SPRITE

Enlarge

The temporal parallelism of the two pioneering European hyperspectral missions PRISMA (PRecursore IperSpettrale della Missione Applicativa, ASI, launch: March 22, 2019; Loizzo et al. 2018) and EnMAP (Environmental Mapping and Analysis Program, DLR, launch: April 1, 2022; Guanter et al. 2015, Chabrillat et al. 2024) is a real stroke of luck from the perspective of the hyperspectral data user community! While individual images from hyperspectral Earth observation systems already allow for considerable advances in the quantitative derivation of biophysical and biochemical variables, the great added value for sustainable land surface management only arises from the temporal repeatability of the spectroscopic images. This applies in particular to research and application in the field of vegetation remote sensing, and especially to agricultural applications where the quantitative observation of temporally dynamic, spatially heterogeneous, and above all non-linear processes (growth, nutrient uptake and transfer, disease and pest infestation, etc.) are of primary importance.

Each of the two current hyperspectral satellite-based imaging systems has only very limited imaging capacity. This is mainly due to the challenges involved in operating these complex systems, which are at the limits of what is technically feasible. At the same time, competition for the limited imaging capacity of both satellites is very high, since hyperspectral data, due to its generic physical nature—a broad range of the electromagnetic spectrum is continuously covered with the aid of a large number of narrow channels—is naturally usable for many and very diverse disciplines, from atmospheric research to geology to hydrology. Finally, the weather also determines the probability of success for hyperspectral satellite images in the optical wavelength range. The result of the above-mentioned limitations is that, to date, very few hyperspectral satellite images of the same area have been able to be taken repeatedly over time. Large parts of the data archives achieved to date consist of isolated images of different areas. Mapping of temporal dynamic changes on the Earth's surface has therefore been virtually impossible to date and has remained the preserve of operational multispectral Earth observation systems. Although these offer the advantage of greater repeatability over time, they cannot measure variables that are essential for sustainable management. One example of this is the measurement of nitrogen content in above-ground plant parts, as this requires continuous coverage of the short-wave infrared range in order to identify the flat and partially overlapping absorptions of protein-based plant components.

Answering questions about the usability of hyperspectral time series is particularly interesting in view of the preparation of the ESA CHIME mission (Copernicus Hyperspectral Imaging Mission for the Environment; Nieke et al. 2023), which is scheduled for launch in 2029 and will be capable of recording hyperspectral time series for the first time with a potential repeat rate of 11 days (with two identical satellites). The preparation of algorithms for processing hyperspectral time series can only be advanced through the synergistic combination of various current sensors, whereby the differences between the sensors must be overcome. A possible extension of the EnMAP mission could even theoretically bridge the time gap until the launch of CHIME.

The aim of the SPRITE (Synergistic PRisma and enmap Integrated Time-series for crop Evaluation) project is therefore to investigate the usability of harmonized multi-sensor hyperspectral time series for vegetation monitoring.