1. Introduction
Recent studies suggest that annual precipitation over the southwestern United States has begun to decrease over the last decade as part of a multidecadal cycle (Meko et al. 2007). When combined with other variations in the hydroclimatology that impact water resources and availability—including variations in evaporation, transpiration, and runoff (Milly et al. 2005; Oki and Kanae 2006)—this decline appears to be driving long-term decreases in streamflow (Woodhouse et al. 2006) and increases in drought severity (Cook et al. 1999; Stahle et al. 2000). Conversely, paleoclimate records indicate historically large increases in summertime rainfall rates over the last 70 yr (compared with the previous 1400) for localized regions in northern Arizona and southern Utah (Salzer and Kipfmueller 2005). Similar results have been observed for other warm-season monsoon regions (Hennessy et al. 1999; Goswami et al. 2006), suggesting that historical rainfall variations during the summertime season may differ from trends during other times of the year (Knowles et al. 2006). Here, we use two disparate analytical and modeling techniques to analyze the observed seasonal-mean precipitation and daily precipitation characteristics over the southwestern United States during the peak of the summertime North American monsoon system [NAMS; July–September; see Adams and Comrie (1997)], with an eye toward detecting systematic trends in these summertime precipitation characteristics. Although observationally based trends in seasonal-mean precipitation, daily precipitation characteristics, and extreme events have previously been documented (e.g., Karl and Knight 1998; Groisman et al. 2004, 2005; Alexander et al. 2006), most of the results focus on the traditional summertime period (June–August), thereby convolving periods of significant dryness (June) with periods of significant monsoonal precipitation [July–August; see Adams and Comrie (1997), Higgins et al. (1997), and Guirguis and Avissar (2008)]. In addition, most studies convolve regions that experience significantly different year-to-year variations in precipitation (e.g., Comrie and Glenn 1998). In this paper, we will account for both the temporal and regional structure of monsoon rainfall across this region.
2. Observational data
The daily precipitation data used here are taken from the serially complete daily maximum and minimum temperatures and precipitation compiled by Eischeid et al. (2000), which was created using an objective-analysis methodology (Eischeid et al. 1995) to identify and remove erroneous entries and fill in missing data; it was later augmented by removing unreliable stations from the record and employing improved consistency checks (Di Luzio et al. 2008). Here, the methodology is applied to the quality-controlled (Reek et al. 1992) National Climatic Data Center (NCDC) Summary of the Day dataset, which is also used by other long-term historical climate databases, such as the U.S. Historical Climatology Network (Easterling et al. 1996). The latest version of the Eischeid et al. (2000) dataset comprises daily precipitation observations from at least January 1948 to August 2003 for 14 317 sites in the United States, although most stations include observations prior to 1948. Summertime (July–September) daily precipitation data at 78 stations covering the southwestern United States are extracted to examine their statistical characteristics. These stations are chosen because their complete data series span at least 70 yr (1931–2000), thereby avoiding issues associated with changes in station locations, which can impact the temporal evolution of gridded observational datasets (Hamlet and Lettenmaier 2005). In addition, the use of this station dataset ensures that results are not biased by interpolation algorithms or scale selection, as they may be with gridded data.
3. Analysis and results
To obtain information about coherent, regional-scale variations in summertime precipitation and its characteristics, we use two regionalization (or cluster) methodologies: Ward’s method (Ward 1963) and the K-means method (Gong and Richman 1995). In addition, we use two criteria to determine the optimal cluster number for each method: the pseudo-T 2 criterion (Duda and Hart 1973) and the Cubic Clustering Criterion (CCC; Sarle 1983). Here, the two climate regionalization methods are applied to the year-to-year variations in summertime total precipitation amounts across the 78 stations (Anderson et al. 2009). Identical optimal cluster numbers (4) and station partitionings are returned by the separate algorithms and criteria (Fig. 1).
Once the stations are assigned to a particular cluster, the station-averaged total precipitation amount for a given cluster and a given year can be determined. Although this value does not represent the area-averaged rainfall amount, we will use it as an internally consistent metric for estimating year-to-year variations, as well as long-term trends, in seasonal total precipitation across the set of stations within a given cluster. To test whether trends in this total precipitation (and characteristics of precipitation, discussed later in the study) are significantly different from those expected by chance, we simulate the daily rainfall over each cluster using stochastic chain-dependent models, which treat the frequency, coverage, and intensity of daily-rainfall events separately (Katz 1977; Swift and Schreuder 1981; Wilks 1999). The construction of the chain-dependent models is described in more detail in Anderson et al. (2009). Briefly, for each region we create a chain-dependent model for the region comprising a previous dependent-occurrence chain, an empirical rainfall coverage distribution, and an empirical rainfall amount distribution. Using the stochastic daily rainfall model system, 142 separate 70-yr sequences of 92-day summertime precipitation time series are generated for each cluster. A bootstrapping methodology is then applied to these time series to arrive at 14 200 separate 70-yr sequences for each region.
Because the daily rainfall statistics within these models are kept constant, year-to-year variations in station-averaged total precipitation amount, rainfall frequency, rainfall intensity, and rainfall coverage within the simulated datasets are purely the result of the random evolution of the daily time series model. Conversely, observed variations that lie outside of the chain-dependent model’s envelope of variability cannot have been produced simply by the stochastic (i.e., random) evolution of daily precipitation events constrained by fixed daily precipitation characteristics. Instead, these observed variations result from a systematic change in the underlying climate of the system (Singh and Kripalani 1986; Shea and Madden 1990; Gregory et al. 1993; Katz and Parlange 1998; Wang et al. 2006).
In previous research, these chain-dependent models have been used to identify nonstochastic year-to-year variations in seasonal-mean precipitation at individual stations (Wang et al. 2006) and across regions (Anderson et al. 2009), as well as to quantify the influence of variations in daily rainfall characteristics upon observed anomalies in seasonal-mean precipitation during a given year. Here, we use these same model systems to determine whether observed trends in total precipitation and precipitation characteristics are the result of long time-scale shifts in the underlying hydroclimatology for a given region, or whether these trends could have arisen simply from the stochastic evolution of daily rainfall events absent of any change in the underlying daily precipitation characteristics of the system (as they do in the stochastic models themselves).
First, we examine seasonal total precipitation and precipitation characteristics (e.g., number of rainfall days and mean rainfall coverage). For each region, a linear trend is applied to the given (observed) variable and the size of the trend is computed based upon the slope of the linear fit. Then, for each of the 14 200 seventy-year simulations, a similar linear trend is computed. The probability distribution function (PDF) of these simulated trends is then compared with the observed value. Table 1 provides a summary of the results (it should be noted that all of the results are valid at the same significance level if only the original 142 seventy-year simulation periods are used).
Overall, the northern half of the domain contains significant (at the 95% level) trends in total precipitation and precipitation characteristics, whereas the southern half indicates no significant trends in any of the variables (these latter results hold even if the significance level is reduced to 90%). Figure 2 shows the observed time series of these three variables as well as the location of the overall trend within the PDF generated by the simulated time series for the two northern subregions (region 1 and 2). In both regions, there is a significant increase in the number of rainfall events and the coverage during these rainfall events, as well as the total seasonal rainfall. To confirm that these results are not sensitive to the choice of start dates—particularly the onset and demise of the “Dust Bowl” during the early- and mid-1930s, located just to the east of the region examined here (Schubert et al. 2004)—we calculate the observed trends separately for the full time period (70 yr) as well as shorter time periods [50 and 25 yr, as in Solomon et al. (2007)]. As can be seen, all of the trends remain consistent and significant across these three time periods (although such significance still allows for the small but finite possibility that they could have arisen spuriously).
Next, we examine trends in extreme events, namely, intense storm activity and extreme dry-spell durations. For the change in intense storm activity, we calculate the amount of total precipitation during a given year generated by daily rainfall events exceeding the 95% threshold [as determined by the cumulative distribution function of nonzero daily rainfall amounts across the 1931–2000 period; Groisman et al. (2004); Alexander et al. (2006)]. For changes in extreme dry spells, we select the period with the longest number of consecutive no-rainfall days as the extreme event for each season (Peterson et al. 2001; May 2004; Alexander et al. 2006). We then compute the linear trend for intense rainfall activity and extreme dry-spell durations based upon year-to-year variations in these two indices. A similar analysis is then applied to each of the 70-yr simulations of daily rainfall for the given region, from which the PDF is generated. Table 1 provides a summary of the results.
As with the seasonal-mean quantities, only the northern portion of the domain indicates significant trends in the amount/duration of intense/extreme events (again, the results hold even if the significance level is reduced to 90%). Figure 3 shows the observed time series of these two variables as well as the location of the overall trend within the PDF generated by the simulated time series for the two northern subregions (region 1 and 2). In region 2, the precipitation produced by intense storms is increasing while the length of extreme dry spells is decreasing. This increase in accumulated precipitation associated with intense events in region 2 coincides with a significant (at the 95% level) increase in the number of such events (on the order of 3.3 events year−1 century−1; not shown). In region 1, no significant change is found in intense storm activity; however, the length of extreme dry spells is decreasing in this region as well. As before, 50- and 25-yr trends are consistent with the full 70-yr trends, with the exception of the extreme dry-spell lengths over region 1 during the final 25-yr period, which appear to have remained relatively constant over this shorter period.
Unfortunately, to see whether similar trends have continued into the present century (2001–08) we cannot use the station-based data (which only covers the summertime period through 2002). However, we can utilize the NCDC Climate Division Dataset (NCDC 1994) available through the National Oceanic and Atmospheric Administration (NOAA)/Earth System Research Laboratory (ESRL). From this dataset, we archive seasonal (July–September) total precipitation for the climate divisions over the southwestern United States, covering the period 1931–2008. Then, for the two regions of interest (region 1 and region 2), we perform a multivariate fit of the relevant climate division data to the station-based seasonal-mean precipitation for the period of overlap (1931–2000). For region 1, we select the climate division data from eastern Utah (divisions 3–7) and western Colorado (division 2). For region 2, we select the climate division data from eastern Colorado (divisions 1, 3, and 4) and the panhandle of Nebraska (division 1).
Once we have determined the regression coefficients for each of the respective climate divisions, we apply them to the extended climate division time series (1931–2008) to produce an extended record of seasonal-mean precipitation amounts that matches the station-based amounts found in the given region (the correlation between the station-based time series and the climate division-based time series for the period of overlap are r = 0.96 and r = 0.90 for regions 1 and 2, respectively). We can then calculate trends of the seasonal-mean precipitation amounts over the extended time series to see whether results are consistent with the shorter 1931–2000 period (Fig. 4). In region 1, the trend in seasonal-mean precipitation over the longer time period (17.6 mm century−1) is smaller than that found in the station-based record, principally because of a substantial decline in precipitation over the region during 2008, but the trend remains significant (at the 90% level). The precipitation decline in 2008 also affects the short-term (25 yr) trend, which may be more strongly related to known decadal-scale variations driven by changes in tropical and extratropical sea surface temperatures in the Pacific (Castro et al. 2007). However, the long-term trends (78 and 50 yr) remain consistent with one another across the extended observational period. In region 2, the overall trend is again less than that found in the station-based record (35.9 mm century−1), but it remains significantly different from that expected by chance (at the 95% level). In addition, it is consistent with trends across the 50- and 25-yr time periods. Although we cannot perform a similar analysis upon the summertime precipitation characteristics (e.g., event frequency and coverage, intense storm amounts, and extreme dry-spell conditions), this analysis does suggest that the long-term trend of increasing summertime precipitation in the periphery of the monsoon region has persisted into the current century.
4. Discussion and conclusions
Overall, our research indicates that there has been a significant and systematic change in the summertime precipitation characteristics over a large portion of the southwestern United States during the last 70 yr, which confirm paleoclimate records indicating historically large increases in summertime rainfall rates during this time (Salzer and Kipfmueller 2005). Results presented here indicate that in the regions north of the core monsoon region, including the elevated regions of Utah and western Colorado along with the Rocky Mountain Front Range farther east, there have been positive multidecadal trends in seasonal precipitation amounts, number of daily rainfall events, and coverage of daily-rainfall events. In addition, intense storm activity appears to be increasing, whereas the length of extreme droughts is decreasing. Over the core (southwestern United States) monsoon region (centered in Arizona and western New Mexico), there does not appear to be any similar systematic trends.
It is important to note that this study has focused only on precipitation during the summertime monsoon (July–September); although summertime monsoon rainfall can contribute up to 40%–50% of the overall annual precipitation amount in the “core” monsoon region (Carleton et al. 1990), historical trends during this season may differ from trends that occur during other times of year (Knowles et al. 2006), which will need to be investigated separately. In addition, other variations in the hydroclimatology of the region—including variations in evaporation, transpiration, and run-off—can also impact water resources and availability (Milly et al. 2005; Oki and Kanae 2006), and hence should also be investigated further.
In addition, we note that previous studies have suggested that the declines in annual precipitation over the southwestern United States during the last decade [as part of a multidecadal cycle; Meko et al. (2007)] may be exacerbated by anthropogenic global climate change over the coming century (Seager et al. 2007); however, it is unclear whether such projected trends are consistent across the seasons (Solomon et al. 2007). Our results show that during the summertime monsoon a systematic shift in the precipitation characteristics over large portions of the southwestern United States is already detectable within the observed record, which could serve as an indication of the possible impacts of a changing climate on summertime precipitation patterns in a region where model projections over the coming century provide little guidance (Solomon et al. 2007). Although our results do not unequivocally attribute the northward expansion of the NAMS precipitation to increasing global-mean temperatures, they are in agreement with the northward expansion of the summertime North American monsoon into Colorado and Utah during periods of increasing global temperatures, associated with the climatic optimum of a.d. 900–1300 (Petersen 1994). They are also in agreement with the apparent intensification of the summertime North American monsoon during the relatively warm mid-Holocene period (Harrison et al. 2003; Mock and Brunelle-Daines 1999; Poore et al. 2005).
Acknowledgments
Dr. Anderson’s research was supported by a visiting scientist appointment to the Grantham Institute for Climate Change, administered by Imperial College of Science, Technology, and Medicine. This research was also funded by Cooperative Agreement NOAA-NA040AR431002. The views expressed here are those of the authors and do not necessarily reflect the views of NOAA. The authors wish to thank Jon Eischeid at NOAA–CIRES Climate Diagnostic Center for producing and providing the station-based precipitation data products. The NCDC Climate Division data are provided by the NOAA/ESRL/Physical Sciences Division, Boulder, Colorado (available online at http://www.cdc.noaa.gov/). We also wish to thank David Gochis and two other anonymous reviewers for all their insightful and constructive comments.
REFERENCES
Adams, D. K., and A. C. Comrie, 1997: The North American monsoon. Bull. Amer. Meteor. Soc., 78 , 2197–2213.
Alexander, L. V., and Coauthors, 2006: Global observed changes in daily climate extremes of temperature and precipitation. J. Geophys. Res., 111 , D05109. doi:10.1029/2005JD006290.
Anderson, B. T., J. Wang, G. Salvucci, and S. Gopal, 2009: Influence of daily rainfall characteristics upon regional summertime precipitation over the Southwestern United States. J. Hydrometeor., 10 , 1218–1230.
Carleton, A. M., D. A. Carpenter, and P. J. Weser, 1990: Mechanisms of interannual variability of the Southwest United States summer rainfall maximum. J. Climate, 3 , 999–1015.
Castro, C. L., R. A. Pielke Sr., J. O. Adegoke, S. D. Schubert, and P. J. Pegion, 2007: Investigation of the summer climate of the contiguous United States and Mexico using the Regional Atmospheric Modeling System (RAMS). Part II: Model climate variability. J. Climate, 20 , 3866–3887.
Comrie, A. C., and E. C. Glenn, 1998: Principal components-based regionalization of precipitation regimes across the southwest United States and northern Mexico, with an application to monsoon precipitation variability. Climate Res., 10 , 201–215.
Cook, E. R., D. M. Meko, D. W. Stahel, and M. K. Cleaveland, 1999: Drought reconstructions for the continental United States. J. Climate, 12 , 1145–1162.
Di Luzio, M., G. L. Johnson, C. Daly, J. K. Eischeid, and J. G. Arnold, 2008: Constructing retrospective gridded daily precipitation and temperature datasets for the conterminous United States. J. Appl. Meteor. Climatol., 47 , 476–497.
Duda, R. O., and P. E. Hart, 1973: Pattern Classification and Scene Analysis. Wiley, 482 pp.
Easterling, D. R., T. R. Karl, E. H. Mason, P. Y. Hughes, and D. P. Bowman, 1996: United States Historical Climatology Network (U.S. HCN): Monthly temperature and precipitation data. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory Rep. ORNL/CDIAC-87, NDP-019/R3, 83 pp.
Eischeid, J. K., C. B. Baker, T. R. Karl, and H. F. Diaz, 1995: The quality control of long-term climatological data using objective data analysis. J. Appl. Meteor., 34 , 2787–2795.
Eischeid, J. K., C. B. Baker, T. R. Karl, and H. F. Diaz, 2000: Creating a serially complete, national daily time series of temperature and precipitation for the western United States. J. Appl. Meteor., 39 , 1580–1591.
Gong, X., and M. B. Richman, 1995: On the application of cluster analysis to growing season precipitation data in North America east of the Rockies. J. Climate, 8 , 897–931.
Goswami, B. N., V. Venugopal, D. Sengupta, M. S. Madhusoodanan, and P. K. Xavier, 2006: Increasing trend of extreme rain events over India in a warming environment. Science, 314 , 1442–1445.
Gregory, J. M., T. M. L. Wigley, and P. D. Jones, 1993: Application of Markov models to area-average daily precipitation series and interannual variability in seasonal totals. Climate Dyn., 8 , 299–310.
Groisman, P. Ya, R. W. Knight, T. R. Karl, D. R. Easterling, B. Sun, and J. H. Lawrimore, 2004: Contemporary changes of the hydrological cycle over the contiguous United States: Trends derived from in situ observations. J. Hydrometeor., 5 , 64–85.
Groisman, P. Ya, R. W. Knight, D. R. Easterling, T. R. Karl, G. C. Hegerl, and V. N. Razuvaev, 2005: Trends in intense precipitation in the climate record. J. Climate, 18 , 1326–1350.
Guirguis, K. J., and R. Avissar, 2008: A precipitation climatology and dataset intercomparison for the western United States. J. Hydrometeor., 9 , 825–841.
Hamlet, A. F., and D. P. Lettenmaier, 2005: Production of temporally consistent gridded precipitation and temperature fields for the continental United States. J. Hydrometeor., 6 , 330–336.
Harrison, S. P., J. E. Kutzbach, Z. Liu, P. J. Bartlein, D. Mubs, I. C. Prentice, and R. Thompson, 2003: Mid-Holocene climates of the Americas: A dynamical response to changed seasonality. Climate Dyn., 20 , 663–688.
Hennessy, K. J., R. Suppiah, and C. M. Page, 1999: Australian rainfall changes, 1910–1995. Aust. Meteor. Mag., 48 , 1–13.
Higgins, R. W., Y. Yao, and X. L. Wang, 1997: Influence of the North American monsoon system on the U.S. summer precipitation regime. J. Climate, 10 , 2600–2622.
Karl, T. R., and R. W. Knight, 1998: Secular trends of precipitation amount, frequency, and intensity in the United States. Bull. Amer. Meteor. Soc., 79 , 231–241.
Katz, R. W., 1977: Precipitation as a chain-dependent process. J. Appl. Meteor., 16 , 671–676.
Katz, R. W., and M. B. Parlange, 1998: Overdispersion phenomenon in stochastic modeling of precipitation. J. Climate, 11 , 591–601.
Knowles, N., M. D. Dettinger, and D. R. Cayan, 2006: Trends in snowfall versus rainfall in the western United States. J. Climate, 19 , 4545–4559.
May, W., 2004: Variability and extremes of daily rainfall during the Indian summer monsoon in the period 1901–1989. Global Planet. Change, 44 , 83–105.
Meko, D. M., C. M. Woodhouse, C. A. Baisan, T. Knight, J. J. Lukas, M. K. Hughes, and M. W. Salzer, 2007: Medieval drought in the upper Colorado River Basin. Geophys. Res. Lett., 34 , L10705. doi:0.1029/2007GL029988.
Milly, P. C. D., K. A. Dunne, and A. V. Vecchia, 2005: Global pattern of trends in streamflow and water availability in a changing climate. Nature, 438 , 347–350.
Mock, C. J., and A. R. Brunelle-Daines, 1999: A modern analogue of western United States summer palaeoclimate at 6000 years before present. Holocene, 9 , 541–545.
NCDC, 1994: Time bias corrected divisional temperature–precipitation–drought index. Documentation for dataset TD-9640, 12 pp.
Oki, T., and S. Kanae, 2006: Global hydrological cycles and world water resources. Science, 313 , 1068–1072.
Petersen, K. L., 1994: A warm and wet little climatic optimum and a cold and dry little ice age in the Southern Rocky Mountains, U.S.A. Climatic Change, 26 , 243–269.
Peterson, T. C., C. Folland, G. Gruza, W. Hogg, A. Mokssit, and N. Plummer, 2001: Report on the activities of the working group on climate change detection and related rapporteurs 1998–2001. WMO Rep. WCDMP-47, WMO-TD 1071, 143 pp. [Available online at http://eprints.soton.ac.uk/30144/01/048_wgccd.pdf].
Poore, R. Z., M. J. Pavich, and H. D. Grissino-Mayer, 2005: Record of the North American southwest monsoon from Gulf of Mexico sediment cores. Geology, 33 , 209–212.
Reek, T., S. R. Doty, and T. W. Owen, 1992: A deterministic approach to the validation of historical daily temperature and precipitation data from the cooperative network. Bull. Amer. Meteor. Soc., 73 , 754–762.
Salzer, M. W., and K. F. Kipfmueller, 2005: Reconstructed temperature and precipitation on a millennial timescale from tree-rings in the southern Colorado plateau, U.S.A. Climatic Change, 70 , 465–487.
Sarle, W. S., 1983: The cubic clustering criterion. SAS Tech. Rep. A-108, 56 pp.
Schubert, S. D., M. J. Suarez, P. J. Pegion, R. D. Koster, and J. T. Bacmeister, 2004: On the cause of the 1930s Dust Bowl. Science, 303 , 1855–1859.
Seager, R., and Coauthors, 2007: Model projections of an imminent transition to a more arid climate in southwestern North America. Science, 316 , 1181–1184.
Shea, D. J., and R. A. Madden, 1990: Potential for long-range prediction of monthly mean surface temperatures over North American. J. Climate, 3 , 1444–1451.
Singh, S. W., and R. H. Kripalani, 1986: Potential predictability of lower-tropospheric monsoon circulation and rainfall over India. Mon. Wea. Rev., 114 , 758–763.
Solomon, S., D. Qin, M. Manning, M. Marquis, K. Averyt, M. M. B. Tignor, H. L. Miller Jr., and Z. Chen, 2007: Climate Change 2007: The Physical Science Basis. Cambridge University Press, 996 pp.
Stahle, D. W., E. R. Cook, M. K. Cleaveland, M. D. Therrell, D. M. Meko, H. D. Grissino-Mayer, E. Watson, and B. H. Luckman, 2000: Tree-ring data document 16th century megadrought over North America. Eos, Trans. Amer. Geophys. Union, 81 , 121–125.
Swift, L. W., and H. T. Schreuder, 1981: Fitting daily precipitation amounts using the SB distribution. Mon. Wea. Rev., 109 , 2535–2541.
Wang, J., B. T. Anderson, and G. Salvucci, 2006: Stochastic modeling of daily summertime rainfall over the southwestern United States. Part I: Interannual variability. J. Hydrometeor., 7 , 739–754.
Ward, J. H., 1963: Hierarchical grouping to optimize an objective junction. J. Amer. Stat. Assoc., 58 , 236–244.
Wilks, D. S., 1999: Interannual variability and extreme-value characteristics of several stochastic daily precipitation models. Agric. For. Meteor., 93 , 153–169.
Woodhouse, C. A., S. T. Gray, and D. M. Meko, 2006: Updated streamflow reconstructions for the upper Colorado River Basin. Water Resour. Res., 42 , W05415. doi:10.1029/2005WR004455.
Spatial clusters of stations having similar year-to-year variations in July–September precipitation (see text for details). Numbers represent stations belonging to the same cluster. Values are arbitrary and are used for identification purposes only. Contours represent topography plotted at 1000-m intervals starting at 1000 m.
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
Spatial clusters of stations having similar year-to-year variations in July–September precipitation (see text for details). Numbers represent stations belonging to the same cluster. Values are arbitrary and are used for identification purposes only. Contours represent topography plotted at 1000-m intervals starting at 1000 m.
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
Spatial clusters of stations having similar year-to-year variations in July–September precipitation (see text for details). Numbers represent stations belonging to the same cluster. Values are arbitrary and are used for identification purposes only. Contours represent topography plotted at 1000-m intervals starting at 1000 m.
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
Observed variations in (a),(b) precipitation accumulation (mm), (c),(d) frequency (events year−1), and (e),(f) coverage (percentage of stations experiencing rainfall during a given event) for stations in regions 1 and 2 for the period 1931–2000. All of the time series are shifted such that the initial 70-yr trend line (thick black line) starts at 0. Thick-dashed (gray) lines are 50-yr (25-yr) trend lines. On the RHS of each panel is the PDF of trends returned by 14 200 seventy-year time sequences of stochastic daily rainfall simulations with fixed occurrence, coverage, and intensity characteristics. Indicated are the ±2.5-percentile threshold (dotted lines), ±5-percentile threshold (dashed lines), and observed 70-yr trend (thick black lines).
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
Observed variations in (a),(b) precipitation accumulation (mm), (c),(d) frequency (events year−1), and (e),(f) coverage (percentage of stations experiencing rainfall during a given event) for stations in regions 1 and 2 for the period 1931–2000. All of the time series are shifted such that the initial 70-yr trend line (thick black line) starts at 0. Thick-dashed (gray) lines are 50-yr (25-yr) trend lines. On the RHS of each panel is the PDF of trends returned by 14 200 seventy-year time sequences of stochastic daily rainfall simulations with fixed occurrence, coverage, and intensity characteristics. Indicated are the ±2.5-percentile threshold (dotted lines), ±5-percentile threshold (dashed lines), and observed 70-yr trend (thick black lines).
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
Observed variations in (a),(b) precipitation accumulation (mm), (c),(d) frequency (events year−1), and (e),(f) coverage (percentage of stations experiencing rainfall during a given event) for stations in regions 1 and 2 for the period 1931–2000. All of the time series are shifted such that the initial 70-yr trend line (thick black line) starts at 0. Thick-dashed (gray) lines are 50-yr (25-yr) trend lines. On the RHS of each panel is the PDF of trends returned by 14 200 seventy-year time sequences of stochastic daily rainfall simulations with fixed occurrence, coverage, and intensity characteristics. Indicated are the ±2.5-percentile threshold (dotted lines), ±5-percentile threshold (dashed lines), and observed 70-yr trend (thick black lines).
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
As in Fig. 2, but for (a),(b) the observed variations in extreme dry-spell duration (days) and (c),(d) the July–September accumulated precipitation generated from intense storms (mm) for stations in regions 1 and 2.
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
As in Fig. 2, but for (a),(b) the observed variations in extreme dry-spell duration (days) and (c),(d) the July–September accumulated precipitation generated from intense storms (mm) for stations in regions 1 and 2.
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
As in Fig. 2, but for (a),(b) the observed variations in extreme dry-spell duration (days) and (c),(d) the July–September accumulated precipitation generated from intense storms (mm) for stations in regions 1 and 2.
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
As in Figs. 2a,b, but for the observed variations in precipitation accumulation (mm) reconstructed from NCDC Climate Division data in regions 1 and 2 for the period 1931–2008 (see text for details).
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
As in Figs. 2a,b, but for the observed variations in precipitation accumulation (mm) reconstructed from NCDC Climate Division data in regions 1 and 2 for the period 1931–2008 (see text for details).
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
As in Figs. 2a,b, but for the observed variations in precipitation accumulation (mm) reconstructed from NCDC Climate Division data in regions 1 and 2 for the period 1931–2008 (see text for details).
Citation: Journal of Climate 23, 7; 10.1175/2009JCLI3317.1
Trends (century−1) in the regional July–September precipitation characteristics, including precipitation accumulation (mm), frequency (events year−1), coverage (percentage of stations experiencing rainfall during a given event), extreme dry-spell duration (number of days), and intense storm accumulation (mm) based upon the period 1931–2000 for stations in regions 1–4, as identified in Fig. 1. Bold text indicates observed trends significant at the 95% level.
