Contents

## Statistical downscaling using Localized Constructed Analogs (LOCA)David Pierce a

Slagowski, Joyce, News Reporter has reference to this Academic Journal, PHwiki organized this Journal Statistical downscaling using Localized Constructed Analogs (LOCA)David Pierce in addition to Dan CayanScripps Institution of OceanographyBridget Thrasher, Edwin Maurer, John Abatzoglou, Katherine HegewischDevelopment sponsored by The Cali as long as nia Energy Commission Department of Interior/US Geological Survey via the Southwest Climate Science Center NOAA RISA Program through the Cali as long as nia Nevada Applications ProgramProduction runs sponsored byU.S. Army Core of Engineers/USBRNASA via computing resourcesDownscaling systemQuantile Mapping (QM)Constructed Analogs (CA; Hidalgo et al. 2008)Bias Correction in addition to Constructed Analogs (BCCA; Maurer et al. 2010)Multivariate Adapted Constructed analogs (MACA; Abatzoglou & Brown 2012)Bias Correction with Spatial Disaggregation (BCSD; Wood et al. 2004)Global Regridding Bias SpatialModels Correction DownscalingIssues with bias correction

This Particular University is Related to this Particular Journal

TmaxDifference between original model-predicted change in addition to change after bias correction 2070-2100 minus 1976-2005Ensemble averaged across 21 GCMsdeg-C1. QM does not preserve model-predicted changes (Maurer in addition to Pierce, HESS, 2014)EDCDFm reference:Li, H., J. Sheffield, in addition to E. F. Wood, 2010: Bias correction of monthly precipitation in addition to temperature fields from Intergovernmental Panel on Climate Change AR4 models using equidistant quantile matching. J. Geophys. Res. Atmos., 115 (D10101), doi:10.1029/2009JD012882.What about precipitationEvaluate temperature changes as a difference (degrees C)Evaluate precipitation changes as a ratio (percent)Positive definiteWide dynamic rangeRain shadow regions

PrecipitationDifference between original model-predicted change in addition to change after bias correction in percentage points2070-2100 minus 1976-2005Ensemble averaged across 21 GCMsPresRat schemeLike EDCDFm (Li et al. 2010) except:Preserves the ratio of model-predicted changes (not the difference)Zero-precipitation threshold (preserve observed number of dry days in historical period)Final correction factor to preserve mean changePresRat schemeLike EDCDFm (Li et al. 2010) except:Preserves the ratio of model-predicted changes (not the difference)Zero-precipitation threshold (preserve observed number of dry days in historical period)Final correction factor to preserve mean change

Correction factors necessary to preserve model-predicted changes (2070-2099 vs. 1976-2005) in mean precipitationAveraged across 21 GCMsCorrection factorPrecipitationDifference between original model-predicted change in addition to change after bias correction in percentage points2070-2100 minus 1976-2005Ensemble averaged across 21 GCMsIf log-RMSE is f, then models are off by factor of (1 + f), on average2. Model Errors can be a Function of Frequency

Log-RMSE metricsHow much does frequency-dependent bias correction change values3. St in addition to ard QM not multivariateTemperature on precipitating days affects snow cover (Abatzoglou et al.)Bias correct temperature conditional on precipitation > 0 or not

Issues with spatial downscalingHidalgo, H.G., Dettinger, M.D., in addition to Cayan, D.R., 2008Slide from Mike Dettinger, USGS (tenaya.ucsd.edu)Spatial Downscaling with constructed analogsSlide from Mike Dettinger, USGS (tenaya.ucsd.edu)Spatial Downscaling with constructed analogsHidalgo, H.G., Dettinger, M.D., in addition to Cayan, D.R., 2008

Slide from Mike Dettinger, USGS (tenaya.ucsd.edu)Spatial Downscaling with constructed analogsHidalgo, H.G., Dettinger, M.D., in addition to Cayan, D.R., 2008Slide from Mike Dettinger, USGS (tenaya.ucsd.edu)Spatial Downscaling with constructed analogsHidalgo, H.G., Dettinger, M.D., in addition to Cayan, D.R., 2008Issues with current downscaling (BCCA)BCCA = Bias correction with Constructed AnalogsAveraging step reduces temporal variance (i.e., mute extremes)

2. Frequency of occurrence -> percent of amountTake an extreme example as long as illustration:Slide 2260% of the time40% of the timeContributes to reduction in extremes3. Drizzle problem from downscalingSlide 23New downscaling (LOCA) (Step 1 of 2)BCCA uses 30 best matching analog days over entire domainLOCA starts with 30 best matching analog days over the region around the pointRegion: everywhere correlation with point being downscaled is > 0 (in obs)Regions are calculated by season (DJF, MAM, JJA, SON) in addition to variable (pr, tasmax, tasmin, etc.)Gives a natural domain independence to LOCA (extending domain past region does not affect results at the point)Example shown as long as precipitation

New downscaling (LOCA) (Step 2 of 2)Once 30 regional analog days are selected:Find best one (of the 30) matching days in a small localized region (~1 degree) around each pointThis two step process means each point:Is consistent with whats happening regionallyIs the best match locallyPoints whose selected analog day is different from a neighbors (edge points) use a weighted average of the relevant analog days~30% of points are edge pointsGreatly reduced averaging means:Better extremesBetter spatial coherenceFar less drizzle problem 4. Run out of analogs as long as extreme days1950-99 .2070-99Existing methods:1950-99 2000-2009 . 2010-2039 2040-2069 .2070-99Anomaly w.r.t. 30-year climatologyUse LOCA to downscale changes in climatologyLOCA:Anomaly w.r.t. historical period (Tmin, Tmax)Project in association with Keith Dixon, GFDL5. Averaging increases spatial coherence precip (red = more coherent)

Evaluation:Seasonal mean of daily precipitation (mm/day) in CCSM4Error in %Evaluation:Seasonal mean of daily Tmax (degC) in CCSM4Error in degCEvaluation:St in addition to ard deviation of daily precip (mm/day), averaged by seasonCCSM4Error in %

Summary of Production Runs32 CMIP5 modelsHistorical: 1950-2005. RCP 4.5 in addition to RCP 8.5: 2006-2100 (2099 some models)Climatological period: 1950-99Interpolated model calendars to st in addition to ard calendar w/leap daysNorth America 24.5 N to 52.8 N at 1/16th degree resolutionDaily Tmin, Tmax, Precip (specific humidity 23 models). ACCESS1-0ACCESS1-3CCSM4CESM1-BGCCESM1-CAM5CMCC-CMCMCC-CMSCNRM-CM5CSIRO-Mk3-6-0CanESM2EC-EARTHFGOALS-g2GFDL-CM3GFDL-ESM2GGFDL-ESM2MGISS-E2-HGISS-E2-RHadGEM2-AOHadGEM2-CCHadGEM2-ESIPSL-CM5A-LRIPSL-CM5A-MRMIROC-ESMMIROC-ESM-CHEMMIROC5MPI-ESM-LRMPI-ESM-MRMRI-CGCM3NorESM1-Mbcc-csm1-1bcc-csm1-1-minmcm4SummaryMany bias correction & downscaling schemes Quantile mapping, BCCA:Muted extremesDifferent biases at different frequenciesToo much spatial coherenceDrizzle problemsWrong temperature of precipitationNew bias correction in addition to LOCA downscalingExtremes preserved pretty well, along with seasonal means in addition to std deviationsReasonable preservation of original model-predicted changesFrequency dependent bias correctionSpatial coherence not degraded as muchGreatly reduces drizzle problemBias correct temperature conditional on precipitationPierce, D. W., D. R. Cayan, in addition to B. L. Thrasher, 2014: Statistical downscaling using Localized Constructed Analogs (LOCA). Journal of Hydrometeorology, v. 15, page 2558-2585Analysis plots: loca.ucsd.edu39

## Slagowski, Joyce News Reporter

Slagowski, Joyce is from United States and they belong to Daily News-Sun and they are from Sun City, United States got related to this Particular Journal. and Slagowski, Joyce deal with the subjects like City/Metropolitan News; Education; Health and Wellness; Local News

## Journal Ratings by Institute of Textile Technology

This Particular Journal got reviewed and rated by Institute of Textile Technology and short form of this particular Institution is US and gave this Journal an Excellent Rating.