Back to search

Michigan State University

Integrated remote sensing and crop system modeling for precision agriculture across spatial and temporal scales

Abstract

dc:description

In light of global environmental change, population pressure, and food production demands, there is considerable value in mapping biogeographic crop niche and characterizing crop productivity at multiple scales to enhance the impact of agricultural improvement across Africa. Crop system research has advanced sustainable strategies for intensifying food production; however, questions regarding where to implement innovative technologies are largely unresolved.This dissertation focuses on four geographic questions: (1) Where is the fundamental climate niche of maize, pigeonpea, and sorghum across Africa? (2) Where are marginal lands in Malawi and what are the underlying drivers of marginality? (3) Based on the drivers of marginal maize production, what are geographic scaling options for integration of pigeonpea into maize-based cropping systems? (4) What spatial resolutions are effective for conducting precision agriculture at the farm scale in smallholder systems? Overarching themes within the geographic discipline such as the modifiable areal unit problem and ecological fallacy problem underpin this research. Marginal areas for maize are highlighted at the Africa and Malawi scales and overlain with the optimal climate niche for crops such as sorghum and pigeonpea that offer multiple ecosystem services (e.g., soil rehabilitation through nitrogen fixation). Crop productivity is evaluated at scales relative to policy making delineations in Malawi (i.e., country, district, and extension planning area) to disentangle heterogeneity at local scales that may appear homogeneous at broader scales. At the Malawi farm scale, this research included the use of a small unmanned aerial system (sUAS), national government satellites (e.g., Sentinel-2), and commercial satellites (e.g., SPOT 6). Spectral measurements of crop status were evaluated at multiple spatial resolutions (ranging from 0.07-20-m) to determine what spatial resolutions and what spectral indices are most effective for estimating crop yields and crop chlorophyll.Results of this research include high spatial resolution maps of maize, pigeonpea, and sorghum suitability across Africa, indicating that pigeonpea and sorghum occupy unique agroecological zones throughout the continent (e.g., sorghum in the Sahel region). Similarly, pigeonpea suitability in Malawi occupies a greater land area than the extent to which it is currently cultivated, demonstrating that integration into maize-based cropping systems, particularly where soil is marginal, can have beneficial scaling outcomes. For the smallholder farm scale, problems of clouds and satellite revisit rates have not yet been overcome for precision agriculture. In this regard, sUAS are a promising option for relating spectral signals to on-farm measurements of crop status. Evidence from drone flights conducted at two experimental farms in the central region of Malawi (Nyambi and Ntubwi) suggest that spatial resolutions closer to the plant scale (i.e., 14-27-cm) are most effective for relating spectral imagery to crop status. Moreover, the green normalized difference vegetation index (GNDVI) and green soil adjusted vegetation index (GSAVI) were consistently correlated with crop chlorophyll and yield, illustrating that a broad range of indices should be evaluated for precision agriculture.

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Peter, Bradley George
Contributors dc:contributor
  • Messina, Joseph
  • Zulu, Leo
  • Dahlin, Kyla
  • Snapp, Sieglinde

Subjects

dc:subject × 7

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivatives 4.0 International
Language dc:language
English

Identifiers

dc:identifier.*
Identifier
etd:48037
isbn:9781085727846
isbn:108572784X
oclc:1138555794
umi:22589958
local:Peter_grad.msu_0128D_17014

Chain of custody

source
Harvested from
Michigan State University
Base URL
d.lib.msu.edu/oai
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Peter, Bradley George. Integrated remote sensing and crop system modeling for precision agriculture across spatial and temporal scales. 2019.