parameters retrieved from observations of an airborne P-band SAR instrument across nine different biomes in North America. These measurements are part of the NASAs AirMOSS mission, and data have been collected between 2012 and 2015. We use a three component decomposition algorithm to separate the contribution of surface and vegetation scattering, and subsequently retrieve surface and vegetation parameters. Applying the retrieval algorithm to data across all the campaign sites, we characterize the dynamics of the parameters across different North American biomes and assess their characteristic range.

Physically-based retrieval of SMAP active-passive measurements covariation and vegetation structure parameters

The NASA Soil Moisture Active Passive (SMAP) mission aims at producing high-resolution (9 km) global maps of surface soil moisture based on L-band radar and radiometer measurements. In this study, a physically-based retrieval of the active-passive covariation parameter β from one active-passive (single-pass) SMAP acquisition couple is proposed, circumventing empirical time-series regressions. The key to single-pass retrieval of β is the vegetation correction of the backscatter signal. This can be achieved by use of the measured cross-polarized backscatter signal and parameters appropriately describing the structure of the vegetation volume. These parameters can be derived from the observed Γ-parameters of the SMAP baseline algorithm enabling a fully SMAP data-driven, single-pass estimation of the covariation parameter β without any auxiliary information. Moreover, vegetation structural parameters, indicative of preferential vegetation shape and orientation, are retrieved using the observed Γ-parameters.