Presentation summarizing comparison of station data with MODIS LST and TRMM data as well as methods for interpolation of tmax, tmin, and ppt in the Cape Floristic Region of South Africa.

This is a short powerpoint presentation of the first interpolation results for one date in Oregon. Results were obtained by running GAM and linear models in r for the day July 15 for the year 2010.RMSE values were corrected following the meeting.

This presentation reports on the results for Task 361: GAM interpolation for 10 dates for the Oregon Case study. There are 186 stations in Oregon and the GAM models are assessed using RMSE by holding out 30% of the stations for each date.

This presentation provides an update and summary of three ongoing tasks:
1) Prediction of tmax using GWR with 0,30,50 and 70% of hold-out stations.
2) Screening of out range values and assessment of GAM predictions.
3) Production of daily mean LST from MOD11A1 product....

A brief overview of a comparison between TRMM satellite derived precipitation product and station observations in Oregon.

This presentation reports on the prediction of tmax for the Oregon case study. There are three main analyses documented:
1) Visualization of residuals and RMSE from GWR models.
2) Processing and calculation of daily mean LST over the 2001-2010 time period.
...

Overview of processing the MOD06 cloud product

Four summary figures from current draft of short communication about land-cover bias in MODIS cloud products.

Incorporating satellite derived cloud climatologies to improve high resolution interpolation of daily precipitation
Presented at the International Biogeography Society Meeting in Miami, FL ****************************************************************
...

Remotely Sensed Metrics of Habitat Heterogeneity for Spatial Biodiversity Modeling

Presentation to the Yale Climate and Energy Institute on March 5, 2013.

NASA/NCEAS/iPlant Update
Adam Wilson
January 28, 2013

Update for the biweekly meeting detailing the changes made to the MOD35 cloud mask product from Collection 5 to Collection 6.

An oral presentation at the US-IALE conference in Austin, Texas

Here are a few slides on where I am at and some questions to guide my next steps. I appreciate any leads or good advice you all might have.

Some example metrics derived from the consensus land cover data and a time series of MODIS EVI for quantifying habitat heterogeneity within 1km pixels (updated)

Presenting preliminary results to identify pits and spikes in EarthEnv-DEM90 dataset.

Some slides with preliminary analysis of climate station gaps to prioritize search for new datasets by William Freedberg (at Yale). See [[https://docs.google.com/presentation/d/1I_aecSaYm2AWCLp3Pd_MKBmrCrHiIzjV4g_hHFBV8-U/edit?usp=sharing]] for the live version....

Topographic variables obtained from the EarthEnv-DEM90.

Median values in 10x10 pixels of the following variables
altitude
slope
roughness
Terrain Ruggedness Index (TRI)
Topographic Position Index (TPI)

obtained by gdaldem

Preliminary results of modeling bird species richness using EVI-based and landcover-based heterogeneity metrics

An update describing progress on development of the MOD09 Cloud Climatology.

This is a brief presentation of the state of the project in terms of coding as well as a summary of the review paper. The presentation goes in some more depth on the accuracy assessment of climate product.

This presentation presents mean monthly LST averages derived from daily MOD11A1. Results from Kriging and co-kriging are presented as well as GAM results using LST monthly averages as variables.

This document provides updates presented at the IPLANT meeting 05012012:
- Results from GAM models run for the year 2010 (365 dates).
- Residuals maps used to examine patterns of GAM residuals
- First results from the study of the effect of sampling on GAM models for 10 dates
...

This document summarizes ongoing work for the IPLANT meeting on 05152012:
-GAM assessment over year 2010: average per month
-GAM+Kriging results on 10 dates
-Sampling influence on GAM RMSEs for 10 dates

This presentation shows early results for the fusion method.

This update shows predictions at every location in Oregon for the first time. Results for kriging include automated fitting of variograms.

First summary of the method comparisons using five methods: Kriging, GWR, GAM, CAI and Fusion. All predictions are done for the Oregon case study. Methods of assessment include: 1)accuracy metrics (MAE&RMSE), 2)multisampling, 3)accuracy in term of distance to closest training station, 4)map visualization, 5)accuracy in term of station density and 6)accuracy profiles at specific stations. This presentation is will be updated and edited at later stages....

This is a condensed version of analyses performed following the IPLANT meeting of 10/22/2012.
It includes interpolation of Tmax using all station at the monthly time scale, simplified models for CAI and screening of LST and ELEV_SRTM.

This document presents maximum temperature predictions using the GAM CAI method for Venezuela for year 2010.

This power point displays results for four tmax prediction runs:

1) Tmax,VE,monthly average 2000-2010, year 2010 (additional info in issue #568)

2) Tmax,VE,monthly average 1980-2010, year 2010

3) Tmax,VE,monthly average 2000, year 2000

4) Tmax,VE,monthly average 1980-2000, year 2000
...

Update reporting on the new code design for interpolation and displaying accuracy results for 4 TMAX predictions and 2 TMIN predictions in Venezuela. First results of TMAX predictions using new code is also recorded.

This presentation shows the latest tmax interpolation results using the new code. It includes plots and accuracy statistics for Oregon using GAM_daily, Kriging_daily and GWR_daily methods.
First test tmax predictions for the Queensland region are also reported using the GAM_fusion method....

Here is an initial draft concerning the description of the current code and workflow for temperature predictions. This will be expanded further later on.

Evaluation of the GMTED2010 products

Median Statistic
Minimum Statistic
Mean Statistic
Maximum Statistic
Standard Dev. Statistic
Breakline Emphasis
Systematic Subsample

to derive topographic variables at 1km resolution....

Topographic variables obtained by the GMTED2010 using Median Statistic product.
Median values aggregation at 1km (10x10 pixels)
Scripts:
root/terrain/procedures/dem_variables/gmted2010_res_x10/sc2a_dem_variables.sh
root/terrain/procedures/dem_variables/gmted2010_res_x10/sc3a_dem_variables_merge.sh...

GeoTiff that may be used as reference tif for the production of environmental variables.
The following metadata are reported in the file

gdalinfo geo30arcsec_reference.tif
Driver: GTiff/GeoTIFF
Files: ../data_process/geo_area/reference_tif/geo30arcsec_reference.tif
...

One page summary of fusion approach (equations).

Miscellaneous reports and figures associated with differences between the SRTM and ASTER DEMs.

Data Portals currently used by Rick Reeves to create Fused Terrain Data Layer:

Data Search platforms:

http://srtm.csi.cgiar.org/ - CGIAR Consortium data archive - CGIAR SRTM data

https://wist.echo.nasa.gov/~wist/api/imswelcome/ - NASA WIST portal - ASTER GDEM Data
...