I have time series values of Normalised Differential Vegetation Index (NDVI) along stretches of River channel from 1990 to 2018. Please help me what kind of inferences could be drawn.
I think the inferences you draw from these NDVI measurements would really depend on the aims of your study. But before making inferences, I assume you are aware that there are slight spectral differences between the Landsat sensors such as ETM+ and OLI. Search for the article by Roy et al. (2016) 'Characterization of Landsat-7 to Landsat-8 reflective wavelength and normalized difference vegetation index continuity'. They say that statistical transformations are required in order to compare indices from two different landsat sensors before actually interpreting the results for time series applications. However, if you have already processed/transformed the time series data, you may want to take a look at NDVI Anomaly Maps such as the one illustrated and described in the attached link. This may give you some ideas and help you decide whether a similar approach is required for your study.
https://earthobservatory.nasa.gov/features/MeasuringVegetation/measuring_vegetation_3.php
Indeed they gives you quite different results, but usually, the difference is not really big. You can check how to do it following Fan and Liu (2016) and Fann and Liu (2017). "Furthermore, following the methodology proposed by Fan and Liu (2016) a total of 220 vegetation spectra were downloaded from the USGS spectral library (Kokaly et al., 2017). Comparing these spectra a linear interpolation was pursued to determine reflectance values at given wavelengths for both Landsat 5 TM and Landsat 8 OLI and quantify the spectral band adjustment factor (Fan and Liu, 2017)."
Here, you can check some manuscript using the same methodology.
Article Linking land use changes to variation in soil properties in ...
Article NDVI, 137Cs and nutrients for tracking soil and vegetation d...
I hope this will help you
For time series analysis, you can use, for example, moving average, harmonic analysis test and random walk test.
You can derived some statistical conclusions from these tests.
Thank you all for your answers.
could you all please elaborate... How could I relate my chart with flood events.
I think the inference to me w.r.t to NDVI is the crop cycle. You can easily identify the crop cycle and cropping pattern from this chart. the second thing to note is in between 2000-2005. There is only one cycle which shows that probably there was less vegetation during 2000-2005. you can explore the reason for this.
- Badges
- Science topic
- Similar topics
- Methodology
- Laboratory