Package {ClimateAnomaly}

ClimateAnomaly: Rainfall and Temperature Anomaly Indices for Climate Analysis

Description

ClimateAnomaly provides a unified interface for computing six widely-used rainfall anomaly indices and a standardised temperature anomaly from any user-supplied monthly or annual climate data frame.

Details

Indices available

The convenience wrapper calc_all_indices computes all of the above in a single call on a user-supplied data frame, grouped by any combination of spatial and temporal identifiers.

Typical workflow

library(ClimateAnomaly)

# Load your own data (must have at least a precipitation column
# plus grouping columns such as state/district/grid and month)
my_data <- read.csv("my_climate_data.csv")

result <- calc_all_indices(
  data       = my_data,
  group_cols = c("state", "month_num"),   # or c("district", "month_num")
  rain_col   = "rainfall_mm",
  tmax_col   = "tmax_c",                  # NULL to skip temperature
  tmin_col   = "tmin_c"
)
  

Data format requirements

The input data frame must contain:

• One or more grouping columns (e.g. state, month_num).

• A precipitation column (numeric, mm).

• Optionally, tmax and tmin columns (numeric, degrees Celsius).

No fixed column names are required; you pass the actual names as arguments.

Author(s)

Maintainer: Ranjit Kumar Paul ranjitstat@gmail.com

Authors:

• Ranjit Kumar Paul ranjitstat@gmail.com

• Md Yeasin

• Pushkar Bora

References

Van Rooy, M.P. (1965). A rainfall anomaly index independent of time and space. Notos, 14, 97–104.

Ekpetere, K.O. (2025). Development and evaluation of the modified and standardized rainfall anomaly indices for extreme variability analysis. Journal of Environmental Management, 375, 124160. doi:10.1016/j.jenvman.2025.124160

Raziei, T. (2021). Revisiting the Rainfall Anomaly Index to serve as a Simplified Standardized Precipitation Index. Journal of Hydrology, 602, 126761. doi:10.1016/j.jhydrol.2021.126761

Raziei, T. (2022). Improving the normalization procedure of the simplified standardized precipitation index (SSPI) using Box-Cox transformation. Stochastic Environmental Research and Risk Assessment, 37, 925–951. doi:10.1007/s00477-022-02317-9

McKee, T.B., Doesken, N.J., & Kleist, J. (1993). The relationship of drought frequency and duration to time scales. Proc. 8th Conf. Applied Climatology, AMS, 179–184.

Compute all rainfall and temperature anomaly indices

Description

This is the main user-facing function. It accepts a data frame (or data.table / tibble) with monthly (or annual) climate records and computes all six anomaly indices in one call, stratified by any combination of grouping columns the user specifies.

Usage

calc_all_indices(
  data,
  group_cols = c("state", "month_num"),
  rain_col = "rainfall_mm",
  tmax_col = "tmax_mean_c",
  tmin_col = "tmin_mean_c",
  indices = c("RAI", "MRAI", "SRAI", "SSPI", "SPI", "TEMP"),
  n_extreme = 10L,
  classify = TRUE
)

Arguments

Details

The function applies each index to every group defined by group_cols. In the typical use case, group_cols = c("state", "month_num") so that every index is computed relative to the long-term climatology of each state in each calendar month.

The user must supply column names for:

• precipitation (rain_col)

• maximum temperature (tmax_col), optional

• minimum temperature (tmin_col), optional

Columns that are NULL are simply skipped and not added to the output.

Value

The original data with additional columns appended:

RAI, RAI_class

Rainfall Anomaly Index + label

MRAI, MRAI_class

Modified RAI + label

SRAI, SRAI_class

Standardized RAI + label

SSPI, SSPI_class

Simplified SPI + label

SPI, SPI_class

Standardized Precipitation Index + label

tmax_anomaly, tmax_anom_class

Tmax z-score + label (if tmax_col supplied)

tmin_anomaly, tmin_anom_class

Tmin z-score + label (if tmin_col supplied)

tavg_c

Mean of the tmax and tmin columns (if both T columns supplied)

tavg_anomaly, tavg_anom_class

Tavg z-score + label (if both T columns supplied)

References

Van Rooy, M.P. (1965). Notos, 14, 97–104.
Ekpetere, K.O. (2025). J. Env. Mgmt., 375, 124160. doi:10.1016/j.jenvman.2025.124160
Raziei, T. (2021). J. Hydrology, 602, 126761. doi:10.1016/j.jhydrol.2021.126761
McKee, T.B. et al. (1993). 8th Conf. Applied Climatology, AMS, 179–184.

See Also

calc_rai, calc_mrai, calc_srai, calc_sspi, calc_spi, calc_temp_anomaly

Examples

# --- Example with built-in sample dataset ---
data(india_monthly)

result <- calc_all_indices(
  data       = india_monthly,
  group_cols = c("state", "month_num"),
  rain_col   = "rainfall_mm",
  tmax_col   = "tmax_mean_c",
  tmin_col   = "tmin_mean_c"
)

# View first few rows for one state
subset(result, state == "Maharashtra" & year >= 2015,
       select = c("state","year","month_num","rainfall_mm",
                  "RAI","MRAI","SPI","tmax_anomaly"))

# --- Compute only RAI and SPI (skip temp) ---
result2 <- calc_all_indices(
  data       = india_monthly,
  group_cols = c("state", "month_num"),
  rain_col   = "rainfall_mm",
  tmax_col   = NULL,
  tmin_col   = NULL,
  indices    = c("RAI", "SPI")
)

Modified Rainfall Anomaly Index (MRAI)

Description

Computes the Modified Rainfall Anomaly Index (MRAI) proposed by Ekpetere (2025). Unlike RAI, MRAI uses the absolute minimum and maximum of the full record as reference thresholds, removing the arbitrary choice of the top/bottom 10 values and ensuring the index always lies within [-3, +3].

Usage

calc_mrai(x)

Arguments

Details

MRAI = \begin{cases} +3\,\dfrac{P_n - \bar{P}}{P_{\max} - \bar{P}} & \text{if } P_n \geq \bar{P} \\[6pt] -3\,\dfrac{P_n - \bar{P}}{P_{\min} - \bar{P}} & \text{if } P_n < \bar{P} \end{cases}

Value

Numeric vector of MRAI values in [-3, +3].

References

Ekpetere, K.O. (2025). Development and evaluation of the modified and standardized rainfall anomaly indices for extreme variability analysis. Journal of Environmental Management, 375, 124160. doi:10.1016/j.jenvman.2025.124160

See Also

calc_rai, calc_srai

Examples

set.seed(42)
rain <- c(120, 85, 200, 60, 150, 95, 170, 40, 110, 130,
          75, 220, 90, 160, 50, 140, 100, 180, 70, 115)
mrai_values <- calc_mrai(rain)
data.frame(rain = rain, MRAI = round(mrai_values, 3))

Rainfall Anomaly Index (RAI)

Description

Computes the Rainfall Anomaly Index (RAI) introduced by Van Rooy (1965). The index ranks positive and negative precipitation anomalies through a normalisation that uses the mean of the n_extreme highest values (\bar{M}) and the mean of the n_extreme lowest values (\bar{X}) as reference thresholds.

Usage

calc_rai(x, n_extreme = 10L)

Arguments

Details

RAI = \begin{cases} +3\,\dfrac{N - \bar{N}}{\bar{M} - \bar{N}} & \text{if } N \geq \bar{N} \\[6pt] -3\,\dfrac{N - \bar{N}}{\bar{X} - \bar{N}} & \text{if } N < \bar{N} \end{cases}

where N is the current value, \bar{N} is the long-term mean, \bar{M} is the mean of the n_extreme highest values, and \bar{X} is the mean of the n_extreme lowest values.

Value

Numeric vector of RAI values (same length and order as x).

References

Van Rooy, M.P. (1965). A rainfall anomaly index independent of time and space. Notos, 14, 97–104.

See Also

calc_mrai, calc_srai, calc_all_indices

Examples

set.seed(42)
rain <- c(120, 85, 200, 60, 150, 95, 170, 40, 110, 130,
          75, 220, 90, 160, 50, 140, 100, 180, 70, 115)
rai_values <- calc_rai(rain)
data.frame(rain = rain, RAI = round(rai_values, 3))

Standardized Precipitation Index (SPI)

Description

Computes the Standardized Precipitation Index (SPI) of McKee et al. (1993). A Gamma distribution is fitted via maximum-likelihood estimation to the non-zero precipitation values; zero-inflation is handled explicitly through the mixed probability H(x) = q + (1-q) \cdot F_\Gamma(x). The resulting cumulative probability is converted to a standard normal z-score, which is the SPI value.

Usage

calc_spi(x)

Arguments

Details

If fewer than 4 non-zero observations are available (e.g. an arid month with almost no rain), the function falls back to a simple z-score (x - \bar{x}) / s.

Value

Numeric vector of SPI values (standard normal scores).

References

McKee, T.B., Doesken, N.J., & Kleist, J. (1993). The relationship of drought frequency and duration to time scales. Proceedings of the 8th Conference on Applied Climatology, 17–22 January 1993, Anaheim, CA. American Meteorological Society, pp. 179–184.

See Also

calc_sspi, calc_all_indices

Examples

set.seed(42)
rain <- c(120, 85, 200, 60, 150, 95, 170, 40, 110, 130,
          75, 220, 90, 160, 50, 140, 100, 180, 70, 115)
spi_values <- calc_spi(rain)
data.frame(rain = rain, SPI = round(spi_values, 3))

Standardized Rainfall Anomaly Index (SRAI)

Description

Computes the Standardized Rainfall Anomaly Index (SRAI) introduced by Ekpetere (2025). SRAI incorporates the standard deviation (\sigma) of the series to better detect extreme rainfall events relative to average conditions. Output is clamped to [-3, +3].

Usage

calc_srai(x, clamp = TRUE)

Arguments

Details

SRAI = \begin{cases} +3\,\dfrac{P_n - \sigma}{P_{\max} - \sigma} & \text{if } P_n \geq \bar{P} \\[6pt] -3\,\dfrac{P_n - \sigma}{P_{\min} - \sigma} & \text{if } P_n < \bar{P} \end{cases}

Value

Numeric vector of SRAI values.

References

Ekpetere, K.O. (2025). Development and evaluation of the modified and standardized rainfall anomaly indices for extreme variability analysis. Journal of Environmental Management, 375, 124160. doi:10.1016/j.jenvman.2025.124160

See Also

calc_mrai, calc_rai

Examples

set.seed(42)
rain <- c(120, 85, 200, 60, 150, 95, 170, 40, 110, 130,
          75, 220, 90, 160, 50, 140, 100, 180, 70, 115)
srai_values <- calc_srai(rain)
data.frame(rain = rain, SRAI = round(srai_values, 3))

Simplified Standardized Precipitation Index (SSPI)

Description

Computes the Simplified Standardized Precipitation Index (SSPI) as proposed by Raziei (2021). SSPI is designed as a non-parametric alternative to SPI that handles zero-dominated precipitation series (arid/semi-arid climates) without fitting a probability distribution.

Usage

calc_sspi(x, eps = 1e-06)

Arguments

Details

The algorithm follows Equation (2) of Raziei (2021):

1. Compute the probability of zero precipitation (PZP).

2. Reduce skewness of the series:

   • If skewness > 0.5: apply cube-root transformation (x^{1/3}).

   • If skewness < -0.5: apply cube transformation (x^3).

   • Otherwise: no transformation.

3. Compute anomaly \Delta_i = P_i^* - \bar{P}^* in the transformed domain.

4. Derive the 5th (P_B) and 95th (P_T) percentiles of the ranked transformed data.

5. Apply the appropriate SSPI formula based on the sign of \Delta_i and whether PZP \geq 40%.

Value

Numeric vector of SSPI values.

References

Raziei, T. (2021). Revisiting the Rainfall Anomaly Index to serve as a Simplified Standardized Precipitation Index. Journal of Hydrology, 602, 126761. doi:10.1016/j.jhydrol.2021.126761

See Also

calc_spi, calc_all_indices

Examples

set.seed(42)
rain <- c(120, 85, 200, 60, 150, 95, 170, 40, 110, 130,
          75, 220, 90, 160, 50, 140, 100, 180, 70, 115)
sspi_values <- calc_sspi(rain)
data.frame(rain = rain, SSPI = round(sspi_values, 3))

Standardised Temperature Anomaly

Description

Computes the standardised temperature anomaly (z-score) for a single group:

Usage

calc_temp_anomaly(x)

Arguments

Details

Z_T = \frac{T_i - \bar{T}}{s_T}

where \bar{T} is the long-term mean temperature for the group and s_T is its sample standard deviation. The grouping is typically location \times calendar month so that, for example, every January at a given station is compared only against other Januaries.

Value

Numeric vector of standardised anomaly values (z-scores).

Examples

# Tmax values for one station, all Januaries 1980-2020
set.seed(7)
tmax_jan <- rnorm(41, mean = 28, sd = 2)
z <- calc_temp_anomaly(tmax_jan)
summary(z)

Classify RAI / MRAI / SRAI values

Description

Assigns a drought/wet severity class label to a numeric vector of RAI, MRAI, or SRAI values using the classification scheme of Van Rooy (1965) and Raziei (2021).

Usage

classify_rai(x)

Arguments

Details

Value

Character vector of class labels (same length as x).

References

Van Rooy, M.P. (1965). Notos, 14, 97–104. Raziei, T. (2021). Journal of Hydrology, 602, 126761.

Examples

vals <- c(-3.5, -2.1, -0.8, 0.1, 0.7, 1.5, 2.5, 3.2)
classify_rai(vals)

Classify SPI / SSPI values

Description

Assigns a drought/wet severity class label using the standard SPI classification (McKee et al., 1993).

Usage

classify_spi(x)

Arguments

Details

Value

Character vector of class labels.

References

McKee, T.B., Doesken, N.J., & Kleist, J. (1993). Proceedings of the 8th Conference on Applied Climatology, AMS, pp. 179–184.

Examples

vals <- c(-2.5, -1.6, -1.2, -0.7, 0.0, 0.6, 1.1, 1.7, 2.3)
classify_spi(vals)

Classify temperature anomaly values

Description

Assigns a temperature departure class label to standardised temperature anomaly z-scores.

Usage

classify_temp_anomaly(x)

Arguments

Details

Value

Character vector of class labels.

Examples

vals <- c(-2.5, -1.2, -0.6, 0.1, 0.7, 1.3, 2.1)
classify_temp_anomaly(vals)

Sample monthly climate data for Indian states (1976–2026)

Description

A subset of monthly state-level rainfall and temperature records for India, included as a demonstration dataset for the ClimateAnomaly package.

Usage

india_monthly

Format

A data frame with columns:

state

Character. State/territory name.

year

Integer. Year (1976–2026).

month_num

Integer. Calendar month (1 = January, 12 = December).

month

Character. Month label in "YYYY-MM" format.

rainfall_mm

Numeric. Monthly total rainfall (mm).

rainy_days

Integer. Number of rainy days in the month.

tmax_mean_c

Numeric. Mean daily maximum temperature (degrees Celsius).

tmin_mean_c

Numeric. Mean daily minimum temperature (degrees Celsius).

tmax_peak_c

Numeric. Peak daily maximum temperature (degrees Celsius).

tmin_low_c

Numeric. Lowest daily minimum temperature (degrees Celsius).

complete

Logical. Whether the month has complete data coverage.

source

Character. Data source identifier.

Source

Open-Meteo historical climate API (open-meteo.com).

Examples

data(india_monthly)
head(india_monthly)
table(india_monthly$state)