ENVIRONMENTAL LAYERS MEETING IPLANT TUCSON Roundup Benoit Parmentier
What I have been doing working on: 1)Visualization of RMSE fit for Geographically Weighted Regression Writing a code in R to visualize the RMSE using -Stations location -Kriged error surface from stations 2)Producing LST daily mean Python script (with IDRISI API but with GDAL in mind) to calculate: -Daily mean -Number of valid observation per day. 3) GAM prediction Some GAM predictions with interaction terms Including daily mean LST and LC in the GAM regression
1)Visualization of RMSE fit for Geographically Weighted Regression Writing a code in R to visualize the RMSE using -Stations location -Kriged error surface from stations 1)VISUALIZATION OF RMSE Moving beyond aggregate statistic…
RMSE FIT USING GWR WITH 30% RETAINED FOR VALIDATION
Run 10-Fit residuals from gwr using
Run 9-Fit residuals from gwr using
Run 8-Fit residuals from gwr using
Run 7-Fit residuals from gwr using
Fit residuals from gwr using Run 6-Fit residuals from gwr using Potentially useful to have the 2 sd thresholds…
Run 5-Fit residuals from gwr using
Run 4-Fit residuals from gwr using
Run 3-Fit residuals from gwr using NO KRIGED FIT
Run 8-Fit residuals from gwr using Run 2-Fit residuals from gwr using
Run 8-Fit residuals from gwr using Run 9-Fit residuals from gwr using Run 1-Fit residuals from gwr using
Python script (with IDRISI API but with GDAL in mind) to calculate: -Daily mean -Number of valid observation per day. LST DAILY MEAM PRODUCTION
MOD11A1 hdf OR83M.rst Mosaic Reprojection QC flags Level 1 and 2 Masking Low quality Daily MeanDaily Valid Obs. WORKFLOW DAILY MEAN CALCULATION Part of the process is automated in python with IDRISI API. Downloading Missing Data Assessment
OREGON- DAILY MEAN FOR DOY 001 mean_day001_rescaled.rst
OREGON-NUMBER OF VALID OBSERVATION FOR DOY 001 mean_day_valid_obs_001_Sum.rst
OREGON- DAILY MEAN FOR DOY 182 mean_day182_rescaled.rst
OREGON-NUMBER OF VALID OBSERVATION FOR DOY 182 mean_day_valid_obs_182_Sum.rst
SUMMARY INFORMATION OF THE DAILY MEAN CALCULATION A full assessment of the temporal and spatial distribution of mean would be necessary: -Most dates have 10 images (on average 9.88 images). -The number of valid values seems to be lower in Winter (more check needed). -Average per month may be quite helpful. Missing data: The average was done over the time period and there were 45 missing images (out of a total of 3652). Missing DOY 78 to 88: to Missing DOY 166 to 181: to (with July 01 missing 2) Missing DOY 301 to 305 Missing DOY 351 to 357: to (355 to 357 missing 2)
3)GAM MODELING USING LST AND LC GAM regressions: Some GAM predictions with interaction terms Including daily mean LST and LC in the GAM regression
Aggregated Classification class Class No. GLC UMDMODISGlobCover 2 Forest1 1,2,3,4,5,6,7, 8 1,2,3,4,5, 6 1,2,3,4,5, 8 40,50,60,70,90,100,160,17 0 Shrub29,10,11,12,147,8,96,7,9110,120,130,150 Grass Crop ,14 Mosaic 3 517, ,30 Urban Barren Snow Wetland Water body Table 5. Legend for the 10 aggregated land cover classes and the corresponding classes from the six individual global land cover legends. Modified from (Nakaegawa 2011). 1 I added class 3 to ‘forest’ since it was missing in original table. The class 2 entry under ‘shrub’ is probably an error and so is removed. 2 GlobCover class assignment needs to be finalized. 3 Mosaic is composed of cropland and natural vegetation. LAND COVER CONSENSUS CATEGORIES
GAM MODELS USED FOR THIS ANALYSIS mod1<- tmax~ s(lat) + s (lon) + s (ELEV_SRTM) mod2<- tmax~ s(lat,lon,ELEV_SRTM) mod3<- tmax~ s(lat) + s (lon) + s (ELEV_SRTM) + s (Northness)+ s (Eastness) + s(DISTOC) mod4<- tmax~ s(lat) + s (lon) + s(ELEV_SRTM) + s(Northness) + s (Eastness) + s(DISTOC) + s(LST) mod5<- tmax~ s(lat,lon) +s(ELEV_SRTM) + s(Northness,Eastness) + s(DISTOC) + s(LST) mod6<- tmax~ s(lat,lon) +s(ELEV_SRTM) + s(Northness,Eastness) + s(DISTOC) + s(LST,LC1) mod7<- tmax~ s(lat,lon) +s(ELEV_SRTM) + s(Northness,Eastness) + s(DISTOC) + s(LST,LC3)
RMSE FOR DIFFERENT DATES AND MODELS
RMSE FOR ALL DATES AND MODELS
PROBLEM WITH MISSING DATA If screening is used such as LST> 258 & LST<313) … the number of observations can drop to 48 and 20 for training and testing compared to 120 and 50 stations.
What's next..? 1)Continue the Visualization of RMSE for GAM and GWR 2) Influence of sampling on results GWR Prediction 3)Producing LST monthly 4) GAM using LST and Land Cover 5) Use Kriging and co-kriging to predict tmax 6) Documentation of the analysis