A blended census and multiscale remote sensing approach to probabilistic cropland mapping in complex landscapes

Issamaldin Mohammed, M.T. Marshall*, C.A.J.M. de Bie, Lyndon Estes, A.D. Nelson

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Remote sensing data are used to map the extent of croplands. They are especially useful in sub-Saharan Africa (SSA) where landscapes are complex and farms are small, i.e. less than two ha. In this study, a hierarchical remote sensing approach was developed to estimate field fractions at 30 m spatial resolution in a highly fragmented agricultural region of Ethiopia. The landscape was stratified into crop production system (CPS) zones with ten-day SPOT Proba-V 1 km normalized difference vegetation index (NDVI) composites. The CPS zones were used to disaggregate agricultural census statistics to 1 km field fractions and mask “wet” and “dry” seasons. Long-term average wet-dry season NDVI and topographic information derived from 30 m Landsat-8 (OLI) surface reflectance and the SRTM digital elevation model were combined with 1 km field fractions in a Generalized Additive Model (GAM) to produce the field fractions. Sample dot grids were manually interpreted from very high-resolution DigitalGlobe imagery on the Google Earth platform for training and testing. The model yielded an Area Under the Curve (AUC) of 0.71 and R 2 of 0.65 in the holdout sample set. The high AUC reveals the model was effective at classifying 30 m pixels as “crop” or “not crop” while the high R 2 indicated leveraging at the extremes (100 and 0% probability), meaning at 30 m resolution, subpixel variations were difficult to discern. The improved model skill compared to previous cropland mapping studies using GAMs can be attributed to the stratification and decomposition of the Landsat time series using CPS-defined phenology. Additional remote sensing model inputs, such as Sentinel-1 radar backscatter and Sentinel-2 red-edge reflectance, could provide additional explanatory power. Wall-to-wall national coverage for agricultural production estimation or other food security related application could be achieved by manually digitizing additional sample data in other regions of Ethiopia or using existing crowd-sourced databases, such as Geo-Wiki.

Original languageEnglish
Pages (from-to)233-245
Number of pages13
JournalISPRS journal of photogrammetry and remote sensing
Volume161
Early online date29 Jan 2020
DOIs
Publication statusE-pub ahead of print/First online - 29 Jan 2020

Fingerprint

farmlands
census
crops
remote sensing
Remote sensing
Crops
crop production
Ethiopia
production system
normalized difference vegetation index
wet season
NDVI
Landsat
dry season
Proba
phenology
reflectance
SPOT (French satellite)
digital elevation models
Shuttle Radar Topography Mission

Keywords

  • Agricultural production
  • Landscape stratification
  • GAMs
  • NDVI
  • Proba-V
  • Landsat
  • ITC-ISI-JOURNAL-ARTICLE

Cite this

@article{b8eddc2ff3244739925215a55cadba64,
title = "A blended census and multiscale remote sensing approach to probabilistic cropland mapping in complex landscapes",
abstract = "Remote sensing data are used to map the extent of croplands. They are especially useful in sub-Saharan Africa (SSA) where landscapes are complex and farms are small, i.e. less than two ha. In this study, a hierarchical remote sensing approach was developed to estimate field fractions at 30 m spatial resolution in a highly fragmented agricultural region of Ethiopia. The landscape was stratified into crop production system (CPS) zones with ten-day SPOT Proba-V 1 km normalized difference vegetation index (NDVI) composites. The CPS zones were used to disaggregate agricultural census statistics to 1 km field fractions and mask “wet” and “dry” seasons. Long-term average wet-dry season NDVI and topographic information derived from 30 m Landsat-8 (OLI) surface reflectance and the SRTM digital elevation model were combined with 1 km field fractions in a Generalized Additive Model (GAM) to produce the field fractions. Sample dot grids were manually interpreted from very high-resolution DigitalGlobe imagery on the Google Earth platform for training and testing. The model yielded an Area Under the Curve (AUC) of 0.71 and R 2 of 0.65 in the holdout sample set. The high AUC reveals the model was effective at classifying 30 m pixels as “crop” or “not crop” while the high R 2 indicated leveraging at the extremes (100 and 0{\%} probability), meaning at 30 m resolution, subpixel variations were difficult to discern. The improved model skill compared to previous cropland mapping studies using GAMs can be attributed to the stratification and decomposition of the Landsat time series using CPS-defined phenology. Additional remote sensing model inputs, such as Sentinel-1 radar backscatter and Sentinel-2 red-edge reflectance, could provide additional explanatory power. Wall-to-wall national coverage for agricultural production estimation or other food security related application could be achieved by manually digitizing additional sample data in other regions of Ethiopia or using existing crowd-sourced databases, such as Geo-Wiki.",
keywords = "Agricultural production, Landscape stratification, GAMs, NDVI, Proba-V, Landsat, ITC-ISI-JOURNAL-ARTICLE",
author = "Issamaldin Mohammed and M.T. Marshall and {de Bie}, C.A.J.M. and Lyndon Estes and A.D. Nelson",
year = "2020",
month = "1",
day = "29",
doi = "10.1016/j.isprsjprs.2020.01.024",
language = "English",
volume = "161",
pages = "233--245",
journal = "ISPRS journal of photogrammetry and remote sensing",
issn = "0924-2716",
publisher = "Elsevier",

}

A blended census and multiscale remote sensing approach to probabilistic cropland mapping in complex landscapes. / Mohammed, Issamaldin ; Marshall, M.T.; de Bie, C.A.J.M.; Estes, Lyndon; Nelson, A.D.

In: ISPRS journal of photogrammetry and remote sensing, Vol. 161, 03.2020, p. 233-245.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - A blended census and multiscale remote sensing approach to probabilistic cropland mapping in complex landscapes

AU - Mohammed, Issamaldin

AU - Marshall, M.T.

AU - de Bie, C.A.J.M.

AU - Estes, Lyndon

AU - Nelson, A.D.

PY - 2020/1/29

Y1 - 2020/1/29

N2 - Remote sensing data are used to map the extent of croplands. They are especially useful in sub-Saharan Africa (SSA) where landscapes are complex and farms are small, i.e. less than two ha. In this study, a hierarchical remote sensing approach was developed to estimate field fractions at 30 m spatial resolution in a highly fragmented agricultural region of Ethiopia. The landscape was stratified into crop production system (CPS) zones with ten-day SPOT Proba-V 1 km normalized difference vegetation index (NDVI) composites. The CPS zones were used to disaggregate agricultural census statistics to 1 km field fractions and mask “wet” and “dry” seasons. Long-term average wet-dry season NDVI and topographic information derived from 30 m Landsat-8 (OLI) surface reflectance and the SRTM digital elevation model were combined with 1 km field fractions in a Generalized Additive Model (GAM) to produce the field fractions. Sample dot grids were manually interpreted from very high-resolution DigitalGlobe imagery on the Google Earth platform for training and testing. The model yielded an Area Under the Curve (AUC) of 0.71 and R 2 of 0.65 in the holdout sample set. The high AUC reveals the model was effective at classifying 30 m pixels as “crop” or “not crop” while the high R 2 indicated leveraging at the extremes (100 and 0% probability), meaning at 30 m resolution, subpixel variations were difficult to discern. The improved model skill compared to previous cropland mapping studies using GAMs can be attributed to the stratification and decomposition of the Landsat time series using CPS-defined phenology. Additional remote sensing model inputs, such as Sentinel-1 radar backscatter and Sentinel-2 red-edge reflectance, could provide additional explanatory power. Wall-to-wall national coverage for agricultural production estimation or other food security related application could be achieved by manually digitizing additional sample data in other regions of Ethiopia or using existing crowd-sourced databases, such as Geo-Wiki.

AB - Remote sensing data are used to map the extent of croplands. They are especially useful in sub-Saharan Africa (SSA) where landscapes are complex and farms are small, i.e. less than two ha. In this study, a hierarchical remote sensing approach was developed to estimate field fractions at 30 m spatial resolution in a highly fragmented agricultural region of Ethiopia. The landscape was stratified into crop production system (CPS) zones with ten-day SPOT Proba-V 1 km normalized difference vegetation index (NDVI) composites. The CPS zones were used to disaggregate agricultural census statistics to 1 km field fractions and mask “wet” and “dry” seasons. Long-term average wet-dry season NDVI and topographic information derived from 30 m Landsat-8 (OLI) surface reflectance and the SRTM digital elevation model were combined with 1 km field fractions in a Generalized Additive Model (GAM) to produce the field fractions. Sample dot grids were manually interpreted from very high-resolution DigitalGlobe imagery on the Google Earth platform for training and testing. The model yielded an Area Under the Curve (AUC) of 0.71 and R 2 of 0.65 in the holdout sample set. The high AUC reveals the model was effective at classifying 30 m pixels as “crop” or “not crop” while the high R 2 indicated leveraging at the extremes (100 and 0% probability), meaning at 30 m resolution, subpixel variations were difficult to discern. The improved model skill compared to previous cropland mapping studies using GAMs can be attributed to the stratification and decomposition of the Landsat time series using CPS-defined phenology. Additional remote sensing model inputs, such as Sentinel-1 radar backscatter and Sentinel-2 red-edge reflectance, could provide additional explanatory power. Wall-to-wall national coverage for agricultural production estimation or other food security related application could be achieved by manually digitizing additional sample data in other regions of Ethiopia or using existing crowd-sourced databases, such as Geo-Wiki.

KW - Agricultural production

KW - Landscape stratification

KW - GAMs

KW - NDVI

KW - Proba-V

KW - Landsat

KW - ITC-ISI-JOURNAL-ARTICLE

UR - https://ezproxy2.utwente.nl/login?url=https://doi.org/10.1016/j.isprsjprs.2020.01.024

UR - https://ezproxy2.utwente.nl/login?url=https://library.itc.utwente.nl/login/2020/isi/marshall_blen.pdf

U2 - 10.1016/j.isprsjprs.2020.01.024

DO - 10.1016/j.isprsjprs.2020.01.024

M3 - Article

VL - 161

SP - 233

EP - 245

JO - ISPRS journal of photogrammetry and remote sensing

JF - ISPRS journal of photogrammetry and remote sensing

SN - 0924-2716

ER -