arXiv subjects

Martin Claverie

Publications and source records attributed to Martin Claverie.

6 recordsLinked to original sources

arXiv2025

From Rows to Yields: How Foundation Models for Tabular Data Simplify Crop Yield Prediction

We present an application of a foundation model for small- to medium-sized tabular data (TabPFN), to sub-national yield forecasting task in South Africa. TabPFN has recently demonstrated superior performance compared to traditional machine learning (ML) models in various regression and classification tasks. We used the dekadal (10-days) time series of Earth Observation (EO; FAPAR and soil moisture) and gridded weather data (air temperature, precipitation and radiation) to forecast the yield of summer crops at the sub-national level. The crop yield data was available for 23 years and for up to 8 provinces. Covariate variables for TabPFN (i.e., EO and weather) were extracted by region and aggregated at a monthly scale. We benchmarked the results of the TabPFN against six ML models and three baseline models. Leave-one-year-out cross-validation experiment setting was used in order to ensure the assessment of the models capacity to forecast an unseen year. Results showed that TabPFN and ML models exhibit comparable accuracy, outperforming the baselines. Nonetheless, TabPFN demonstrated superior practical utility due to its significantly faster tuning time and reduced requirement for feature engineering. This renders TabPFN a more viable option for real-world operation yield forecasting applications, where efficiency and ease of implementation are paramount.

cs.AI↗

arXiv2024

From parcels to people: development of a spatially explicit risk indicator to monitor residential pesticide exposure in agricultural areas

The increase in global pesticide use has mirrored the rising demand for food over the last decades, resulting in a boost in crop yields. However, concerns about the impact of pesticides on biodiversity, ecosystems, and human health, especially for populations residing close to cultivated areas, are growing. This study investigates how exposure and possible risks to residents can be estimated at high spatial granularity based on plant protection product data. The complexities of such analysis were explored in France, where relevant data with good granularity are publicly available. Integrating sets of spatial datasets and exposure assessment methodologies, we have developed an indicator to monitor the levels of pesticide risk faced by residents. By spatialising pesticide sales data according to their authorization on specific crops, we developed a detailed map depicting potential pesticide loads at parcel level across France. This spatial distribution served as the basis for an exposure and risk assessment, modelled following the European Food Safety Authority's guidelines. Combining the risk map with population distribution data, we have developed an indicator that allows to monitor patterns in non-dietary exposure to pesticides. Our results show that in France, on average, 13% of people might be exposed to pesticides due to living in the proximity to treated crops. This exposure is in the lower range for 34%, moderate range for 40% and higher range for 25% of the exposed population. The risk evaluation is based on worst case assumptions and values should not be taken as a regulatory risk assessment but as indicator to use, for example, for monitoring time trends. The purpose of this indicator is to demonstrate that more granular pesticide data can improve risk reduction strategies. Harmonized and high-resolution data can help in identifying regions where to focus on sustainable farming.

q-bio.QM↗

arXiv2023

Earth Observation based multi-scale analysis of crop diversity in the European Union: first insights for agro-environmental policies

To understand the resilience of farms and the agricultural sector, as well as the provision of ecosystem services, we need to characterize and quantify crop diversity. Using a 10m resolution satellite-derived product, we created datasets of crop diversity across spatial and administrative scales for 27 EU countries and the UK in 2018. We define local crop diversity, or $\alpha$-diversity, at a 1km scale, corresponding to large or clusters of small-to-medium-sized farms. $\alpha$ crop diversities range from 2.3 to 4.4, with higher levels in systems with many small farms (averaging less than 10 ha). $\gamma$-diversity, the number and area of crops grown independently of location, increases from 2.85 at 1km to 3.86 at 10km, and levels off at 4.27 at 100km. These levels are higher than those reported in the U.S., possibly due to differences in farm structure and practices. $\beta$-diversity, the ratio of $\gamma$ and $\alpha$ diversities, measures the difference between agroecosystems and ranges from 1.2 to 2.3 across EU countries. We classify countries' crop diversities into four groups based on the magnitude and change of $\gamma$-diversity across scales, with implications for regional to national agro-environmental policy recommendations. Continental Copernicus crop type maps will enable temporal comparisons, and exploring ecosystem co-variates will deepen our understanding of the link between crop diversity and agro-ecosystem services.

cs.CE↗

arXiv2023

Boosting Crop Classification by Hierarchically Fusing Satellite, Rotational, and Contextual Data

Accurate in-season crop type classification is crucial for the crop production estimation and monitoring of agricultural parcels. However, the complexity of the plant growth patterns and their spatio-temporal variability present significant challenges. While current deep learning-based methods show promise in crop type classification from single- and multi-modal time series, most existing methods rely on a single modality, such as satellite optical remote sensing data or crop rotation patterns. We propose a novel approach to fuse multimodal information into a model for improved accuracy and robustness across multiple years and countries. The approach relies on three modalities used: remote sensing time series from Sentinel-2 and Landsat 8 observations, parcel crop rotation and local crop distribution. To evaluate our approach, we release a new annotated dataset of 7.4 million agricultural parcels in France and Netherlands. We associate each parcel with time-series of surface reflectance (Red and NIR) and biophysical variables (LAI, FAPAR). Additionally, we propose a new approach to automatically aggregate crop types into a hierarchical class structure for meaningful model evaluation and a novel data-augmentation technique for early-season classification. Performance of the multimodal approach was assessed at different aggregation level in the semantic domain spanning from 151 to 8 crop types or groups. It resulted in accuracy ranging from 91\% to 95\% for NL dataset and from 85\% to 89\% for FR dataset. Pre-training on a dataset improves domain adaptation between countries, allowing for cross-domain zero-shot learning, and robustness of the performances in a few-shot setting from France to Netherlands. Our proposed approach outperforms comparable methods by enabling learning methods to use the often overlooked spatio-temporal context of parcels, resulting in increased preci...

cs.CV↗

arXiv2023

Deep-Learning-based Change Detection with Spaceborne Hyperspectral PRISMA data

Change detection (CD) methods have been applied to optical data for decades, while the use of hyperspectral data with a fine spectral resolution has been rarely explored. CD is applied in several sectors, such as environmental monitoring and disaster management. Thanks to the PRecursore IperSpettrale della Missione operativA (PRISMA), hyperspectral-from-space CD is now possible. In this work, we apply standard and deep-learning (DL) CD methods to different targets, from natural to urban areas. We propose a pipeline starting from coregistration, followed by CD with a full-spectrum algorithm and by a DL network developed for optical data. We find that changes in vegetation and built environments are well captured. The spectral information is valuable to identify subtle changes and the DL methods are less affected by noise compared to the statistical method, but atmospheric effects and the lack of reliable ground truth represent a major challenge to hyperspectral CD.

cs.CV↗

arXiv2022

Multimodal Crop Type Classification Fusing Multi-Spectral Satellite Time Series with Farmers Crop Rotations and Local Crop Distribution

Accurate, detailed, and timely crop type mapping is a very valuable information for the institutions in order to create more accurate policies according to the needs of the citizens. In the last decade, the amount of available data dramatically increased, whether it can come from Remote Sensing (using Copernicus Sentinel-2 data) or directly from the farmers (providing in-situ crop information throughout the years and information on crop rotation). Nevertheless, the majority of the studies are restricted to the use of one modality (Remote Sensing data or crop rotation) and never fuse the Earth Observation data with domain knowledge like crop rotations. Moreover, when they use Earth Observation data they are mainly restrained to one year of data, not taking into account the past years. In this context, we propose to tackle a land use and crop type classification task using three data types, by using a Hierarchical Deep Learning algorithm modeling the crop rotations like a language model, the satellite signals like a speech signal and using the crop distribution as additional context vector. We obtained very promising results compared to classical approaches with significant performances, increasing the Accuracy by 5.1 points in a 28-class setting (.948), and the micro-F1 by 9.6 points in a 10-class setting (.887) using only a set of crop of interests selected by an expert. We finally proposed a data-augmentation technique to allow the model to classify the crop before the end of the season, which works surprisingly well in a multimodal setting.

cs.CV↗