Using Diffuse Reflectance Spectroscopy to Quantify and Predict Soil Carbon Content in Agricultural Soils of Hawai'i

• Research Team
• Time
• Funding Source
• Summary
• Results

Research Team

PI:
Gregory L. Bruland, University of Hawai'i Manoa

Co-PIs:
Sabine Grunwald, Soil and Water Science Department, University of Florida
Jonathan Deenik, University of Hawai'i Manoa
Ray Uchida, University of Hawai'i Manoa

Post-Doc:
Meryl L. McDowell, University of Hawai'i Manoa

Time: 09/2009 to 09/2011

Funding Source: UU TSTAR-Pacific

Summary

In Hawai‘i there is an increasing demand for techniques that can quantify soil carbon (C) over space and time to assess changes in fertility and C sequestration with changes in management.  However, traditional analysis required for accurate assessment of soil C is so time-consuming and expensive that it limits the degree to which this variability can be characterized.  Thus, there is a need for new techniques to measure C.  Visible/near-infrared (VNIR)- and mid-infrared (MIR)-diffuse reflectance spectroscopy (DRS) has the potential to revolutionize soil monitoring by allowing for samples to be scanned rapidly, inexpensively, and non-destructively.  The spectra can be related to laboratory measurements of C with chemometrics.  This allows for better characterization of soil C across Hawai‘i’s agroecosystems.  Thus, our project objectives are to:

1. Employ VNIR/MIR-DRS to scan samples from the NRCS National Soils Database for which laboratory measurements of soil C are available
2. Model relationships between spectra and laboratory-measured C
3. Use chemometric models to predict C in freshly-collected samples from agricultural fields
4. Validate predictive models for use by UH Mānoa Agricultural Diagnostic Service Center (ADSC). 

The project tests two hypotheses:

1. Fundamental wavelength regions in the MIR range will produce better predictions of C than combination or overtone regions in the VNIR range
2. Subsetting the data (i.e. by texture, taxonomy) will result in better predictive models of C than will global datasets. 
   

Results

Peer-reviewed Publications: (in progress)

Oral and poster presentations:

McDowell M.L., G.L. Bruland, J.L. Deenik, S. Grunwal and R.S. Uchida. 2010. Diffuse reflectance spectroscopy for total carbon analysis of Hawaiian soils. American Geophysical Union (AGU) Meeting, San Francisco, CA, Dec. 13-17, 2010.

Bruland G.L., M.L. McDowell, J. Deenik, R. Uchida and S. Grunwald. 2010. Using Diffuse Reflectance Spectroscopy to Assess Soil Total Carbon in Agricultural Soils of Hawaii and Other Pacific Islands. ASA-CSSA-SSSA Meeting, Long Beach, CA, Oct. 31- Nov. 4, 2010.