Seasonal Variability of the Observed and the Projected Daily Temperatures in Northern

Seasonal variability of the observed and the projected daily temperatures in northern Saudi Arabia

H. Athar

Centerof Excellence for Climate ChangeResearch, Department of Meteorology,

KingAbdulazizUniversity,

Jeddah, Saudi Arabia

SUPPLEMENTARY INFORMATION

______

Correspondence to: Dr. H. Athar, Centerof Excellence for Climate ChangeResearch, Department of Meteorology, KingAbdulazizUniversity, P.O. Box 80208, Jeddah21589, Saudi Arabia. E-mail:

S1 Quality control of station datasets

The quality control of the station-based daily maximum and minimum temperature datasets was performed in the following two steps: first numerical and visual checks were performed to identify whether any daily minimum temperature value is greater than the daily maximum temperature value, and then for each climatological mean value of the daily temperature variable, it was checked weather it falls within the range of 4 sigma standard deviations or not, following Zhang et al. (2005). The flagged outliers for each station were checked against several possible causes, on case by case basis. These include but not limited to, for instance, the occurrence and/or the passage of the short lived meso-scale high/low pressure systems, and the heating during the El-Niño years (such as during the year 1997 and 2002), and the spatial coherence of both of these, with respect to the nearest stations. Heating anomalies were noticeable in the re-analysis dataset (not shown). Occurrence of the rainfall resulting in relative temperature change was also considered, after taking into account the topographic details of the station’s location. Some of the above mentioned causes are the same as noted by AlSarmi and Washington (2011). A few unphysical values were set as missing values.

S1.1 Calculation of temperature extremes

The station-wise temperature extremes were computed from the observed dataset, and from the single run of the RCM output (PRECIS), to further support the study of variability in the temperature extremes on an individual station basis. A modified version of the definition given in Zhang et al. (2005) is used, since the nSA region is more of the arid climate nature rather than the Mediterranean climate (Fisher and Membery 1998). The lower threshold for the day time extreme temperature (TX) was taken to be > 35°C. The TX was also computed using a lower threshold of 25°C and the obtained results were compared with those by Zhang et al. (2005) and by Lelieveld et al. (2012), where applicable, and an agreement was found (not shown). Only 6 stations (including Hail, Dhahran, Riyadh, and Madina) were considered by Zhang et al. (2005), and 2 stations (Dhahran and Riyadh) were considered by Lelieveld et al. (2012), from SA for extreme temperatures analysis. The night time minimum temperature (TN) was computed, with TN > 25°C.

S2 Observed temperature extremes

The results for the observed and the climate model based temperature extremes are displayed in Table S5. For TX, for the inland stations, with relatively lower topographic heights, the observed diurnal temperature cycle is well resolved by the downscaled high resolution RCM (PRECIS). For TN, the model outputis in even better agreement with the observations. For Turaif (see Table S1), despite the elevated terrain, the PRECIS model performed better. Table S5 indicates that the day time temperature representation in mountainous areas is not well resolved by the RCM.

Additionally, another index related to the temperature extremes was also computed using both the observed and the climate model dataset (not shown), on station by station basis; the growing season length (GSL). Because of the relatively higher prevailing temperatures, the GSL values remains large (comparable to the number of calendar days in a year) and do not show any significant annual variability during the based period (1978–2008).

S3 A meteorological implication of results for human heat stress

Another possible meteorological implication of the increased occurrence of temperature extremes is the recurrent appearance of the mid-latitude blocking anti-cyclones (see, for instance, Lupo et al. 2008). The effect of prolonged Atlantic basin based mid-latitude high-pressure systems (blocking anti-cyclones) on the mean surface temperature of SA,for a 40-year period (1968–2007), was studied by Athar et al. (2012). It was noticed that the occurrence of mid-latitude blocking anti-cyclones leads to an upwards seasonal shift in the mean surface temperature in SA, depending upon the season, by about 1–1.5°C. The detailed adaptation and mitigation studies are thus called for to gauge the impacts of increased warmer temperatures in nSA/SA on human health and ecosystems.

References

AlSarmi S, Washington R(2011) Recent observed climate change over the Arabian Peninsula.J Geophys Res116:D11109. doi:10.1029/2010JD015459

Athar H, Almazroui M, Islam MN, Abid MA, Ehsan MA (2012) Effect of mid-latitude blocking anticyclones on the weather of the Arabian Peninsula. Int J Climatol. doi:10.1002/joc.3450

Fisher M, Membery DA (1998) Climate. In: Ghazanfar S, Fisher M (ed) Vegetation of the ArabianPeninsula, vol. 1, 1st edn. Kluwer Acad, Dordrecht, Netherland,pp 5–38

Lelieveld J, Hadjinicolaou P, Kostopoulou E, Chenoweth J, Giannakopoulos C,Hannides C, Lange MA, El Maayar M, Tanarhte M, Tyrlis E, Xoplaki E (2012)Climate change and impacts in the Eastern Mediterranean and the Middle East. Clim Chang. 114: 667–687.doi: 10.1007/s10584–012–0418–4

Lupo A, Clark J, Hendin A, Kelly A, Mihalka K, Perrin B, PuricelliK(2008) The global increase in blocking occurrences. Preprint, The 20th Conference on Global Climate Change, New Orleans, LA, USA, American Meteorological Society, P 2.14

Zhang X, Aguilar E, Sensoy S, Melkonyan H, Tagiyeva U, Ahmed N, Kutaladze N, Rahimzadeh F, Taghipour A, Hantosh TH, Albert P, Semawi M, Ali MK, Al-Shabibi MHS, Al-Oulan Z, Zatari T, Khelet IAlD, Hamoud S, Sagir R, Demircan M, Eken M, Adiguzel M, Alexander L, Peterson TC, Wallis T (2005) Trends in Middle East climate extreme indices from 1950 to 2003. J Geophys Res 110:D22104

Table S1 The detailed description of the stations in nSA used in this study

No. / Station name / ICAO code / WMOcode / Lat (°N) / Lon (°E) / Elevation (m)
1 / Turaif / 40356 / OETR / 31.68 / 38.73 / 852
2 / Al-Jouf / 40361 / OESK / 29.78 / 40.10 / 670
3 / Tabuk / 40375 / OETB / 28.37 / 36.60 / 770
4 / Al-Qaysumah / 40373 / OEPA / 28.33 / 46.12 / 360
5 / Hail / 40394 / OEHL / 27.44 / 41.69 / 1000
6 / Gassim / 40405 / OEGS / 26.30 / 43.77 / 648
7 / Dhahran / 40416 / OEDR / 26.26 / 50.16 / 22
8 / Wejh / 40400 / OEWJ / 26.20 / 36.47 / 20
9 / Riyadh Old / 40438 / OERY / 24.71 / 46.73 / 610
10 / Madina / 40430 / OEMA / 24.54 / 39.70 / 630
11 / Yenbo / 40439 / OEYN / 24.14 / 38.06 / 8

Table S2The difference of 10th percentile (ΔP10), 50th percentile (ΔP50), 90thpercentile (ΔP90), mean value (ΔMean), model standard deviation value (ΔStd), and skewness (ΔSkewness), for the PDFs based on the simulations of the PRECIS, the CM2.0, and the CM2.1models relative to the observed PDF of the base period (1978–2008), for the minimum temperature

Model / Statistic / DJF / MAM / JJA / SON
PRECIS
ΔP10 (°C) / -4.01 / -2.40 / -0.51 / -2.81
ΔP50 (°C) / -3.73 / -1.11 / 0.39 / -1.75
ΔP90 (°C) / -0.76 / 0.19 / 3.46 / 0.86
ΔMean (°C) / -2.99 / -1.05 / 0.75 / -1.31
ΔStd (°C) / 1.09 / 1.07 / 1.29 / 1.25
ΔSkewness / 0.32 / 0.18 / 0.42 / 0.36
CM2.0
ΔP10 (°C) / -5.61 / -6.72 / -5.73 / -8.07
ΔP50 (°C) / -6.72 / -6.83 / -4.52 / -6.61
ΔP90 (°C) / -6.98 / -6.10 / -3.20 / -4.58
ΔMean (°C) / -6.38 / -6.57 / -4.51 / -6.41
ΔStd (°C) / -0.52 / 0.34 / 1.18 / 1.22
ΔSkewness / 0.32 / 0.24 / 0.19 / 0.32
CM2.1
ΔP10 (°C) / -4.73 / -5.07 / -3.80 / -7.08
ΔP50 (°C) / -5.02 / -4.09 / -1.79 / -4.87
ΔP90 (°C) / -2.58 / -2.62 / 0.51 / -1.23
ΔMean (°C) / -4.16 / -3.91 / -1.74 / -4.40
ΔStd (°C) / 0.73 / 0.98 / 1.68 / 2.03
ΔSkewness / 0.36 / 0.19 / 0.00 / 0.37

Table S3Same as Table S2, except for mean temperature

Model / Statistic / DJF / MAM / JJA / SON
PRECIS
ΔP10 (°C) / -2.30 / -0.63 / 0.86 / -2.06
ΔP50 (°C) / -2.49 / -0.34 / 1.44 / -1.46
ΔP90 (°C) / -2.59 / -0.60 / 0.79 / -0.53
ΔMean (°C) / -2.49 / -0.45 / 1.12 / -1.38
ΔStd (°C) / -0.11 / 0.02 / 0.00 / 0.55
ΔSkewness / 0.00 / -0.04 / -0.33 / 0.08
CM2.0
ΔP10 (°C) / -5.22 / -5.41 / -4.58 / -7.27
ΔP50 (°C) / -5.70 / -5.41 / -3.39 / -5.71
ΔP90 (°C) / -5.38 / -5.01 / -1.23 / -3.15
ΔMean (°C) / -5.49 / -5.27 / -3.11 / -5.41
ΔStd (°C) / 0.03 / 0.26 / 1.32 / 1.46
ΔSkewness / 0.15 / 0.20 / 0.33 / 0.38
CM2.1
ΔP10 (°C) / -3.91 / -3.58 / -1.85 / -5.76
ΔP50 (°C) / -3.77 / -3.07 / -0.20 / -3.78
ΔP90 (°C) / -2.03 / -2.22 / 1.71 / -0.06
ΔMean (°C) / -3.34 / -2.93 / -0.13 / -3.32
ΔStd (°C) / 0.70 / 0.60 / 1.35 / 1.91
ΔSkewness / 0.16 / 0.22 / -0.06 / 0.34

Table S4The 10th percentile (P10), 50th percentile (P50), 90thpercentile (P90), mean, standard deviation (Std), and the skewness for thePRECIS based PDFs (displayed in Fig. S3, Fig. S4, and Fig. 3), on a seasonal basis, for all the three temperature variables (minimum temperature, abbreviated as Tmin, and like wise) for the first (2070–2079) and the last decade (2090–2099), and their decadal differences, during the eighties (2070–2099), respectively

Season / Period / P10 (°C) / P50 (°C) / P90 (°C) / Mean(°C) / Std (°C) / Skewness
Tmin
DJF / 2070–2079 / 0.50 / 7.90 / 17.34 / 8.44 / 6.40 / 0.20
2090–2099 / 0.68 / 8.26 / 17.87 / 8.78 / 6.48 / 0.18
Δ(Last-First) / 0.18 / 0.37 / 0.53 / 0.34 / 0.09 / -0.01
MAM / 2070–2079 / 9.08 / 19.11 / 28.32 / 18.87 / 7.31 / -0.03
2090–2099 / 10.23 / 20.82 / 29.31 / 20.28 / 7.20 / -0.15
Δ(Last-First) / 1.15 / 1.71 / 0.99 / 1.40 / -0.11 / -0.11
JJA / 2070–2079 / 24.27 / 29.80 / 37.54 / 30.21 / 4.83 / 0.20
2090–2099 / 25.22 / 31.18 / 38.65 / 31.52 / 5.02 / 0.12
Δ(Last-First) / 0.95 / 1.38 / 1.11 / 1.30 / 0.19 / -0.08
SON / 2070–2079 / 11.28 / 22.32 / 31.19 / 21.74 / 7.28 / -0.15
2090–2099 / 13.73 / 23.44 / 32.64 / 23.23 / 7.06 / -0.06
Δ(Last-First) / 2.45 / 1.12 / 1.45 / 1.49 / -0.22 / 0.09
Tmean
DJF / 2070–2079 / 8.19 / 15.23 / 21.66 / 15.05 / 5.14 / -0.12
2090–2099 / 8.47 / 15.72 / 22.19 / 15.47 / 5.25 / -0.09
Δ(Last-First) / 0.28 / 0.49 / 0.53 / 0.42 / 0.11 / 0.03
MAM / 2070–2079 / 17.70 / 26.65 / 34.51 / 26.32 / 6.31 / -0.26
2090–2099 / 18.92 / 28.08 / 35.62 / 27.71 / 6.19 / -0.35
Δ(Last-First) / 1.22 / 1.43 / 1.12 / 1.39 / -0.12 / -0.08
JJA / 2070–2079 / 33.03 / 38.19 / 41.29 / 37.60 / 3.25 / -0.68
2090–2099 / 34.07 / 39.24 / 43.12 / 38.85 / 3.54 / -0.51
Δ(Last-First) / 1.04 / 1.05 / 1.83 / 1.25 / 0.29 / 0.17
SON / 2070–2079 / 18.34 / 28.98 / 37.07 / 28.40 / 6.90 / -0.35
2090–2099 / 20.80 / 30.39 / 38.16 / 29.89 / 6.61 / -0.34
Δ(Last-First) / 2.47 / 1.41 / 1.09 / 1.49 / -0.29 / 0.01
Tmax
DJF / 2070–2079 / 15.78 / 22.02 / 29.19 / 22.21 / 5.12 / 0.14
2090–2099 / 16.28 / 22.57 / 29.45 / 22.73 / 5.16 / 0.19
Δ(Last-First) / 0.51 / 0.55 / 0.26 / 0.52 / 0.03 / 0.05
MAM / 2070–2079 / 24.52 / 33.16 / 41.39 / 32.99 / 6.46 / -0.11
2090–2099 / 25.90 / 34.64 / 42.82 / 34.41 / 6.35 / -0.13
Δ(Last-First) / 1.38 / 1.48 / 1.43 / 1.42 / -0.11 / -0.01
JJA / 2070–2079 / 39.65 / 43.86 / 48.67 / 44.00 / 3.55 / -0.09
2090–2099 / 40.30 / 45.14 / 50.53 / 45.26 / 3.92 / -0.01
Δ(Last-First) / 0.66 / 1.28 / 1.86 / 1.26 / 0.37 / 0.08
SON / 2070–2079 / 25.11 / 35.80 / 44.61 / 35.15 / 7.25 / -0.23
2090–2099 / 27.20 / 37.21 / 45.33 / 36.65 / 6.93 / -0.32
Δ(Last-First) / 2.09 / 1.41 / 0.72 / 1.50 / -0.32 / -0.08

Table S5 The station-wise comparison of the temperature extremes (TX > 35°C and TN > 25°C) using the observed datasets, and the PRECIS datasets for the base period (1978–2008). The displayed values are number of days per year, average over the base period

Station name / TX > 35°C / TN > 25°C
OBS / PRECIS / OBS / PRECIS
Turaif / 79 / 91 / 198 / 198
Al-Jouf / 131 / 123 / 226 / 225
Tabuk / 134 / 96 / 245 / 221
Al-Qaysumah / 178 / 169 / 257 / 253
Hail / 131 / 123 / 238 / 229
Gassim / 172 / 171 / 266 / 265
Dhahran / 177 / 93 / 276 / 235
Wejh / 26 / 90 / 306 / 224
Riyadh Old / 177 / 173 / 278 / 271
Madina / 193 / 184 / 317 / 297
Yenbo / 175 / 182 / 352 / 305

Fig. S1 All nSA observed daily minimum temperature PDFs for the DJF, MAM, JJA and SON seasons (row wise from top to bottom), during the base period, for the first decade (1978–1987, black bars), and the last decade (1999–2008, red face bars). The left vertical line indicates the 10th percentile limit, whereas the right vertical line indicates the 90th percentile limit, both for the first decade

Fig. S2Same as Fig. S1, except for mean temperature

Fig. S3 All nSA PRECIS based daily minimum temperature PDFs for the DJF, MAM, JJA and SON seasons (row wise from top to bottom), for the base period (black bars), and for the 30-year periods centered in the twenties (red face bars), fifties (green face bars), and eighties (blue face bars). The left vertical line indicates the 10th percentile limit, whereas the right vertical line indicates the 90th percentile limit, both for the base period

Fig. S4Same as Fig. S3, except for mean temperature

Fig. S5 Same as Fig. S3, except for CM2.0 minimum temperature

Fig. S6Same as Fig. S3, except for CM2.0 mean temperature

1