Data Analysis for Paper ONLY

[cehatch]

To do list for this paper ONLY [Data Processing and plots]:

1. Probe transect data:

• [x] Break into segments for analysis: TE AWC, TE Hummock, TE Undisturbed; TW Dry, TW Woods, TW West Bog, TW East Bog
• [x] Plot Autocorrelation for each segment
• [ ] Interpret Autocorrelation
• [ ] Meet with Bill Clement
• [ ] Try Wavelets??? Plot distance vs. moisture:
• [ ] TE AWC
• [ ] TE Hummock
• [ ] TE Undisturbed
• [x] TW All Plot distance vs. Deviation from the mean (see stats for numbers):
• [ ] TE AWC
• [ ] TE Hummock
• [ ] TE Undisturbed
• [ ] TW Dry, TW Woods
• [ ] TW West Bog
• [ ] TW East Bog

**2. Calibration: Compare gravimetric data to probe data***

• [x] 33 samples at Tidmarsh East
• [x] 71 samples at Tidmarsh East
• [x] calculate gravimetric moisture from weights
• [x] calculate probe moisture from mV
• [x] stats
• [x] 1:1 plots
• [x] RMSE calcs

3. Context: What were conditions like when we measured?

• [ ] Get background Climate Data (hourly from 2013-present)
• [ ] Pull data ~30 days prior to 7/22-23/2014; 7/26-8/3/2017; 7/10-12/2018
• [ ] Calculate cumulative prior precipitation beginning on the LAST day of sampling and working backwards to ~30 days prior to the first day of sampling for:
• [ ] 2014
• [ ] 2017
• [ ] 2018 PLOT cumulative prior precipitation for:
• [ ] 2014
• [ ] 2017
• [ ] 2018 What were the temperatures like? Find a good way to represent these ---
• [ ] maybe plot mean daily plus min/max temperatures during the month prior, and
• [ ] report the average for the day(s) of sampling?
• [ ] Was this a particularly hot or dry year? Are there almanac discussions about this?
• [ ] Try asking Mike Rawlins for a good way to assess this.

4. Comparisons: is East really wetter? How do we know?

• [ ] Interpretation of different kinds of plots: autocorrelation, deviation from a mean, etc.
• [x] TE Change plot from GSM data 2014 to 2107
• [x] TE Change plot from GSM data 2014 to 2108
• [x] TE Change plot from GSM data 2017 to 2108
• [x] TW Change plot from GSM data 2017 to 2108
• [ ] What else can we characterize? Should parse out other environments like grassy uplands, frequency and depth of inundated areas, etc. – how to plot? --> YES! DONE!
• [ ] Compare climate context to results from plots in this section
• [ ] Assess significance of change(s) numerically/ statistically

5. Other analyses we haven't thought of yet

• [ ] Are there other statistics we should be doing here? Talk to Bill?
• [ ] LIDAR: Create change in topography map for hummocky transect region???
• [ ] HYPROP Data?
• [ ] LOI information? (Organic content)
• [ ] Incorporate SMAP data into context??? (Ask Mike Cosh)

[cehatch]

@ItoErika 2018_02_07_Tidmarsh_Activities.xlsx