NDMC Blends

Metadata

Blends

The Blends datasets are computed weekly using weighted sums of the ranked inputs. The results are then ranked to provide normalized 0.0-1.0 values for comparison.

Three blends are computed using the following inputs:

Short-Term:

  • 7% SPEI (9-mon. precipitation totals)
  • 13% Soil Moisture (0-100cm root zone: 1-week anomaly)
  • 20% SPI (1-mon. precip. totals)
  • 25% SPI (3-mon. precip. totals)
  • 35% SPEI (1-mon. precip. totals)

Long-Term:

  • 10% Soil Moisture (0-200cm total column: 12-week anomaly)
  • 10% SPI (60-mon. precipitation totals)
  • 15% SPI (6-mon. precip. totals)
  • 20% SPI (12-mon. precip. totals)
  • 20% SPI (24-mon. precip. totals)
  • 25% SPEI (9-mon. precip. totals)

Long-Term West:

  • 10% Soil Moisture (0-200cm total column: 12-week anomaly)
  • 10% SPI (12-mon. precip. totals)
  • 10% SPI (24-mon. precip. totals)
  • 10% SPI (60-mon. precip. totals)
  • 30% SPEI (9-mon. precip. totals)
  • 30% SPEI (60-mon. precip. totals)

Source

The blends were developed by the National Drought Mitigation Center. The period of record runs from 1980 to current. The data are projected in EPSG:4326 with the SW corner at 125 o W, 24.5o N. The cell spacing is 1/24 deg (4km) with 1393 columns and 601 rows.

Ranking

The inputs to the Blends, and the final Blends data are ranked on a 0.0-1.0 scale using the average ranking approach: a rescaled mean of the strict and weak rankings. Strict rankings compare values using a ‘greater than’ condition, whereas weak rankings use a ‘greater than or equal to’ condition.

Once both rankings have been computed, the mean of the values is taken, and the final range is determined. If the range is not 0.0 to 1.0 for any given cell, the data is rescaled for that cell for the minimum is 0.0 and the maximum is 1.0.

The average ranking method is the default setting for the SciPy ‘rankdata’ function.

Blends Categories

  • D4 Exceptional Drought (0.00 - 0.02)
  • D3 Extreme Drought (0.03 - 0.05)
  • D2 Severe Drought (0.06 - 0.10)
  • D1 Moderate Drought (0.11 - 0.20)
  • D0 Abnormally Dry (0.21 - 0.30)
  • Near Normal (0.31 - 0.70)
  • Abnormally Wet (0.71 - 0.80)
  • Moderately Wet (0.81 - 0.90)
  • Severely Wet (0.91 - 0.95)
  • Extremely Wet (0.96 - 0.98)
  • Exceptionally Wet (0.99 - 1.00)
  • No Data

Soil Moisture:

Source

The North American Land Data Assimilation System (NLDAS) Noah Land Surface Model version 2 produced by NASA, is used as the base input for the Soil Moisture parameter. The grid is 1/8° x 1/8° (12.5km x 12.5km) spacing and covers 25.0° to 53.0° latitude: -125.0° to -67.0° longitude for 1979 to present. The data contains four depth ranges of 0-10cm, 10-40cm, 40-100cm, and 100-200cm.

NLDAS Noah Summary Page

Processing

The data is rescaled to 1/24° x 1/24° (4km x 4km) to be properly blended with the other inputs. Since the grid box centers of the other inputs are not aligned with the Noah grid, the Noah data is shifted to use the same grid box centers. This is accomplished using an extended bilinear approach so that cells in the new spacing which center on the edges of the original boxes have the appropriate equivalent values.

The four input levels are depth weighted into our three internal zone values:

  • Root Zone 1 (0-40cm): 20% 0-10cm, 80% 10-40cm
  • Root Zone 2 (0-100cm): 10% 0-10cm, 30% 10-40cm, 60% 40-100cm
  • Total Column (0-200cm): 5% 0-10cm, 15% 10-40cm, 30% 40-100cm, 50% 100-200cm

Precipitation Indices:

Source

The Applied Climate Information System (ACIS) NRCC Interpolated (Grid 1) is used as the base input for the Standardized Precipitation Index (SPI) and Standardized Precipitation-Evapotranspiration Index (SPEI). The grid is 1/24° x 1/24° (4km x 4km) spacing and covers 24.0° to 50.0° latitude: -125.0° to -66.0° longitude for 1950 to present. ACIS is maintained by the NOAA Regional Climate Centers.

ACIS Grid Descriptions Page

Standardized Precipitation Index (SPI)

The SPI is calculated according to Edwards and McKee (1997) using the Gamma Distribution with maximum likelihood as shown in Guenang and Kamga (2014).

The fit parameters for the distribution are determined using the estimation defined in Thom (1958).

The cumulative distribution is calculated using the Python SciPy package Gamma CDF function, which uses the same integration process found in MATLAB.

The distribution is standardized according to Abramowitz and Stegun (1965) and limited to the range of -3.0 to 3.0

The grid size for the SPI is 1/24° x 1/24° (4km x 4km).

The SPI data are calculated by the National Drought Mitigation Center.

Standardized Precipitation Evapotranspiration Index (SPEI)

The SPEI is calculated according to Vincente-Serrano and Begueria (2010, 2014) using the 3-parameter Log Logistic Distribution (LLD3) with unbiased probability weighted moments.

The PWMs and L-Moments are calculated according to the methods described in Greenwood et al. (1979) and Hosking (1990).

The fit parameters and cumulative distribution are determined using methods described in Hosking (1990, 1997) following Hosking’s FORTRAN routines.

Code for the distribution process was verified using the R-language “lmomco” package and the following functions:

The distribution is standardized according to Abramowitz and Stegun (1965) and limited to the range of -3.0 to 3.0

The grid size for the SPEI is 1/24° x 1/24° (4km x 4km).

The SPEI data are calculated by the National Drought Mitigation Center.

References

  • Abramowitz, M. and I.A. Stegun, 1965: Handbook of Mathematical Functions. Dover Publications, New York.
  • Beguería, S., Vicente-Serrano, S.M., Fergus Reig, Borja Latorre. Standardized Precipitation Evapotranspiration Index (SPEI) revisited (2014): parameter fitting, evapotranspiration models, kernel weighting, tools, datasets and drought monitoring. International Journal of Climatology, 3 4: 3001-3023.
  • McKee, T. B., Doesken, N. J. and j. Kleist, 1993: The Relationship of Drought Frequency and Duration to TIme Scales, Eighth Conference on Applied Climatology, 17–22 January 1993, Anaheim, California [Available online at https://climate.colostate.edu/pdfs/relationshipofdroughtfrequency.pdf.]
  • Greenwood, J.A., Landwehr, J.M., Matalas, N.C., and Wallis, J.R., 1979, Probability weighted moments—Definition and relation to parameters of several distributions expressable in inverse form: Water Resources Research, v. 15, pp. 1,049–1,054.
  • Guenang, G. M., & Kamga, F. M. (2014). Computation of the Standardized Precipitation Index (SPI) and Its Use to Assess Drought Occurrences in Cameroon over Recent Decades, Journal of Applied Meteorology and Climatology, 53(10), 2310-2324.
  • Hosking, J.R.M., 1990, L-moments—Analysis and estimation of distributions using linear combinations of order statistics: Journal of the Royal Statistical Society, Series B, v. 52, pp. 105–124.
  • Hosking, J.R.M., and Wallis, J.R., 1997, Regional frequency analysis—An approach based on L-moments: Cambridge University Press.
  • Thom, H. C. S., 1958: A note on the gamma distribution. Mon. Wea. Rev., 86, 117–122.
  • Vicente-Serrano S.M., Santiago Beguería, Juan I. López-Moreno, (2010) A Multi-scalar drought index sensitive to global warming: The Standardized Precipitation Evapotranspiration Index - SPEI. Journal of Climate 23: 1696-1718.

Funding

The NDMC blends work is funded through a cooperative agreement with USDA’s Office of the Chief Economist.