Can Tree Phenology Be Observed from 700 km Away in Space?

Overview

The objective of this project was to create a phenology model by analyzing known vegetation pixels, to extract key transition dates from each year’s phenological data, and to compare them to recorded data from the USA National Phenology Network (USA-NPN). It also aimed to create annual phenology maps of the study area, Indianapolis, IN, from 2000 to 2014 using NDVI data that were derived from NASA’s MODIS (Moderate Resolution Imaging Spectroradiometer) imagery. It is hypothesized that a mathematical model can be built based on satellite-derived NDVI data to stimulate the phenology cycle of a study area and that it is feasible to use the developed model to extract key transition dates in the phenology cycle. To test the hypotheses, a methodology was created with the following procedures: extraction of vegetation pixel vectors; vegetation growth analysis; modeling phenological dates; NDVI image visualization; and observation-based transition dates. Results showed that vegetation growth analysis and annual phenology modeling provided reasonably accurate phenology metrics; however, the model-based and data-based phenology both showed notable discrepancies with the USA-NPN observation dates. Based on these results, my hypothesis was proven correct. For future studies, I plan to experiment with different types of fitting models and to develop a phone app to collect more observation data. Furthermore, I plan to identify factors that lead to differences in annual phenology and to use these factors to calibrate future models. These models could then be used to predict the response of vegetation growth to global/regional climate change.

Competition history

• AJAS 2017 Category not listed

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Earth and Environmental Sciences projects
• AJAS Earth and Environmental Sciences projects
• Every AJAS 2017 project