The Copernicus Hyperspectral Imaging Mission for the Environment (CHIME) is expected to deliver key information to support the EU policies on the management of natural resources. Once in orbit, CHIME mission is foreseen to provide systematic VNIR-SWIR hyperspectral images with high radiometric accuracy at high spatial resolution. In the framework of the CHIME preparatory activities, ESA established a collaboration with NASA/JPL, University of Zurich and Italian Space Agency (ASI) to acquire hyperspectral data with the Next Generation Airborne Visible Infrared Imaging Spectrometer (AVIRIS-NG) over several representative test site in Europe. The aim of this collaboration was the acquisition of PRISMA hyperspectral data and in situ measurements, synchronously to AVIRIS-NG overpasses, in order to simulate CHIME-like data and test some of the CHIME High Priority Prototype Product retrievals, as well as to support collaboration and synergies with current and planned hyperspectral missions.

On some sites, simultaneous PRISMA and AVIRIS-NG acquisitions were made coupling remote sensing with in-situ observations, both spectral reference measurements and bio-physical variables. Match-ups between the L2 products (surface reflectance) for the two sensors regarding imagery acquired in June 2021 over 4 CAL/VAL sites in Italy - Jolanda di Savoia (44.85N; 11.94E) and Braccagni (42.83N; 11.07E) (Cropland), Venezia (45.35N; 12.44E) (Coastal sea water) and Lago Trasimeno (43.12N; 12.13E) (Freshwaters) - and reference in-situ measurements are considered in this study. The high reliability of AVIRIS data, coupled with the large surface covered, enabled to substantially increase the statistical representativeness of the match-ups.

The comparison provided an evaluation of the quality of L2D PRISMA product on different surfaces. In Croplands, it was possible to highlight differences in discrepancies related to land use: minor discrepancies in the VIS-NIR regions on vegetation and a much larger difference for bare soils; higher discrepancies for wavelength greater than 2300nm. On Coastal and Freshwaters, the comparison of PRISMA products, both L1 and L2, with AVIRIS-NG is well promising as spectra are corresponding well for the whole VNIR range. Overall the best agreement is for turbid and shallow waters; in clearer waters some divergence were found although mostly related to the different illumination-viewing geometries between PRISMA and AVIRIS-NG.

Despite the focus of this study is on PRISMA, it might be relevant to note that for some cases (e.g. Freshwater, Cropland) few match-ups between AVIRIS-NG and PRISMA are also including DESIS which is further contributing to CHIME development.

Monitoring Non-Photosynthetically active Vegetation (NPV), as it plays an important role in the cycling of carbon, nutrients and water, is relevant to different studies including ecosystem dynamics, climate change, ecology, and hydrology, and hence is a topic of interest for remote sensing environmental applications.
In croplands NPV represents a key information in the field of sustainable agriculture, given that the crop residue (CR) management affects the agri-ecological functions of soil. A proposed conservation agropractice is to leave CR in field and perform minimum tillage.

In the perspective of monitoring CR presence and management from farm to regional scales, two main information are requested: i) recognition of spatial distribution of different land surface conditions (soil, vegetation and NPV both from CR and dead standing vegetation) at parcel level and ii) the characterisation of NPV classes in terms of abundance of carbon base constituent (CBC) on surface unit.

Some preliminary studies with PRISMA proved that the lignin-cellulose absorption band centered at 2100nm is apparent in such data and it is reliable for the detection of NPV, besides it is promising for the characterisation of CR abundance by spectral modelling (Pepe et al. 2020).

Given that the requirements for assessing crop residue cover and soil tillage activity (when) and intensity (which type) are: i) an accurate land use map (Daughtry et al. 2005); ii) the knowledge of surface conditions changes as related to the timing of tillage or planting (Zheng et al. 2012); a classification paradigm is proposed to map PRISMA data in terms of five different surface status categories: bare soil, crop residue, vegetation at emergence (including plant regrowing on crop residues), crop in vegetative stage (green vegetation) and senescence phase (dead (dying) standing vegetation).
To this purpose, the method previously proposed by (Pepe et al. 2020) is improved by extending the analysis to spectral intervals other than that of lignin-cellulose, including those of leaf chlorophyll pigments (centered around 690 nm) and water content (centered around 1200 nm). Such absorption bands, representing diagnostic features to assess the presence of the different surface category, are modelled by the Exponential Gaussian Optimization method (Pompilio et al. 2009, 2010). Parameters extracted from PRISMA spectra, from a supervised training set, are used for inferring classification rules using a decision tree approach. The training set comes from a reference imagery for which information on ground conditions from an intensive field campaign are available for the study area corresponding to a large farm (3800ha) in Jolanda di Savoia, Northern Italy.
The classification paradigm (spectral modelling and decision tree) is run at pixel level, afterwards the results are post-processed to obtain a final map at parcel level (which is the extent of interest).

The mapping approach is applied to a set of images acquired during two crop seasons (2019-2020 and 2020-2021) over the study area as the site belongs to the network of the PRISMA mission cal/val project (PRISCAV). A total of 12 images (6 per crop season), were available for the experiment.
The performance of the approach is quantitatively assessed by traditional statistics for the image where ground reference data exists and qualitatively evaluated in terms of crop conditions trajectories derived from time series analysis and compared to crop map information and management knowledge of the estate.

Results proved the method to be viable and reliable for identifying practices related to land management able to recognise existence and periods of crop residue presence and confirmed that PRISMA hyperspectral data are promising for monitoring and verification actions on the implementation of conservation agriculture. Moreover, even if the mission is not intended to be operative, the revisit time and tasking characteristics of PRISMA, seems actually not optimal but sufficient, to provide a number of cloud-free images during planting season useful for monitoring CR. Such results are also important in the perspective of the new and forthcoming operational hyperspectral missions such as DLR-ENMAP and ESA-CHIME.

Future studies will be devoted to evaluate the reliability and consistency of these spectroscopic approaches in the characterisation of CR abundance.

Imaging spectroscopy (IS) is a powerful tool for monitoring Earth surface properties. IS can provide important information on the state and dynamics of the global cryosphere. In particular, these data allow the retrieval of several physical properties of the surface such as albedo, snow/ice grain size, liquid water content, and concentration of light-absorbing particles (e.g., mineral dust, black carbon, cryospheric algae). The recent launch of the PRISMA mission (April 2019) opened interesting perspectives for the quantitative estimation of snow and ice properties from satellite IS data. In this contribution, we present results from research activities aimed at evaluating the quality of PRISMA products (L1 as Top-Of-Atmosphere Radiance, and L2D as Surface Reflectance) for studying the cryosphere. Furthermore, we present some preliminary results for the retrieval of snow and ice parameters in polar areas.
The calibration and validation activities were accomplished in predefined periods, which were representative of the surface condition during the season, i.e. fresh and aged snow. The study was performed in the European Alps making use of two test sites located at different altitudes: Torgnon (2160 m) and Plateau Rosa (3500 m). Field reflectance measurements were collected in both sites using the Spectral Evolution spectrometer, which operates in the 300-2500 nm wavelength domain. Atmospheric Aerosol Optical Thickness (AOT) was measured as well. At the Torgnon site, an automatic system for continuous monitoring of spectral reflectance in the visible to near-infrared (VNIR) provided additional field data to compare with PRISMA observations. Field spectra were propagated to Top-Of-Atmosphere with MODTRAN and then compared with L1 PRISMA products. L2 PRISMA products were validated by using a direct comparison with both field data and by an additional intercomparison with a different retrieval method based on Optimal Estimation and Sentinel-2 data. The agreement between the in situ measurements and satellite data is generally good, and for both L1 and L2 products the mean absolute difference is around 5%. Underestimation of radiance and reflectance for wavelengths below 500nm was observed both for fresh and aged snow.
A further preliminary analysis was also conducted regarding the retrieval of snow and ice parameters in polar glaciers, where we tested two different algorithms on rather flat areas. In particular, we analysed a PRISMA scene acquired in August 2020 over the “k-transect” (South-West Greenland), and another scene acquired in December 2020 over the Nansen Ice Sheet (East Antarctica). In both cases, we obtained reliable estimation of snow and ice parameters such as albedo, grain effective radius, liquid water content, and concentration of impurities and algae. Although the preliminary results are encouraging, further analyses are needed to validate these retrievals with field data.

Wildfires are natural phenomenon which both influence and are influenced by climate change. In the last few decades Earth Observation (EO) satellite have been used to analyze many fire characteristics, including: fire temperature, fire radiative power (FRP), smoke composition and vegetation mortality.
Active detection and monitoring of risk areas is becoming increasingly important to counteract severe and destructive landscape fires. Satellite-based remote sensing (RS) represents acost-effective way to detect, map, and investigate wildfires (Barmpoutis et al., 2020). EO satellites operating in the middle infrared (MIR) and thermal infrared (TIR) spectral band such as AVHRR, Modis, Sentinel 3, Landsat among many others are used to generate operational products (i.e., active fire detection, FRP).
Fire detection products based on hyperspectral (HS) EO satellite are challenged by their general lack of MIR and TIR spectral ranges, the limited revisit time, and limited task availability. However, HS imagery provides unique attributes in support to fire detection (Veraverbeke et al., 2018). Indeed, previous results based on EO-1 Hyperion have shown the HS potentialities for RS applications (Waigl et al., 2019, Amici S. and Piscini A. 2021). With the new PRISMA mission, and many similar ones to come soon, hyperspectral instruments can complement long-wave information useful to characterizing the whole continuum of landscape fire (pre- fire, active and post fire). This work will investigate how PRISMA HS images can be used to support fire detection and related crisis management. Here we present how different detection techniques (indexes and AI-based) have been tested on landscape fires observed by PRISMA.
First, we will start with a descriptive analysis of collected PRISMA images containing wildfires. This phase also leads to the identification of Hyperspectral Fire index for PRISMA (HFDI) and the definition of classification classes, e.g., hot spots, smoke plumes, burnt areas, and healthy vegetation (Griffin et al., 2000). Second, we will describe how deep learning classification models can be designed to perform semantic segmentation of input HS data, where an output image with metadata will be associated to each pixel of the input image. Finally, an estimation of the temperature is carried out by using a linear mixture model and evaluating the temperature of the emitting sources (i.e., the landscape fires) with a least square approach. A critical comparison of the retrieved temperature against ECOSTRESS and Landsat sensors is realized.
REFERENCES
Amici S. and Piscini A., 2021. Exploring PRISMA Scene for Fire Detection: Case Study of 2019 Bushfires in Ben Halls Gap National Park, NSW, Australia, MDPI Remote Sensing, 2021, 13(8), 1410; https://doi.org/10.3390/rs13081410.
Barmpoutis, P., Papaioannou, P., Dimitropoulos, K., Grammalidis, N., 2020. A review on early forest fire detection systems using optical remote sensing. Sensors (Switzerland).
Dennison, P. E., Charoensiri, K., Roberts, D. A., Peterson, S. H., Green, R. O., 2006. Wildfire temperature and land cover modeling using hyperspectral data. Remote Sensing of Environment.
Griffin, M. K., Hsu, S. M., Burke, H. h. K., Snow, J. W., 2000. Characterization and delineation of plumes, clouds, and fires in hyperspectral images. International Geoscience and Remote Sensing Symposium (IGARSS).
Guarini, R., Loizzo, R., Facchinetti, C., Longo, F., Ponticelli, B., Faraci, M., Dami, M., Cosi, M., Amoruso, L., De Pasquale, V., Taggio, N., Santoro, F., Colandrea, P., Miotti, E., Di Nicolantonio, W., 2018. PRISMA hyperspectral mission products. International Geoscience and Remote Sensing Symposium (IGARSS).
Loizzo, R., Daraio, M., Guarini, R., Longo, F., Lorusso, R., DIni, L., Lopinto, E., 2019. Prisma Mission Status and Perspective. International Geoscience and Remote Sensing Symposium (IGARSS).
Piscini, A., Amici, S., 2015. Fire detection from hyperspectral data using neural network approach. Remote Sensing for Agriculture, Ecosystems, and Hydrology XVII.
Spiller, D., Ansalone, L., Amici, S., Piscini, A., Mathieu, P. P., 2021. “Analysis and Detection of Wildfires by Using Prisma Hyperspectral Imagery”, ISPRS 2021.
Veraverbeke, S., Dennison, P., Gitas, I., Hulley, G., Kalashnikova, O., Katagis, T., Kuai, L., Meng, R., Roberts, D., Stavros, N., 2018. Hyperspectral remote sensing of fire: State-of-the-art and future perspectives.
Waigl, C. F., Prakash, A., Stuefer, M., Verbyla, D., Dennison, P., 2019. Fire detection and temperature retrieval using EO-1 Hyperion data over selected Alaskan boreal forest fires. International Journal of Applied Earth Observation and Geoinformation.

Waste management is nowadays considered as an important indicator of sustainable development closely intertwined with many interdependent and cross-border issues.
In Italy, a significant part of waste is still disposed of in landfills as an undifferentiated component. Therefore, the monitoring of landfills in support of their management and planning activities is of great importance. In fact, the European Directive 31/1999 / EC and the Italian law 36/2003 require long-term monitoring of various parameters (air, soil, water) from the opening of the landfill to the post-closure control period. In this perspective, it has already been amply demonstrated in previous projects and studies [1], [2], [3], [4], [5], [6], [7] and [8] that remote sensing can be useful in monitoring the impact, if present, of landfills on the surrounding environment, through the collection of remotely sensed information useful for the identification and classification of potentially contaminated areas using non-invasive methods.
For example, biogas leaks or leachate can produce effects at different spatial and temporal scales that require careful analysis to correctly interpret local environmental dynamics.
Many researchers have explored the possibilities offered by remote sensing in environmental analysis, in particular for:
- the classification and estimate of the quantity of waste stored in landfills;
- the identification of appropriate sites (geology and hydrology studies, waste transport, urban displacement planning);
- in situ management of the landfill (support for operations);
- monitoring the evolution of the landfill over time (compliance with procedures and regulations, prevention of pollution risk) [8];
- the identification of unauthorized landfills;
- identify biogas emissions [5];
- the estimate of leachate not captured [6];
- estimation of the generation and deposition of dust [7].

The main objective of the CLEAR-UP project (funded by ASI) concerns the use of PRISMA hyperspectral images for the study, development and implementation of indicators of the presence of pollutants in the soil and in the air close to areas affected by the presence of landfills. The availability of PRISMA hyperspectral images, with the limitations related to spatial resolution, makes it possible in principle to achieve this goal with unprecedented accuracy. The objective of the study concerns the possibility of:
- Detecting the presence of heavy metals in the soils in the area next to landfills;
- Identifying potentially harmful emissions (CH4, CO2, NOX) caused by spontaneous combustion and/or due to malicious behavior against the material in the landfill;
- Identifying the presence of stress conditions affecting the vegetation close to the area of the landfill;
- Determining the extent of the area possibly affected by the presence of the landfill.


References
1) Nas, B., Cay, T., İşcan, F., and Berktay, A., Selection of MSW landfill site for Konya, Turkey using GIS and multi-criteria evaluation. Environmental Monitoring and Assessment, 160(1-4), 491-500, 2010.
2) Ottavianelli, G., Synthetic Aperture Radar remote sensing for landfill monitoring. Ph.D. Thesis, Cranfield University, United Kingdom, pp. 298, 2007.
3) Schrapp, K. and Al-Mutairi, N., Associated health effects among residences near Jeleeb Al-Shuyoukh landfill. American Journal of Environmental Sciences, 6(2), pp. 184–190, 2010.
4) Shaker, A., Faisal, K., El-Ashmawy, N., and Yan, W.Y., Effectiveness of using remote sensing techniques in monitoring landfill sites using multi-temporal Landsat satellite data. Al-Azhar University Engineering Journal, 5(1), pp. 542-551, 2010.
5) Manzo, C., Studio di nuove tecnologie applicate alla individuazione e caratterizzazione delle emissioni di biogas da discarica. (PhD. Thesis. ed. Siena). Scuola Superiore Santa Chiara, University of Siena 2012, http://dx.doi.org/10.13140/2.1. 2433.5686
6) Slonecker, T., Fisher, G.B., Aiello, D.P., Haack, B., Visible and infrared remote imaging of hazardous waste: a review. Remote Sens. 2 (11), 2474–2508, 2010.
7) Stefanov, W. L., Ramsey, M.S., Christensen, P.R., Identification of fugitive dust generation, transport, and deposition areas using remote sensing. Environ. Eng. Geosci. 9(2), 151–165, 2003.
8) E.G. Cadau; C. Putignano; G. Laneve; R. Aurigemma; V. Pisacane; S. Muto; A. Tesseri; F. Battazza: Optical and SAR data synergistic use for landfill detection and monitoring. The SIMDEO project: Methods, products and results. – IGARSS – 13-18 July 2014, Quebec City.

Hyperspectral data, providing reflectance from visible to shortwave infrared wavelength, can greatly contribute to the retrieval of biophysical and biochemical vegetation traits, which are of high relevance for agricultural and ecological applications. In the framework of the Italian Space Agency project “Sviluppo di Prodotti Iperspettrali Prototipali Evoluti” (Contract ASI N. 2021-7-I.0), a prototype processor has been developed, exploiting PRISMA (PRecursore IperSpettrale della Missione Applicativa) imagery, for quantifying parameters such as Leaf Area Index (LAI), Fraction of Absorbed Photosyntetically Active Radiation (FAPAR), Fractional Vegetation Cover (FCOVER), Chlorophyll-a and Chlorophyll-b (Cab) useful for vegetation characterization.
In the scientific literature, the retrieval methods of vegetation traits are categorized into four groups: parametric regression, non-parametric regression, physically-based (including inversion of Radiative Transfer Models – RTMs – using numerical optimization and Look Up Tables –LUT- approaches), and hybrid regression methods. We have developed a hybrid method that invert of physical models through machine learning (ML) regression algorithms. In our method, the physical models, based on PROSAIL, and relating the vegetation physical parameters to the bottom of atmosphere reflectance, are used to generate simulated plant canopy spectral reflectances (from 400 to 2500 nm at 1 nm spectral resolution). Such simulated data, resampled to the PRISMA band configuration, are used to train the ML regression model.
A contamination with noise has been considered in order to improve the generalization capability of the models. In addition, a subspace of the feature space has been selected by means of dimensionality reduction techniques like PCA (Principal Component Analysis), in order to avoid correlated information that may result in suboptimal performances. Different machine learning algorithms, such as Random Forest, Support Vector Machine, Gaussian Process and Artificial Neural Network, have been evaluated and tested for the regression task.
The proposed approach allows retrieving vegetation indicators with lower computational time than other methodologies presented in the literature. In addition, it has a high power of generalization, thanks to the high representativeness of the training dataset, which has been generated taking into account different combinations of vegetation parameters and illumination/acquisition geometry configurations.
In order to demonstrate the capabilities of the prototype processor and to measure its performances, the trained models have been validated on several PRISMA data acquired over the Maccarese (Italy) study site with respect to ground data collected in situ, showing very promising results.