Using a gridded global data set to characterize regional hydroclimate in central Chile

1Using a gridded global data set to characterize regional hydroclimate in central Chile 3E.M.C Demaria1, E Maurer2*, J Sheffield3, E Bustos1, D Poblete1, S Vicuña1, F Meza1 51Centro de Cambio Global, Pontificia Universidad Católica de Chile, Santiago, Chile 62Civil Engineering Department, Santa Clara University, Santa Clara, CA, USA 73Department of Civil and Environmental Engineering, Princeton University, Princeton, NJ, USA 9*Corresponding author, emaurer@engr.scu.edu, 408-554-2178 10 11Proposed submission to J Hydrometeorology 12 11 13Abstract 14Central Chile is facing dramatic projections of climate change, with a consensus for declining 15precipitation, negatively affecting hydropower generation and irrigated agriculture Rising from 16sea level to 6,000 meters within a distance of 200 kilometers, precipitation characterization is 17difficult due to a lack of long-term observations, especially at higher elevations For 18understanding current mean and extreme conditions and recent hydroclimatological change, as 19well as to provide a baseline for downscaling climate model projections, a temporally and 20spatially complete data set of daily meteorology is essential We use a gridded global daily 21meteorological data set at 0.25 degree resolution for 1948-2008, and adjust it using monthly 22precipitation observations interpolated to the same grid using a cokriging method with elevation 23as covariate For validation, we compare daily statistics of the adjusted gridded precipitation to 24station observations For further validation we drive a hydrology model with the gridded 0.2525degree meteorology and compare stream flow statistics with observed flow We validate the high 26elevation precipitation by comparing the simulated snow extent to MODIS images Results show 27that the daily meteorology with the adjusted precipitation can accurately capture the statistical 28properties of extreme events as well as the sequence of wet and dry events, with hydrological 29model results displaying reasonable agreement for observed flow statistics and snow extent This 30demonstrates the successful use of a global gridded data product in a relatively data-sparse 31region to capture hydroclimatological characteristics and extremes 32 32 34Introduction 35 36Whether exploring teleconnections for enhancing flood and drought predictability or assessing 37the potential impacts of climate change on water resources, understanding the response of the 38land surface hydrology to perturbations in climate is essential This has inspired the development 39and assessment of many large scale hydrology models for simulating land-atmosphere 40interactions over regional and global scales [e.g.,Lawford et al., 2004; Milly and Shmakin, 2002; 41Nijssen et al., 2001a; Sheffield and Wood, 2007] 42 43A prerequisite to regional hydroclimatological analyses is a comprehensive, multi-decadal, 44spatially and temporally complete data set of observed meteorology, whether for historic 45simulations or as a baseline for downscaling future climate projections In response to this need, 46data sets of daily gridded meteorological observations have been generated, both over 47continental regions [e.g.,Cosgrove et al., 2003; Maurer et al., 2002] and globally [Adam and 48Lettenmaier, 2003; Sheffield et al., 2006] These have benefited from work at coarser time scales 49[Chen et al., 2002; Daly et al., 1994; Mitchell and Jones, 2005; New et al., 2000; Willmott and 50Matsuura, 2001], with many products combining multiple sources, such as station observations, 51remotely sensed images, and model reanalyses 52 53While these large-scale gridded products provide opportunities for hydrological simulations for 54land areas around the globe, they are inevitably limited in their accuracy where the underlying 55station observation density is low, the station locations are inadequate to represent complex 56topography, or where the gridded spatial resolution is too large for the region being studied 53 57Central Chile is an especially challenging environment for characterizing climate and hydrology 58since the terrain exhibits dramatic elevation changes over short distances, and the orographic 59effects this drives produce high spatial heterogeneity in precipitation in particular In general, the 60observation station density in South America is inadequate for long-term hydroclimate 61characterization [de Goncalves et al., 2006] While some of South America is relatively well 62represented by global observational datasets [Silva et al., 2007], regions west of the Andes are 63much less so [Liebmann and Allured, 2005] 64 65In this study, we utilize a new high-resolution global daily gridded dataset of temperature and 66precipitation, adjust it with available local climatological information, and assess its utility for 67representing river basin hydrology Recognizing the value in simulating realistic extreme events, 68we assess the new data product for its ability to produce reasonable daily streamflow statistics 69We evaluate the potential to reproduce climate and hydrology in a plausible manner, such that 70historical statistics are reproduced 71 72The principal aim of this study is to produce a gridded representation of the climate and 73hydrology of central Chile, are demonstrate a methodology for producing a reasonable set of data 74products that can be used for future studies of regional hydrology or climate Given these 75regional results, we assess the potential to export the method to other relatively data-sparse 76regions, where representative climatological average information is available but long-term daily 77data are inadequate The paper is organized as follows: Section describes the study area In 78Section we describe the data, the hydrological model and the methodological approach Results 74 79of the adjusted data set validation and model simulations are discussed in Section Finally, the 80main conclusions of the study are presented in Section 81 82Region 83 84The focus area of this study is central Chile (Figure 1), encompassing the four major river basins 85(from north to south, the Rapel, Mataquito, Maule, and Itata Rivers) between latitudes 35.25º S 86and 37.5º S The climate is Mediterranean, with 80% of the precipitation falling in the rainy 87season from May-August [Falvey and Garreaud, 2007] The terrain is dramatic, rising 88approximately 6000 meters within a horizontal distance of approximately 200 km, producing 89sharp gradients in climate [Falvey and Garreaud, 2009] 90 91Driven by the terrain, the area exhibits a dramatic climate gradient, with mean precipitation of 92approximately 500 mm per year at the North end of the study domain, and as much as 3000 mm 93per year in the high elevations at the Southern end of the domain It is evident from Figure that 94the high elevation areas are under-represented by any of the observation stations 95 96The region of Central Chile is especially important from a hydroclimatological standpoint, as it 97contains the largest proportion of irrigated agriculture and reservoir storage of any region in the 98country and provides water supply for some of Chile's largest cities A changing climate is 99evident in recent hydroclimate records [Rubio-Álvarez and McPhee, 2010], and future climate 100projections for the region indicate the potential for very large impacts [Bradley et al., 2006] The 101vulnerability of Central Chile to projected climate change is high, with robust drying trends in 95 10 102General Circulation Model (GCM) projections, and a high sensitivity to changing snow melt 103patterns [Vicuna et al., 2010], who also discuss the challenges in characterizing climate in a 104Chilean catchment with few precipitation observations, and none at high elevations 105 106Methods and data 107 108Gridded data set development 109 110We begin with a gridded global (land surface) dataset of daily precipitation and minimum and 111maximum temperatures at 0.25º spatial resolution (approximately 25 km), prepared following 112Sheffield et al [2006] To summarize, the forcing dataset is based on the NCEP–NCAR 113reanalysis [Kalnay et al., 1996] for 1948-2008, from which daily maximum and minimum 114temperature and daily precipitation are obtained at approximately 2º spatial resolution 115Reanalysis temperatures are based on observations, though precipitation is a model output and 116thus exhibits significant biases 117 118The reanalysis temperatures are interpolated to a 0.25º spatial resolution, lapsing temperatures by 119-6.5ºC/km based on the elevation difference between the large reanalysis spatial scale and the 120elevation in each 0.25º grid cell Precipitation is interpolated to 0.25º using a product of the 121Tropical Rainfall Measuring Mission (TRMM) [Huffman et al., 2007] following the methods 122outlined by Sheffield et al [2006] To ensure large-scale correspondence between this data set 123and the observationally-based monthly 0.5º data from the Climate Research Unit [CRU, Mitchell 124and Jones, 2005], precipitation is scaled so the monthly totals match the CRU monthly values at 116 12 125the CRU spatial scale Maximum and minimum temperatures are also scaled to match the CRU 126time series, using CRU monthly mean temperature and diurnal temperature range 127 128While the incorporation of multiple sources of extensively reviewed data provides an invaluable 129data product for global and continental scale analyses, as discussed by Mitchell and Jones [2005] 130ultimately much of the local characterization is traceable to a common network of land surface 131observations [Peterson et al., 1998], which is highly variable in station density for different 132regions For example, for the region of study shown in Figure 1, an average of 3-4 observation 133stations are included in the CRU precipitation data product, and none are in high-elevation areas 134This results in a few low elevation meteorological stations in Chile on the western side of the 135Andes, and the next observation station to the east is in a more arid area in Argentina Thus, the 136resulting precipitation fields in the gridded product for this region showed a spatial gradient 137opposite to that published by the Dirección General de Aguas [DGA, 1987] Figure 2a shows the 138spatial distribution of gridded global total annual precipitation that displays a notable decrease of 139rainfall with a elevation Conversely the DGA precipitation map is able to capture the 140climatological orographic enhancement of precipitation by the Andes (Figure 2b) The 141precipitation lapse rates for the latitudinal bands -35.125º S and -36.125º S show a negative 142gradient of precipitation with elevation in the global gridded data set whereas the DGA 143precipitation shows a positive gradient for the period 1951-89 (Figures 2c and 2d, respectively) 144 145Local data from the DGA of Chile, some monthly and some daily, were obtained to characterize 146better the local climatology While still biased toward low elevation areas, the stations (Figure 1) 147do cover a wider range and include altitudes up to 2400 m These stations were filtered to include 137 14 148those that had at least 90% complete monthly records for the 25-year period 1983-2007 The 149monthly average precipitation for the 25-year period for these 40 DGA stations was interpolated 150onto the same 0.25º grid using cokriging, with elevation being the covariate This method of 151cokriging has been shown to improve kriging interpolation to include orographic effects induced 152by complex terrain [Diodato and Ceccarelli, 2005; Hevesi et al., 1992] 153 154This process produced 12 monthly mean precipitation maps for the region The same 1983-2007 155period was extracted from the daily gridded data set, and monthly average values were calculated 156for each grid cell Ratios (12, one for each month) of observed climatology divided by the 157gridded data set average were then calculated for each grid cell Daily values in the gridded data 158set were adjusted to create a new set of daily precipitation data, P adj, which matches the 159interpolated observations produced with cokriging, using a simple ratio: Padj  i, j , t   Pgrid  i, j , t   160where Pgrid Pobs ,mon  i, j  (1) Pgrid ,mon  i, j  is the original daily gridded 0.25º data at location (i,j), Pobs is the interpolated 161observed climatology, overbars indicate the 25-year mean, and the subscript “mon” indicates the 162month from the climatology in which day t falls 163 164This same method was applied to a global dataset of daily meteorology in a data sparse region in 165Central America, resulting in improved characterization of precipitation and land surface 166hydrology [Maurer et al., 2009] In addition, this new adjusted data set includes the full 19481672008 period, despite the fact that local observations are very sparse before 1980 168 158 16 169To validate the adjusted precipitation data set, we computed a set of statistical parameters widely 170used to describe climate extremes [dos Santos et al., 2011; X Zhang and Yang, 2004] 171Additionally to evaluate the temporal characteristics of rainfall events we computed the wet, dry 172and transition probabilities Table shows a description of the statistics used 173 174To evaluate if the adjusted precipitation data set was capturing the orographic gradient of 175precipitation we compared VIC simulated Snow Water Equivalent (SWE) to the MODIS/Terra 176Snow Cover data set, which is available at 0.05 degree resolution for 8-day periods starting from 177the year 2000 MODIS snow cover data are based on a snow mapping algorithm that employs a 178Normalized Difference Snow Index [Hall et al., 2006] To estimate snow cover from the 179meteorological data, a hydrological model was employed 180 181Hydrologic Model Simulations 182 183To assess the ability of the daily gridded meteorology developed in this study to capture daily 184climate features across the watersheds, we simulate the hydrology of river basins in the region to 185obtain streamflow and snow cover estimates The hydrologic model used is the Variable 186Infiltration Capacity (VIC) model [Cherkauer et al., 2003; Liang et al., 1994] The VIC model is 187a distributed, physically-based hydrologic model that balances both surface energy and water 188budgets over a grid mesh The VIC model uses a “mosaic” scheme that allows a statistical 189representation of the sub-grid spatial variability in topography, infiltration and vegetation/land 190cover, an important attribute when simulating hydrology in heterogeneous terrain The resulting 191runoff at each grid cell is routed through a defined river system using the algorithm developed by 179 18 192Lohmann et al [1996] The VIC model has been successfully applied in many settings, from 193global to river basin scale [e.g.,Maurer et al., 2002; Nijssen et al., 2001b; Sheffield and Wood, 1942007] 195 196For this study, the model was run at a daily time step at a 0.25º resolution (approximately 630 197km2 per grid cell for the study region) Elevation data for the basin routing are based on the 15198arc-second Hydrosheds dataset [Lehner et al., 2006], derived from the Shuttle Radar Topography 199Mission (SRTM) at arc-second resolution Land cover and soil hydraulic properties were based 200on values from Sheffield and Wood [2007], though specified soil depths and VIC soil parameters 201were modified during calibration The river systems contributing to selected points were defined 202at a 0.25º resolution, following the technique outlined by O’Donnell et al [1999] 203 204Results and Discussion 205 206The adjusted data set was validated in several ways First, daily statistics were compared 207between the adjusted global daily data set and local observations, where available Second, 208hydrologic simulation outputs were compared to observations to investigate the plausibility of 209using the new data set as an observational baseline for studying climate impacts on hydrology 210 211Gridded meteorological data development and assessment 212 213The quality of daily gridded precipitation fields was improved using available monthly observed 214precipitation Rain gauge records from DGA were selected using two criteria: stations with 1910 20 552 553Table - Contingency table summarizing the comparisons of MODIS and VIC simulated snow cover Values 554are relative frequencies calculated as the total number of occurrences in each category divided by the number 555of pixels (1530) No Snow Snow Total 556 5930 60 No Snow 0.65 0.06 0.71 Snow 0.05 0.24 0.29 Total 0.69 0.31 1.0 557List of Figures 558 559Figure - Geographic location of the study area in Central Chile From north to south the basins 560are: Rapel, Mataquito (Mataquito river at Licanten), Maule (Claro river at Rauquen and 561Loncomilla river at Bodega) and Itata river basins Circles indicate the location of DGA rain 562gauges and stars the location of the three stream gauges used in VIC simulations 563Figure - Maps of annual precipitation for the period 1951-1980 Source a) gridded global 564observations and b) DGA Precipitation lapse rates for latitudinal bands -35.125 S and -36.125 S 565for c) global gridded precipitation data set and d) DGA data set 566Figure - Scatterplots of observed and predicted monthly precipitation for the month of July 567Figure - a) Annual adjusted global precipitation for the period 1950-2006 and b) differences 568between the original global gridded and the adjusted global precipitation data sets 569Figure - Location of DGA rain gauge stations and adjusted global precipitation grid points used 570for validation of daily rainfall 571Figure - Boxplots of statistical parameters, green represents observations and purple represents 572adjusted precipitation for each geographic location The bottom and top lines represent the 25th 573and 75th percentiles and the middle line represents the median Whiskers extend from each end 574of the box to the adjacent values in the data within 1.5 times the Inter Quartile Range The Inter 575Quartile Range is the difference between the third and the first quartile, i.e., 25th and 75th 576percentiles Outliers are displayed with a plus sign 577Figure - Probabilities of a) wet and b) dry days, and transition probabilities c) and d) Daily 578observed (black) and adjusted gridded (grey) precipitation for the four selected locations 6131 62 579Figure - Observed and Simulated monthly flows for the Mataquito river at Licanten for the 580calibration period (top panel) and validation period (bottom panel) Summary statistics are 581shown in each panel 582Figure - Monthly observed and simulated flows for the Claro river at Rauquen 583Figure 10 - Same as Figure but for Loncomilla river at Bodega 584Figure 11 - Statistical properties of observed and VIC simulated streamflows in three basins: 585Mataquito river, Claro river and Loncomilla river (a) Center timing, (b) water year volume, (c) 5863-day peak flows and (d) 7-day low flows 587Figure 12 - Comparison of snow coverage for the period August 21-28, 2002 Shaded areas 588indicate snow coverage a) MODIS and b) VIC simulated Snow Water Equivalent 589 6332 64 590 591Figure - Geographic location of the study area in Central Chile From north to south the basins are: Rapel, 592Mataquito (Mataquito river at Licanten), Maule (Claro river at Rauquen and Loncomilla river at Bodega) 593and Itata river basins Circles indicate the location of DGA rain gauges and stars the location of the three 594stream gauges used in VIC simulations 6533 66 595 596Figure - Maps of annual precipitation for the period 1951-1980 Source a) gridded global observations and 597b) DGA Precipitation lapse rates for latitudinal bands -35.125 S and -36.125 S for c) global gridded 598precipitation data set and d) DGA data set 599 600 6734 68 601 602Figure - Scatterplots of observed and predicted monthly precipitation for the month of July 603 6935 70 604 605Figure - a) Annual adjusted global precipitation for the period 1950-2006 and b) differences between the 606original global gridded and the adjusted global precipitation data sets 607 7136 72 608 609Figure - Location of DGA rain gauge stations and adjusted global precipitation grid points used for 610validation of daily rainfall 611 7337 74 612 613Figure - Boxplots of statistical parameters, green represents observations and purple represents adjusted 614precipitation for each geographic location The bottom and top lines represent the 25th and 75th percentiles 615and the middle line represents the median Whiskers extend from each end of the box to the adjacent values 616in the data within 1.5 times the Inter Quartile Range The Inter Quartile Range is the difference between the 617third and the first quartile, i.e., 25th and 75th percentiles Outliers are displayed with a plus sign 7538 76 618 619Figure - Probabilities of a) wet and b) dry days, and transition probabilities c) and d) Daily observed 620(black) and adjusted gridded (grey) precipitation for the four selected locations 621 7739 78 622 623Figure - Observed and Simulated monthly flows for the Mataquito river at Licanten for the calibration 624period (top panel) and validation period (bottom panel) Summary statistics are shown in each panel 625 7940 80 626 627Figure - Monthly observed and simulated flows for the Claro river at Rauquen 8141 82 628 629Figure 10 - Same as Figure but for Loncomilla river at Bodega 630 8342 84 631 632Figure 11 - Statistical properties of observed and VIC simulated stream flows in three basins: Mataquito 633river, Claro river and Loncomilla river (a) Center timing, (b) water year volume, (c) 3-day peak flows and (d) 6347-day low flows 635 8543 86 636 637Figure 12 - Comparison of snow coverage for the period August 21-28, 2002 Shaded areas indicate snow 638coverage a) MODIS and b) VIC simulated Snow Water Equivalent 639 8744 88 ... climatology in which day t falls 163 164This same method was applied to a global dataset of daily meteorology in a data sparse region in 16 5Central America, resulting in improved characterization... basins 560are: Rapel, Mataquito (Mataquito river at Licanten), Maule (Claro river at Rauquen and 561Loncomilla river at Bodega) and Itata river basins Circles indicate the location of DGA rain... 592Mataquito (Mataquito river at Licanten), Maule (Claro river at Rauquen and Loncomilla river at Bodega) 593and Itata river basins Circles indicate the location of DGA rain gauges and stars the location