JohnHalleyGotway
commented
6 months ago Note that we may want to also add logic to METcalcpy to handle longterm climatology-based statistics. Would need a separate METcalcpy issue to do so. This was discussed with @anewman89 on April 5, 2024 at a Land Project meeting.
Describe the New Use Case
The Vegetation-Atmosphere Coupling (VAC) index as described in Zscheischler et al. (2015) and summarized by Paul Dirmeyer here.
The original VAC index is computed using anomalies of fraction of absorbed photosynthetically active radiation (FPAR) or evapotranspiration (ET), and 2 m temperature with the long-term mean seasonal cycle removed. FPAR and ET were from satellite (FPAR) or model/model-obs fusion products (ET). For this use case, we will focus on ET as FPAR is often prescribed in current and likely next generation land models for forecast applications.
The long-term mean data should be required to be user provided, but for the use case, we'll have to compute those ourselves. For observations, we can use ERA5 for 2 m temperature, and possibly FLUXCOM or GLEAM for 'observed' ET.
For the models, we will likely not have access to long-term datasets except for long-term climate simulation datasets. For operational forecasts, there are often only short test periods to draw from. The user will likely need to be tasked to provide what they consider an appropriate model long-term seasonal mean value to compute anomalies from.
Further, the original VAC index estimates percentile anomalies empirically from the long-term anomaly time series. The user may need to provide the observed and model percentile values (or fitted Gaussian distribution parameters?). An alternative would be to use just the anomalies in anomaly units, not percentiles, but that often has more noise than percentile based anomalies.
Finally, the visualization of this metric is often done in a 2-d scatter plot, with varying levels of color coding depending on user preference. We will likely need to coordinate with forthcoming MetPlotPy enhancements across TBD issues.
Use Case Name and Category
Vegetation-Atmosphere Coupling (VAC) index as a land-atmosphere coupling diagnostic
Input Data
Observed ET and 2 m temperature for evaluation period Modeled ET and 2 m temperature for evaluation period User provided long-term seasonal means for both obs and model User provided long-term percentiles for both obs and model
Acceptance Testing
Describe tests required for new functionality. As use case develops, provide a run time here
Time Estimate
3-4 days
Sub-Issues
Consider breaking the new feature down into sub-issues.
Relevant Deadlines
Approximately 6/30/2024
Funding Source
2785031 (Dyn Veg Drought)
Define the Metadata
Assignee
Labels
Projects and Milestone
Define Related Issue(s)
There will be coordination with MetPlotPy and possibly MetCalcPy for VAC plotting and computation respectively.
New Use Case Checklist
See the METplus Workflow for details.
feature_<Issue Number>_<Description>feature <Issue Number> <Description>