Land use, land cover changes, and ecosystem-specific management practices are increasingly recognized for their roles in mediating the climatic effects on ecosystem structure and function. As demonstrated by some scholars, human activities can influence C fluxes and storage far more than climatic changes (IPCC 2014).
Our understanding and forecasting of ecosystem C fluxes cannot rely solely on conventional biophysical regulations at any scale, from the local ecosystem to the globe. We must quantify the magnitude of the C fluxes from managed ecosystems and landscapes over the lifetime of the C cycle and deduct the various energy inputs during management from the amount of C sequestered by an ecosystem (West & Marland 2003). For example, conventional crop management often includes tillage, fertilization, irrigation, applications of pesticides and herbicides, harvesting, transportation to the market, land conversion, etc. All of these activities require a CO2-equivalent (CO2eq) amount of energy (“social C flux”) to offset the actual amount of C sequestered by the ecosystems and landscapes. A complete life cycle assessment (LCA) is needed to account for the actual sequestration strength at different spatial and temporal scales.
Our overall objective is to quantify the landscape-scale C fluxes at annual scale of both managed agricultural-forest landscapes and people, using the Kalamazoo watershed in southwestern Michigan as our testbed.
Questions to be answered
- (1) What are the quantitative contributions of land cover change, specific management practices, and climate changes (means and extremes) to the social and physical C fluxes of managed ecosystems and landscapes?
- (2) What are the spatial and temporal changes of their contributions in managed agricultural-forest landscapes?
- (3) How will future land use changes (including alternative management practices) impact C sequestration in an upper, mid-latitude managed ecosystem?
Project members attended the Annual Legislative Breakfast March 20, 2017, hosted by the Kalamazoo Environmental Council.
Our overarching hypothesis is that social C flux is more responsible than physical C flux for the dynamics, and especially the uncertainty, of the cumulative CO2eq production of these intensively-managed landscapes. However, their proportions vary significantly among the landscapes and over history because of the great variations in land conversions, land use practices, climatic changes and extremes in the watershed.
Figure 1. (above) Proposed research components and their linkages for process-based predictions of the spatiotemporal changes in CO2eq production that will be quantified by estimating “social C flux”, and “physical C flux” at contrasting landscapes (i.e., different land cover compositions) within the Kalamazoo Watershed as well as the entire watershed (Fig. 2, below).
Life cycle assessment (LCA) will be employed for major patch types to quantify the C production at different temporal scales. The statistical downscaling modeling will be used to predict future local climate from Representative Concentration Pathways scenarios. Bayesian structural equation models (SEM) will be constructed to explore the contributions of climate change and human activities.
Figure 2. (above) Current land cover of the Kalamazoo Watershed (NLCD), which includes 127 sub-watersheds (USGS). The entire watershed will be examined for the changes of CO2eq during a 40-year period (1978–2018) using Landsat/Sentinel with the climate and human activities following our working framework (Fig. 1), while four contrasting landscapes will be quantified with high-resolution RS data and historical records and survey statistics over an 80-year period (1938–2018).
Task 1: Dynamics of Physical C Fluxes
Quantify the changes of the physical C flux on an annual scale, which will be converted to CO2eq, by integrating: (1) remotely-sensed land cover type and other surface properties; (2) geospatial records of climate, vegetation, soil, and management practices for model parameterization; (3) direct measurements of net ecosystem exchange of CO2 using EC flux towers for model validation; and (4) a customized ecosystem model (i.e. CLM).
Task 2: Dynamics of Social C Fluxes
Estimate the social C fluxes of major management practices for different land cover types by classifying historical land cover, identifying land ownership, and by surveying historical management practices of individual land-owners (parcel scale). Back-of-the-envelope calculations will be applied to scale up the CO2eq fluxes to the landscapes and the watershed.
Task 3: The dynamics and the regulations of CO2eq in time and space
Diagnose the mechanistic/empirical causal relationships based on biophysical models and SEM, and to quantify the ecosystem, landscape, and watershed C fluxes at multiple temporal scales and under alternative management/climate scenarios.
In alliance with NASA, our researchers openly share their data with the broader community. The latest information and resources from our project can be found here for exploration and education.
Data use policy
Use of this data is for non-profit, personal, or educational use only. To publish this data or apply it in other for-profit endeavours, please contact Dr. Jiquan Chen for information. For help interpreting the data, contact the author listed. Information provided here does not conflict with our committment to confidentiality or expose personal information from survey participants.
Physical Geography of the Kalamazoo Watershed
Kalamazoo Watershed Basin 2001–2011: (download 10.7MB) Shapefiles for Kalamazoo watershed and sub-watersheds used to create the following images below, which demonstrate changes between 2001–2011. Author: Dr. Ranjeet John
- Image features include:
- VENµS overpass
- Landsat WRS footprint
- NLCD 2001
- NLCD 2011
- River basin boundary
- Watershed boundaries
- Flux tower locations
- State boundary
- Country boundary
- NLDC 2001
- NLDC 2011
- Four study clusters
- 27x27 km VENµS swath
- Download files
Kalamazoo watershed landcover types 2001 (left) and 2011 (right). Land cover of the Kalamazoo Watershed (NLCD), which includes 127 sub-watersheds (USGS). Download Kalamazoo_2001_2011 for the complete list of shapefiles and data curated from sources including USGS, NLDC, VENµS, and more.
Kalamazoo Watershed stream gauge data: (download 77.4MB) The sub-hydrologic codes are Daily Data, this data is averaged already for the date. The other sub-hydrologic codes with (CC) listed after them are Current Conditions. This data is taken quarterly or half-hourly instead of daily and is much larger, but does not go back as far as the Daily Data. Some data gauges are listed in both the Daily Data and Current Conditions. In the qualification section, these are as follows: (A) Approved for publication, processing and review completed; and (P) provisional data subject to revision. Some of the data gathered here includes: gauge height, discharge, pH, conductivity turbidity, dissolved oxygen. Data source: USGS, links: USGS Water Resources Link, USGS Surface-Water Daily Data, and USGS Hydrologic region data .
Download graduate position details | Download Postdoc position details Positions starting fall 2017 are now closed. If you are interested in applying for spring 2018 or in the future, please contact Dr. Jiquan Chen with your CV, GRE scores, TOEFL (if applicable), and experience for more information.
|Dr. Jiquan Chen||Professor, PI
Michigan State University
|Dr. Kyla Dahlin||Assistant Professor, Co-PI
Michigan State University
|Dr. Ranjeet John||Research Associate, Co-PI
Michigan State University
|Dr. Phil Robertson||Collaborator
Kellogg Biological Station (GLBRC)
|Dr. Steve Hamilton||Collaborator
Kellogg Biological Station (LTER)
|Dr. David Lusch||Collaborator, RS/GIS
Michigan State University
|Lauren Cooper||Collaborator, Forest Climate Program Michigan State Universityfirstname.lastname@example.org|
|Dr. Arnon Karnieli||CollaboratorBen Gurion University, Israelemail@example.com|
|Giovanni Sylos Labini||CollaboratorPlanetek, Italy||website|