Assessing biomass of diverse coastal marsh ecosystems using statistical and machine learning models

E. Glennie,A. Anyamba

Published 2018 in International Journal of Applied Earth Observation and Geoinformation

ABSTRACT

Abstract A time series of Advanced Very High Resolution Radiometer (AVHRR) derived normalized difference vegetation index (NDVI) data were compared to National Agricultural Statistics Service (NASS) corn yield data in the United States Corn Belt from 1982 to 2014. The main objectives of the comparison were to assess 1) the consistency of regional Corn Belt responses to El Nino/Southern Oscillation (ENSO) teleconnection signals, and 2) the reliability of using NDVI as an indicator of crop yield. Regional NDVI values were used to model a seasonal curve and to define the growing season — May to October. Seasonal conditions in each county were represented by NDVI and land surface temperature (LST) composites, and corn yield was represented by average annual bushels produced per acre. Correlation analysis between the NDVI, LST, corn yield, and equatorial Pacific sea surface temperature anomalies revealed patterns in land surface dynamics and corn yield, as well as typical impacts of ENSO episodes. It was observed from the study that growing seasons coincident with La Nina events were consistently warmer, but El Nino events did not consistently impact NDVI, temperature, or corn yield data. Moreover, the El Nino and La Nina composite images suggest that impacts vary spatially across the Corn Belt. While corn is the dominant crop in the region, some inconsistencies between corn yield and NDVI may be attributed to soy crops and other background interference. The overall correlation between the total growing season NDVI anomaly and detrended corn yield was 0.61(p = 0.00013), though the strength of the relationship varies across the Corn Belt.

PUBLICATION RECORD

  • Publication year

    2018

  • Venue

    International Journal of Applied Earth Observation and Geoinformation

  • Publication date

    2018-06-01

  • Fields of study

    Computer Science, Environmental Science

  • Identifiers
  • External record

    Open on Semantic Scholar

  • Source metadata

    Semantic Scholar

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-27 of 27 references · Page 1 of 1

CITED BY

Showing 1-23 of 23 citing papers · Page 1 of 1