Assessing hyperspectral sensor capabilities: PRISMA vs EnMAP for crop type mapping in a semi-arid region
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Significant advancements are being made through the use of new-generation of hyperspectral sensors to enhance crop type mapping, especially in heterogeneous and fragmented agricultural landscapes. However, the availability and spatial-temporal coverage of hyperspectral satellites such as PRISMA (PRecursore IperSpettrale della Missione Applicativa) and EnMAP (Environmental Mapping and Analysis Program) can pose considerable challenges for crop type mapping. These include limitations in revisit frequency, narrow swath widths, cloud interference, and the relatively recent and still-developing global acquisition strategies, which may result in data gaps over specific agricultural regions during critical crop growth stages. This study presents a comprehensive comparison of EnMAP Level 2A (L2A) and PRISMA level 2D (L2D), acquired on the same day over a semi-arid region in Morocco. To do so, we evaluated two feature selection strategies: Stepwise Discriminant Analysis (SDA) and Recursive Feature Elimination (RFE) and benchmark the performance of four classification models, including Support Vector Machine (SVM), Random Forest (RF), CatBoost, and the transformer-based foundation model for tabular data (TabPFN). Sensor consistency was assessed using spectral correlation metrics and a vegetation index, showing strong agreement between PRISMA and EnMAP in the VNIR and SWIR-1 domains (R 2 > 0.96). Classification results demonstrated that TabPFN outperformed all other models, reaching 93 % overall accuracy with PRISMA and 90 % with EnMAP. These findings highlight the operational value of hyperspectral imaging and foundation models for accurate, scalable crop mapping in complex agricultural settings. The high level of agreement observed between PRISMA and EnMAP in this study further supports the feasibility of synergetic use of multiple hyperspectral missions. This opens the door to constructing dense, multi-sensor hyperspectral time series, which could significantly enhance temporal coverage and support continuous agricultural monitoring and crop dynamics analysis at large scales.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it