Document Type

Dissertation - Open Access

Award Date

2016

Degree Name

Doctor of Philosophy (PhD)

Department

Geospatial Science and Engineering

Keywords

biogeochemical model, carbon cycle, land use change, remote sensing

Abstract

Understanding the major drivers of the cropland carbon fluxes is important for carbon management and greenhouse gas mitigation in agriculture. Past studies found that agricultural land-use and land-cover (LULC) changes, such as changes in cropland production technologies, tillage practices, and planted crop species, could have large impacts on carbon fluxes. However, the impacts remain highly uncertain at regional to global scales. Satellite remote sensing is commonly used to create products with geospatial information on LULC changes. This geospatial information can be integrated into biogeochemical models to simulate the spatial and temporal patterns of carbon fluxes. We used the General Ensemble Biogeochemical Modeling System (GEMS) to study LULC change impacts on cropland carbon fluxes in the Midwest USA. First we evaluated the impacts of LULC change on cropland net primary production (NPP) estimates. We found out the high spatial variability of cropland NPP across the study region was strongly related to the changes in crop species. Ignoring information about crop species distributions could introduce large biases into NPP estimates. We then investigated whether the characteristics of LULC change could impact the uncertainties of carbon flux estimates (i.e., NPP, net ecosystem production (NEP) and soil organic carbon (SOC)) using GEMS and two other models. The uncertainties of all three flux estimates were spatial autocorrelated. Land cover characteristics, such as cropland percentage, crop richness, and land cover diversity all showed statistically significant relationships with the uncertainties of NPP and NEP, but not with the uncertainties of SOC changes. The impacts of LULC change on SOC changes were further studied with historical LULC data from 1980 to 2012 using GEMS simulations. The results showed that cropland production increase over time from technology improvements had the largest impacts on cropland SOC change, followed by expansion of conservation tillage. This study advanced the scientific knowledge of cropland carbon fluxes and the impacts of various management practices over an agricultural area. The findings could help future carbon cycle studies to generate more accurate estimates on spatial and temporal changes of carbon fluxes.

Library of Congress Subject Headings

Landscape changes -- Middle West

Land use -- Middle West

Climatic changes -- Middle West

Soils -- Carbon content -- Middle West

Carbon cycle (Biogeochemistry)

Description

Includes bibliographical references (page 134-138)

Format

application/pdf

Number of Pages

159

Publisher

South Dakota State University

Rights

Copyright © 2016 Zhengpeng Li