Martina Natali

Martina Natali

Auto-generated profile, based on publicly available information

Martina Natali

National Research Council of Italy, Unknown, Perugia, Italy [IT

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 December 2024 in Agricultural and Forest Meteorology
Verified
Journal Article
Open access
JKJerzy Piotr Kabala
Jerzy Piotr Kabala
Christian MassariChristian Massari
Christian Massari
Francesco NiccoliFrancesco Niccoli
Francesco Niccoli
MNM. Natali
M. Natali
FAF. Avanzi
F. Avanzi
GBG. Battipaglia
G. Battipaglia

20 January 2025
Verified
Conference Paper
Open access
JKJerzy Piotr Kabala
Jerzy Piotr Kabala
Christian MassariChristian Massari
Christian Massari
Francesco NiccoliFrancesco Niccoli
Francesco Niccoli
FAFrancesco Avanzi
Francesco Avanzi
MNMartina Natali
Martina Natali
Giovanna BattipagliaGiovanna Battipaglia
Giovanna Battipaglia
Transpiration accounts for more than a half of the evaporative water fluxes from land. However, several criticalities still exist in its quantification over the whole land surface. Sap-flux measurement is a widespread technique that allows to retrieve data, at a high temporal resolution, for individual plants or trees. Unfortunately, due to practical constraints the spatial extent of sap-flux data is limited, and the campaigns of measurement are most often limited to one or few vegetative seasons. Hydro-meteorological data can be obtained in a much simpler way than sap-flux, while the vegetation condition is monitored frequently and at a high spatial resolution and with a wide coverage by several remote sensing satellite missions; both this kind of data are related to the transpiration process: on one hand the meteorological forcings drive the evaporation process, while the vegetation exerts control on the stomatal resistance, in response to both the environmental condition as well as its own physiological conditions. Machine learning (ML) is a suitable methodology for extracting and reproducing complex patterns from data; and thus might be able to predict sap-flux based on its physical drivers or proxies.The objective of this research was to test three different ML algorithms (namely Regression Tree, Random Forest and XGBoost) on timeseries of transpiration based on sap-flux measurements taken in a Fagus sylvatica L. forest located in Southern Italy, during the 2021 and 2022 vegetative seasons, and to evaluate the performance of different vegetation indices (namely NDVI, EVI2 from Sentinel-2 and Cross-Ratio from Sentinel-1) in improving the prediction accuracy. As meteorological predictors Radiation, Air Temperature, Vapour Pressure Deficit, and Soil Moisture were selected. The accuracies obtained by training the algorithms on the meteorological dataset, were compared to those gained with the addition of the different vegetation indices.The results showed that the vegetation indices always improved the prediction accuracy. EVI2 was the most effective vegetation index, and this is the first study to show that the Sentinel-1 Cross-Ratio is a valuable predictor of vegetation transpiration. With respect to algorithm performance Random Forest and XGBoost outperformed the Regression Tree and showed comparable accuracies between them.

14 March 2026
Verified
Conference Paper
Open access
MNMartina Natali
Martina Natali
GLGabriëlle De Lannoy
Gabriëlle De Lannoy
Alessia FlamminiAlessia Flammini
Alessia Flammini
SSSusan Steele-Dunne
Susan Steele-Dunne
Christian MassariChristian Massari
Christian Massari
Synthetic Aperture Radar (SAR) satellites provide all-weather, day-night global Earth surface coverage, enabling continuous monitoring of ecosystems across multiple microwave bands. Over forested areas, SAR backscatter carries information on canopy water content, vegetation structure, and soil moisture. Microwave signals with shorter wavelengths interact with leaves and upper canopy layers, while longer wavelengths penetrate deeper and investigate branches, trunks, and soil. This makes SAR backscatter a proxy for retrieving ecosystem variables related to water and carbon cycles, such as soil moisture and biomass, which are critical inputs in data assimilation schemes for Earth system models. However, SAR applications over forests are limited by backscatter saturation over dense canopies and by the limited penetration depth of shorter wavelengths through vegetation. Understanding how these constraints vary across different microwave bands, SAR variables, and forest types is thus essential before implementing data assimilation experiments. This study explores the sensitivity of Sentinel-1 C-band (5.405 GHz, l ~ 5.55 cm) and SAOCOM L-band (1.275 GHz, l ~ 23.5 cm) SAR observations to soil moisture (SM), evaporation (ET), gross primary productivity (GPP), and Leaf Area Index (LAI), over Mediterranean forest sites. We analyze the sigma nought (σ0) backscattering coefficient in different polarizations (dual-pol for Sentinel-1, quad-pol for SAOCOM), along with its cross-ratio (σ0VH/ σ0VV), backscatter-incidence angle slope, and polarimetric decomposition parameters. We calculate the sensitivity of each parameter by computing linear regression against in-situ measurements of ecosystem variables. We also assess sensitivity changes across different acquisition geometries and timing (morning and evening overpasses), seasonality, and forest types. We consider three study sites in central and northern Italy, namely IT-BFt, IT-Cp2, and IT-SR2, which belong to the ICOS/FLUXNET network and are equipped with soil moisture probes and eddy covariance towers for water and carbon fluxes measurements. The three sites comprise both deciduous (Carpinus betulus, Quercus robur) and evergreen (Pinus pinea l., Quercus ilex) forests with diverse structural characteristics.C-band backscatter from Sentinel-1 exhibits saturation at two out of three study sites, particularly where canopies are the densest. Conversely, L-band backscatter shows higher sensitivity to soil moisture and vegetation growth. By characterizing the sensitivity of SAR parameters to geophysical variables, this study contributes to a better understanding of the potential of SAR retrievals in data assimilation experiments to improve predictions of hydrological and carbon fluxes over forested regions.

18 March 2025
Verified
Conference Paper
Open access
MNMartina Natali
Martina Natali
SMSara Modanesi
Sara Modanesi
GLGabrielle De Lannoy
Gabrielle De Lannoy
DSDomenico De Santis
Domenico De Santis
DDDaniela Dalmonech
Daniela Dalmonech
Christian MassariChristian Massari
Christian Massari
Land surface-atmosphere interactions are strongly influenced by vegetation, since the latter drives the exchange of energy, water and carbon at canopy level via transpiration and photosynthesis. These biochemical processes are related to both the stomatal response to meteorological variations (linking the canopy to the deepest soil layers), and the allocation of carbon in different parts of the plant such as roots and leaves. In recent years, the characterization of these processes has gained increasing attention in land surface models (LSMs), which are powerful tools that reproduce the soil-plant-atmosphere continuum and the mutual feedback of its components. Vegetation in LSMs is described either statically -- based on a prescribed vegetation climatology or cover -- or dynamically, that is, evolving in time its characteristics such as leaf area index and vegetation cover fraction, among the others. However, the dynamic simulation of vegetation is often simplified in LSMs with respect to state-of-the-art bio-geophysical and forest models. In the Noah Multi-Parametrization (Noah-MP, v. 4.0.1) LSM, multiple parametrizations are available for each individual sub-process scheme such as dynamic vegetation, runoff partitioning, groundwater recharge and radiative transfer through the canopy, among others. It is thus important to identify the land cover type, soil and climate characteristics of the specific study site and tailor the parametrization to find the “optimal” combination of sub-process schemes, i.e. the one which best reproduces in-situ observations. In this study, we evaluate point-scale simulations generated using different parametrizations of dynamic vegetation schemes within Noah-MP, run in offline mode within the NASA’s Land Information System (LIS). We compare the LSM results of gross primary productivity, soil moisture and evapotranspiration over several years between 2000 and 2023 to both ground-based estimates and remote sensing datasets derived from multiple observations and platforms such as MODIS, OCO-2, MSG and FLUXCOM. The study focuses on sites along the Italian peninsula, mostly forests, with croplands and grasslands as well, some of which are equipped with Eddy-covariance stations for carbon and water fluxes measurements and are included in the FLUXNET network. The sites are all natural, rain-fed ecosystems mostly located in drought-prone, Mediterranean regions. This study is meant to reveal previously neglected uncertainties in dynamic vegetation simulations, especially in dry regions, and to fine-tune the combination of sub-processes schemes in Noah-MP for future data assimilation experiments.

13 October 2026
Verified
Preprint
Open access
PDF file
Sara ModanesiSara Modanesi
Sara Modanesi
DDDaniela Dalmonech
Daniela Dalmonech
Alessio CollaltiAlessio Collalti
Alessio Collalti
GLGabrielle De Lannoy
Gabrielle De Lannoy
Tommaso MoramarcoTommaso Moramarco
Tommaso Moramarco
Christian MassariChristian Massari
Christian Massari
Land surface models (LSMs) explicitly represent dynamic vegetation processes and their coupling with the water and carbon cycles. Improving these interactions is crucial in Mediterranean ecosystems, where increasing drought frequency and severity can strengthen vegetation–water feedback. However, it remains unclear how improvements in vegetation representation propagate to other eco-hydrological processes. This study investigates the impact of satellite-constrained vegetation parameter optimization on the Noah-MP LSM, implemented in the NASA Land Information System (LIS). Key vegetation parameters are optimized against bias-corrected MODIS leaf area index (LAI) over predominantly forested Mediterranean ecosystems in central and southern Italy. Model performance is evaluated against satellite- and in situ-based datasets for gross primary productivity (GPP), evapotranspiration (ET), runoff (Q) and soil moisture (SM), including analyses during hydrological drought conditions and across different land-cover types. Vegetation calibration substantially corrects the representation of LAI, leading to marked improvements in GPP and consistent reductions in ET errors, although changes in ET temporal dynamics remain comparatively modest. Basin-scale Q simulations also improve across most catchments, particularly through reduced bias and during hydrological droughts, highlighting the role of vegetation in water partitioning under water-limited conditions. In contrast, SM shows only slight changes, suggesting that calibration primarily affects water flux partitioning rather than soil water storage. Calibration is most effective in forest-dominated ecosystems, whereas heterogeneous and human-influenced landscapes show weaker and more variable responses. Overall, constraining vegetation dynamics with satellite observations propagates beyond LAI itself, improving coupled water and carbon flux simulations and basin-scale hydrological predictions, particularly under water-scarcity conditions.

24 March 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
MNMartina Natali
Martina Natali
Sara ModanesiSara Modanesi
Sara Modanesi
DSDomenico De Santis
Domenico De Santis
GLGabrielle De Lannoy
Gabrielle De Lannoy
SSSusan Steele-Dunne
Susan Steele-Dunne
CMChristian Massari
Christian Massari
Irrigation is a significant component of the water cycle in agricultural regions: however, information about how much and where irrigation is applied is scarce and uncertain. Remote sensing-based products have been developed to retrieve irrigation across scales, but they exhibit limitations due to spatial uncertainties and confounding factors, particularly in highly fragmented agricultural landscapes. This paper presents a novel method to estimate irrigation at the field scale by integrating Sentinel-1 C-band Synthetic Aperture Radar (SAR) backscatter data and Sentinel-2 NDVI data within a modeling framework. The proposed approach involves the joint calibration of the Water Cloud Model (WCM) and a two-layer Soil Water Balance (SWB) model to simultaneously simulate backscatter, irrigation, surface and root-zone soil moisture, and other water balance components. The model is driven by reanalysis meteorological data, and selected parameters in the WCM and SWB are calibrated exclusively using Sentinel-1 backscatter data, thus making the framework independent of in situ ground measurements. The model is tested on 38 fields within an irrigation district in the Po Valley, Italy, during 2018, for which irrigation data are available as reference. It is shown that the capability to estimate irrigation events depends on the treatment of the SAR data: after bias-correcting Sentinel-1 observations, average time series correlations of 0.43 to 0.95 could be obtained for daily to bi-weekly irrigation estimates, respectively. The associated bias amounts to -0.01 mm/day and -0.04 mm/15 days, and less than 1\% for the annual estimates.