Survey Protocol Cards for Crop Maps

Zaytar, A., Tadesse, G.A., Robinson, C., Nair, S.S., Blasch, G., Degerickx, J., Subakanya, M., Laso Bayas, J.C. ORCID: https://orcid.org/0000-0003-2844-3842, Hacheme, G.Q., Becker-Reshef, I., Dodhia, R., & Lavista Ferres, J. (2026). Survey Protocol Cards for Crop Maps. IEEE Geoscience and Remote Sensing Letters 23 p. 2504404. 10.1109/LGRS.2026.3708987.

Full text not available from this repository.

Abstract

Crop type maps underpin food security decisions, yet their accuracy depends on label quality, which in turn depends on survey design choices made under tight budgets. Survey planners must allocate limited resources across GPS devices, stratification strategies, sample size, worker training, and verification protocols, but lack quantitative guidance on which investments yield quality crop maps. We address this gap by modeling the full chain from survey design to downstream crop detection accuracy: survey choices map to costs, costs constrain achievable label noise levels, and noise levels affect crop mapping performance. We implement 17 noise functions grounded in documented errors from the agricultural survey literature, and measure degradation on two datasets: EuroCrops and Zambia. Our experiments reveal that label verification matters far more than GPS accuracy: crop misidentification causes up to 99% F1 loss while 30-m GPS jitter causes only 4%. Dataset-specific noise-to-performance surrogate models achieve R-2=0.87 , enabling millisecond what-if queries-but cross-dataset transfer shows mixed results: Spearman rho=0.32 -0.60 indicates rankings transfer asymmetrically, and negative R-2 reveals degradation predictions fail across contexts. We package these findings into a programmable protocol card that optimizes survey design given budget constraints.

Item Type:	Article
Uncontrolled Keywords:	Labeling, Crops, Modeling, Protocols, Surveys, Noise, Training, Degradation, Sensitivity, Testing, Agriculture data collection, image classification, machine learning, remote sensing
Research Programs:	Advancing Systems Analysis (ASA) Advancing Systems Analysis (ASA) > Novel Data Ecosystems for Sustainability (NODES)
Depositing User:	Luke Kirwan
Date Deposited:	24 Jul 2026 07:21
Last Modified:	24 Jul 2026 07:21
URI:	https://pure.iiasa.ac.at/21758

Actions (login required)

View Item