Numerical Simulations of Airflow and Weather over the Island of Oahu Yi-Leng Chen and Hiep Van Nguyen.

Numerical Simulations of Numerical Simulations of Airflow and Weather over the Airflow and Weather over the

Island of OahuIsland of Oahu

Yi-Leng Chen and Hiep Van NguyenYi-Leng Chen and Hiep Van Nguyen

22

Total area is 594 sq. miles - 44 miles long and 30 miles wide. The Total area is 594 sq. miles - 44 miles long and 30 miles wide. The

highest point is Kaala Peak with an elevation of 4,003 ft above sea-level.highest point is Kaala Peak with an elevation of 4,003 ft above sea-level.

33

OutlineOutline IntroductionIntroduction Model descriptionModel description Effects of terrain and diurnal heating cycle Effects of terrain and diurnal heating cycle on on

island weatherisland weather Effects of terrain Effects of terrain Effects of diurnal heating cycleEffects of diurnal heating cycle

Effects of trade wind conditions Effects of trade wind conditions on island-scale on island-scale flow and weather flow and weather

Strong tradesStrong trades Weak tradesWeak trades

Sensitivity tests Sensitivity tests on terrain and diurnal heating on terrain and diurnal heating effectseffects

ConclusionsConclusions

44

IntroductionIntroduction

Effects on airflow: Effects on airflow: The mountain acts as a barrier to the approaching The mountain acts as a barrier to the approaching

airflow; and airflow; and As a heat source (sink) during the day (night) As a heat source (sink) during the day (night)

(Leopold 1948, 1949; Chen and Nash 1994) (Leopold 1948, 1949; Chen and Nash 1994) resulted in mountain-valley winds and land-sea resulted in mountain-valley winds and land-sea breezes (Leopold 1948 ) breezes (Leopold 1948 )

Wakes off the lee-side coastWakes off the lee-side coast

Despite its relatively small size, the island terrain of Despite its relatively small size, the island terrain of Oahu has profound influences on island-scale airflow Oahu has profound influences on island-scale airflow and weatherand weather

55

IntroductionIntroduction

Diurnal cycle of winds over Oahu:Diurnal cycle of winds over Oahu: Throughout the diurnal cycle, the surface winds Throughout the diurnal cycle, the surface winds

are weaker than over the open ocean except over are weaker than over the open ocean except over the northwestern and southeastern corners the northwestern and southeastern corners (Leopold, 1948; Ramage and Oshiro 1977) (Leopold, 1948; Ramage and Oshiro 1977)

A maximum wind speed in the afternoonA maximum wind speed in the afternoon, due to , due to transfer of momentum from aloft (Leopold, 1948)transfer of momentum from aloft (Leopold, 1948)

Complete reversal of wind direction occurs on the Complete reversal of wind direction occurs on the Waianae coast in the afternoon (Leopold,1948) Waianae coast in the afternoon (Leopold,1948)

66

IntroductionIntroduction Island effects on Oahu rainfall Island effects on Oahu rainfall

The trade‑wind belt has a minimum in the global The trade‑wind belt has a minimum in the global distribution of rainfall. However, trade-wind distribution of rainfall. However, trade-wind showers are frequent over the Hawaiian Islands showers are frequent over the Hawaiian Islands because of terrain and local winds because of terrain and local winds

Diurnal cycle of rainfall over Oahu:Diurnal cycle of rainfall over Oahu: Windward and mauka areas: nocturnal and morning Windward and mauka areas: nocturnal and morning

rainfall maxima with a minimum in the afternoon rainfall maxima with a minimum in the afternoon (Loveridge, 1924; Leopold, 1948; Schroeder 1977; Loos (Loveridge, 1924; Leopold, 1948; Schroeder 1977; Loos 2004)2004)

Waianae lee-side coast: afternoon maximum Waianae lee-side coast: afternoon maximum due to the due to the development of sea breeze/onshore flow along the lee-development of sea breeze/onshore flow along the lee-side coast (Leopold, 1948; Loos 2004)side coast (Leopold, 1948; Loos 2004)

77

Previous Modeling Studies of Island effectsPrevious Modeling Studies of Island effects

Theoretical studies of airflow and orographic effects in Theoretical studies of airflow and orographic effects in past have focused on flow past an idealized bell-shape past have focused on flow past an idealized bell-shape mountain.mountain.Land surface forcings (thermal forcing and surface Land surface forcings (thermal forcing and surface friction) are either not included or crudely estimated.friction) are either not included or crudely estimated.NE/E Trade-wind flow is persistent during the summer NE/E Trade-wind flow is persistent during the summer over the Hawaiian islands without the presence of large-over the Hawaiian islands without the presence of large-scale disturbances. The Hawaiian Island chain is an scale disturbances. The Hawaiian Island chain is an idealized situation to study island effects for islands with idealized situation to study island effects for islands with different heights and sizes. different heights and sizes. In this study, we will investigate island effects on airflow In this study, we will investigate island effects on airflow and weather for Oahu using numerical models with full and weather for Oahu using numerical models with full model physics for model physics for real casesreal cases..

88

Previous modeling studies for OahuPrevious modeling studies for Oahu Lavoie (1974) performed modeling studies over OahuLavoie (1974) performed modeling studies over Oahu using using

a simple, single-layer mesoscale model:a simple, single-layer mesoscale model: Focus: effect of topography and the trade wind inversionFocus: effect of topography and the trade wind inversion Interesting finding: Hydraulic jump-like feature to the lee of Interesting finding: Hydraulic jump-like feature to the lee of

both mountain ranges both mountain ranges Limitation: one layer model, 3-km resolution, poor input Limitation: one layer model, 3-km resolution, poor input

data source. Poor treatments of land surface forcings and data source. Poor treatments of land surface forcings and diurnal cycle.diurnal cycle.

Recently:Recently: Zhang et al. (2005): use MSM/LSM to simulate Zhang et al. (2005): use MSM/LSM to simulate

temperature at three stations on Oahu. He showed that temperature at three stations on Oahu. He showed that adequate descriptions of the terrain and vegetation are adequate descriptions of the terrain and vegetation are required to simulate diurnal cycle at these sites.required to simulate diurnal cycle at these sites.

Using MM5/LSM (Land Surface Model) with adequate Using MM5/LSM (Land Surface Model) with adequate descriptions of the terrain and vegetation, Yang et al. descriptions of the terrain and vegetation, Yang et al. (2005) were able to successfully simulate the diurnal (2005) were able to successfully simulate the diurnal evolution of airflow and weather over the Big Island.evolution of airflow and weather over the Big Island.

99

Scientific Objectives:Scientific Objectives: MM5/LSM will used as a research tool to study island effects MM5/LSM will used as a research tool to study island effects

on rainfall and airflow for the island of Oahu.on rainfall and airflow for the island of Oahu. Would it be possible to use MM5/LSM to simulate the island Would it be possible to use MM5/LSM to simulate the island

effects for the island of Oahu which is relatively small?effects for the island of Oahu which is relatively small? Study the relative importance of land surface forcing Study the relative importance of land surface forcing

(daytime heating, surface friction) and orographic blocking on (daytime heating, surface friction) and orographic blocking on the production of westerly reversely flow along the lee-side the production of westerly reversely flow along the lee-side coast in the afternoon hours.coast in the afternoon hours.

Orographic effects on winds and trade-wind inversion above Orographic effects on winds and trade-wind inversion above the lee-side slopesthe lee-side slopes

What are the effects of trade-wind conditions, e.g., strong What are the effects of trade-wind conditions, e.g., strong and weak trades, on the island-scale airflow and orographic and weak trades, on the island-scale airflow and orographic rainfall over Oahu?rainfall over Oahu?

1010

Research planResearch plan Tool: The nTool: The nonhydrostatic Mesoscale Model version 5 (MM5) onhydrostatic Mesoscale Model version 5 (MM5)

(Dudhia, 1993) (Dudhia, 1993) coupled with an advanced land surface coupled with an advanced land surface model (F. Chen and Dudhia 2001)model (F. Chen and Dudhia 2001)

With adequate depiction of the terrain and land surface With adequate depiction of the terrain and land surface properties and vegetation cover (Zhang et al., 2005; Yang et properties and vegetation cover (Zhang et al., 2005; Yang et al., 2005) in the model.al., 2005) in the model.

Input for mm5: Global Forecast System (GFS) data Input for mm5: Global Forecast System (GFS) data provided by NCEP (National Centers for Environmental provided by NCEP (National Centers for Environmental Prediction), hPrediction), horizontal resolution orizontal resolution ~ 100 km~ 100 km

Run model for the period of July-August 2005, and compare Run model for the period of July-August 2005, and compare with observationswith observations

Observations: 13 hourly stations and 69 rain gagesObservations: 13 hourly stations and 69 rain gages Simulations from 10 days of weak trades and 10 days of Simulations from 10 days of weak trades and 10 days of

strong trades are averaged separately to investigate the strong trades are averaged separately to investigate the effects of trade-wind conditions on island-scale airflow and effects of trade-wind conditions on island-scale airflow and weather weather

Sensitivity tests: to investigate the effects of terrain and Sensitivity tests: to investigate the effects of terrain and thermal forcing on island-scale airflow and weather thermal forcing on island-scale airflow and weather

1111

Model descriptionsModel descriptions No of domains: 4 nested domainsNo of domains: 4 nested domains Horizontal resolution of 1.5-km (for Oahu domain) Horizontal resolution of 1.5-km (for Oahu domain) Vertical levels: 28 (sigma)Vertical levels: 28 (sigma) Cumulus parameterization scheme: Grell et al. (1994) (No Cumulus parameterization scheme: Grell et al. (1994) (No

cumulus parameterization for the 1.5-km Oahu domain)cumulus parameterization for the 1.5-km Oahu domain) Explicit precipitation: warm rain (Hsie et al., 1984)Explicit precipitation: warm rain (Hsie et al., 1984) Radiation scheme: Dudhia (1989)Radiation scheme: Dudhia (1989) Planetary boundary layer scheme: MRF (Hong and Pan, 1996)Planetary boundary layer scheme: MRF (Hong and Pan, 1996) NCEP Noah LSMNCEP Noah LSM Soil moisture data are generated for continuous model run for Soil moisture data are generated for continuous model run for

two months prior to simulation period (Yang et al. 2005) two months prior to simulation period (Yang et al. 2005) Each day, a 36-hour forecast is made. Each day, a 36-hour forecast is made. 24 hours of simulation: from the 1224 hours of simulation: from the 12thth to the 36 to the 36thth hour of model hour of model

outputoutput

1212

4.5 km

13.5 km

40.5 km

Note: Horizontal resolution of 1.5-km for Oahu domainNote: Horizontal resolution of 1.5-km for Oahu domain

1.5 km

1313

Average simulated island-scale and wind observations for summer 2005. Weak winds over central Oahu and along the lee-side coast Strong wind speed at northwestern and southeastern corners of Oahu

Observations: (red windbars) at 13 stations

1414

Average simulated island-scale and wind observations for summer 2005. Weak winds over central Oahu and along the lee-side coast Strong wind speed at northwestern and southeastern corners of Oahu

Observations: (red wind barbs) at 13 stations

One flag: 5 m/s,

one full barb: 1 m/s

1515

Simulation of temperature diurnal cycle at stationsSimulation of temperature diurnal cycle at stations

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Kaneohe MCB

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Honolulu AP

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Kalaeloa AP

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Kii

1616

Simulation of temperature diurnal cycle at stationsSimulation of temperature diurnal cycle at stations

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Kalaeloa AP

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Kii

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Kahuku Training

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Kawailoa TA

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Makua Valley

15

19

23

27

31

0 3 6 9 12 15 18 21 24

T T_ob Makua Ridge

1717

Mean 2-m temperature anomalies (K) from an upstream point Mean 2-m temperature anomalies (K) from an upstream point (21.67oN, 157.71oW) at the same height during July-August 2005 at (21.67oN, 157.71oW) at the same height during July-August 2005 at

(a) 14 HST(a) 14 HST

1818

Mean 2-m temperature anomalies (K) from an upstream point Mean 2-m temperature anomalies (K) from an upstream point (21.67oN, 157.71oW) at the same height during July-August 2005 at (21.67oN, 157.71oW) at the same height during July-August 2005 at

(b) 05 HST.(b) 05 HST.

1919

Simulation of mean mixing ratioSimulation of mean mixing ratio

Mean mixing ratio simulation shows:Moist area over windward of Ko’olau because of trade flowDry areas over central Oahu because of island blocking The Driest over the Waianae mountains

2020

Mean 2-m mixing ratio anomalies (g kg-1) from an upstream point Mean 2-m mixing ratio anomalies (g kg-1) from an upstream point

(21.67oN, 157.71oW) during July-August, 2005 at (a) 14 HST(21.67oN, 157.71oW) during July-August, 2005 at (a) 14 HST

2121

Mean 2-m mixing ratio anomalies (g kg-1) from an upstream point Mean 2-m mixing ratio anomalies (g kg-1) from an upstream point (21.67oN, 157.71oW) during July-August, 2005 at (b) 05 HST.(21.67oN, 157.71oW) during July-August, 2005 at (b) 05 HST.

2222

Diurnal cycle of windsDiurnal cycle of windsWith daytime heating and nighttime cooling overland, the airflow With daytime heating and nighttime cooling overland, the airflow also changes considerably during the diurnal cyclealso changes considerably during the diurnal cycleThe simulated island airflow throughout the day agrees well with The simulated island airflow throughout the day agrees well with observations at thirteen stationsobservations at thirteen stations

2323


2424

Accumulate rainfall show maximum at windward side; minimum Accumulate rainfall show maximum at windward side; minimum at central Oahu and the leesideat central Oahu and the leeside

A weak maximum on the leeside of the Waianae mountainsA weak maximum on the leeside of the Waianae mountains

: 62 days

2525

Total rainfall Total rainfall accumulation (mm) accumulation (mm) during July-August during July-August 2005.2005. (a) observation, (a) observation,

(b) simulation.(b) simulation.

2626

Three-hour Three-hour rainfall rainfall

accumulation accumulation during July-during July-

August, 2005: August, 2005: observed (left) observed (left) and simulated and simulated (right) rainfall (right) rainfall

(mm).(mm).

2727

Time series Time series plots of 3-h plots of 3-h simulated simulated meteorological meteorological variables at variables at 21.521.5ooN, N, 157.85157.85ooW for W for (a) equivalent (a) equivalent potential potential temperature temperature (K), (K),

(b) deviations of (b) deviations of vertical motions vertical motions from the daily from the daily mean (cm s-1).mean (cm s-1).

2828

(c) zonal wind (c) zonal wind (m s-1), (m s-1),

(d) LCL and LFC(d) LCL and LFC in km. in km.

2929

Diurnal cycle of winds and vertical motionsDiurnal cycle of winds and vertical motions

cross section at 21.45 Ncross section at 21.45 N

3030

Vertical cross Vertical cross section of section of wind vector wind vector (m s-1) and (m s-1) and vertical vertical velocity velocity (shaded, cm (shaded, cm s-1) along s-1) along 21.45o N for 21.45o N for (a) mean (a) mean state during state during July-August July-August 2005, 2005, (b) anomalies (b) anomalies from the from the mean at 08 mean at 08 HST.HST.

3131

Anomalies Anomalies from the mean from the mean

(c) 14 HST, (c) 14 HST,

(d) 20 HST(d) 20 HST. .

3232

Vertical displacements of the fluid are limited by U/NVertical displacements of the fluid are limited by U/N

3333

3434

Theta (K) and wind vectorsTheta (K) and wind vectors

3535

Theta-E (K) and wind vectorsTheta-E (K) and wind vectors

3636

Strong and weak trade wind categoriesStrong and weak trade wind categories+ Upstream point of Oahu (21N, 152W) in reanalysis data from National + Upstream point of Oahu (21N, 152W) in reanalysis data from National Center for Environment Prediction/National Center for Atmospheric Center for Environment Prediction/National Center for Atmospheric Research (NCEP/NCAR) is used to define the strong and weak tradesResearch (NCEP/NCAR) is used to define the strong and weak trades

+ Strong trades: wind speed > 8m/s+ Strong trades: wind speed > 8m/s wind direction: 70 - 90 degreewind direction: 70 - 90 degree+ Weak trades: wind speed < 7m/s+ Weak trades: wind speed < 7m/s wind direction: 60 - 90 degreewind direction: 60 - 90 degree

3737

Effects of trade wind conditions on mean circulationEffects of trade wind conditions on mean circulation

Strong trades

Weaktrades

mean

One flag: 5 m/s, Full barb: 1m/s Half barb:0.5m/s

Mean simulation Mean simulation of 10 strong, 10 of 10 strong, 10 weak trade days weak trade days show that mean show that mean wind speeds wind speeds entire Oahu are entire Oahu are weaker in weak weaker in weak tradestrades

3838

Effect of different trade wind conditions on total rainfallEffect of different trade wind conditions on total rainfall More rainfall is simulated when trades are strongerMore rainfall is simulated when trades are stronger

Ten days of strong trade winds

Ten days of weak trade winds

3939

At night, the At night, the cool air cool air moves down moves down from the from the windward side windward side of the Ko’olau of the Ko’olau mountains mountains under weak under weak trade-wind trade-wind conditionsconditions

Strong trades

Weaktrades

05HST


Effects of trade wind conditions on diurnal cycle of windEffects of trade wind conditions on diurnal cycle of wind

4040

Strong trades

Weaktrades

In the In the afternoon, the afternoon, the westerly westerly onshore flow onshore flow along the along the Waianae coast Waianae coast and adjacent and adjacent waters is more waters is more pronounced pronounced under weak under weak trades. What trades. What is the relatively is the relatively importance of importance of orographic orographic blocking vs. blocking vs. land surface land surface forcing?forcing?

14HST



4141

Sensitivity testsSensitivity tests Sensitivity tests are to evaluate the effect of terrain and land surface forcing on the production of lee-side wakes

Epifanio and Rotunno (2005): Blocking of the flow leads to the formation of warm anomalies over the lee slope as potentially warm air descends from aloft to replace the colder air deflected laterally around the barrier. The production of lee-side wake is caused by the temperature gradient between

the warm lee-slope air and the colder fluid downstream.

Sensitivity tests: No mountainsNo mountains and no surface frictionNo mountains and 75% of solar heatingNo Mountains and 50% of solar heating

The sensitivity tests are run for one normal trade wind day, 09 Aug, 2005

4242

No mountain case: WindsNo mountain case: Winds

Wind directions are uniform but with weakerspeeds over land than open ocean upstream

Onshore flows off the Waianae coast still exist as aresult of island heating but are weaker than CTRL

05 HST

Control

No mountain

14 HST

4343

No Mountain , No frictions (NMNF)No Mountain , No frictions (NMNF)For the NMNF case, the westerly flow at 14 HST is weaker than the For the NMNF case, the westerly flow at 14 HST is weaker than the NM case. The Westerly flow offshore is mainly caused by solar NM case. The Westerly flow offshore is mainly caused by solar heating.heating.

4444

No Mountains and 50% solar heating (NM50%), the No Mountains and 50% solar heating (NM50%), the afternoon westerly flow off the lee-side coast disappears.afternoon westerly flow off the lee-side coast disappears.

4545

SummarySummary Even with a relatively small size, the island of Oahu Even with a relatively small size, the island of Oahu

has profound influences on island airflow and weatherhas profound influences on island airflow and weather With a 1.5-km resolution, the island effects on airflow With a 1.5-km resolution, the island effects on airflow

and weather are reasonably well simulated by the and weather are reasonably well simulated by the MM5/LSM model, includingMM5/LSM model, including

windward and mauka showers at night and early morning windward and mauka showers at night and early morning over the Ko’olau mountains;over the Ko’olau mountains;

strong easterly downslope winds aloft on the lee-side slopes strong easterly downslope winds aloft on the lee-side slopes of both Ko’olau and Waianae mountains at night; of both Ko’olau and Waianae mountains at night;

weak wind over central Oahu;weak wind over central Oahu; westerly onshore flow in the afternoon offshore off the westerly onshore flow in the afternoon offshore off the

Waianae coast, mainly a result of land surface heating.Waianae coast, mainly a result of land surface heating.

4646

SummarySummary The simulated spatial and diurnal variations of surface The simulated spatial and diurnal variations of surface

winds and rainfall are in good agreement with observations.winds and rainfall are in good agreement with observations. The rainfall is mainly caused by orographic lifting. More The rainfall is mainly caused by orographic lifting. More

rainfall is simulated when trades are stronger.rainfall is simulated when trades are stronger. There are two simulated rainfall maxima at the foothills and There are two simulated rainfall maxima at the foothills and

over the Ko’olau Mountains during the diurnal cycle. over the Ko’olau Mountains during the diurnal cycle. The early morning rainfall maximum is caused by The early morning rainfall maximum is caused by

anomalies in rising motions on the windward side as the anomalies in rising motions on the windward side as the results of flow deceleration over the windward coastal results of flow deceleration over the windward coastal regions and foothills when the land surface is the coolest.regions and foothills when the land surface is the coolest.

The early evening rainfall maximum is caused by stronger The early evening rainfall maximum is caused by stronger winds aloft after sunset. winds aloft after sunset.

The rainfall minimum in the early afternoon is related to The rainfall minimum in the early afternoon is related to relatively weak orographic lifting due to relatively weak relatively weak orographic lifting due to relatively weak winds aloft and a relatively high LFC as a result of vertical winds aloft and a relatively high LFC as a result of vertical mixing.mixing.

4747

Future workFuture work

Use WRF (Weather Research and Forecast) Use WRF (Weather Research and Forecast) model, developed at the National Center for model, developed at the National Center for Atmospheric Research, to conduct high-resolution Atmospheric Research, to conduct high-resolution simulations of heavy rainfall events over the simulations of heavy rainfall events over the Hawaiian Islands . Especially the recent record Hawaiian Islands . Especially the recent record breaking prolonged wet period during February 19 breaking prolonged wet period during February 19 – April 2, 2006.– April 2, 2006.

Data assimilation to improve the initial conditions.Data assimilation to improve the initial conditions.

4848

4949

Over the Hawaiian Islands, large variations in local microclimate ranging from humid tropical on the windward slopes to hot desert over bare lava soil with different vegetation covers are typical (Juvik et al. 1978; Giambelluca et al. 1993).

5050

(a) (b)

(c)High-resolution (30“, ~ 1 km at 20 ºN) (a) vegetation types, (b) soil types, and (c) vegetation fraction for the island of Oahu compiled from the USGS 1:100,000-scale Land Use Land Cover Datasets for Hawaii. 13 vegetation types and 16 soil types are used. Vegetation fraction ranges from 0 to 100%.

5151

Surface roughness length in (a) the current version of MSM and (b) coupled MSM/LSM. Unit is m in (a) and cm in (b). Contour interval is 0.0005 m in (a) and 5 cm in (b).

(a) (b)

5252

Surface albedo (%) in (a) the current version of MSM and (b) coupled MSM/LSM. Contour interval is 1.0 in (b). In (a), the surface albedo was interpolated from the global data and was constant (= 4.002). In (b), the surface albedo was computed based on surface vegetation types and reflects the heterogeneous surface properties over the island.

(a) (b)

5353


5454

Effects of trade wind conditions on mean circulationEffects of trade wind conditions on mean circulation

strong trades

Weaktrades

05HSTMean simulation Mean simulation of 10 strong, 10 of 10 strong, 10 weak trade days weak trade days show that mean show that mean wind speeds wind speeds entire Oahu are entire Oahu are weaker in weak weaker in weak tradestrades

5555

At night, the At night, the cool air cool air moves down moves down from the from the windward side windward side of the Ko’olau of the Ko’olau mountains mountains under weak under weak trade-wind trade-wind conditionsconditions

Strong trades

Weaktrades

05HST


5656

Strong trades

Weaktrades

In the In the afternoon, the afternoon, the westerly westerly onshore flow onshore flow along the along the Waianae coast Waianae coast and adjacent and adjacent waters due to waters due to land surface land surface heating is more heating is more pronounced pronounced under weak under weak tradestrades

14HST


5757

Simulation of mean dew point temperatureSimulation of mean dew point temperature

Diurnal cycle of dew point temperature errors for 13 hourly Oahu stations

02HST 05HST 08HST 11HST 14HST 17HST 20HST 23HST

BIAS 0.7 0.6 -0.7 -1.1 -1.0 -0.5 0.5 0.5

MSE 1.6 1.6 1.7 2.0 2.1 1.9 1.6 1.9

MAE 1.4 1.4 1.5 1.8 1.8 1.6 1.4 1.4

5858

Simulation of mean temperatureSimulation of mean temperature

Diurnal cycle of temperature errors for 13 hourly Oahu stations


BIAS 0.4 0.4 0.0 -0.4 -0.3 -1.1 -0.3 -0.1

RMSE 1.4 1.5 1.5 2.1 2.2 1.9 1.2 1.2

MAE 1.1 1.2 1.2 1.8 1.8 1.6 0.9 1.0

Bias =mean( model –observation); MAE = mean( abs( model –observation)

5959

Wind dirWind dir 02HST 05HST 08HST 11HST 14HST 17HST 20HST 23HST

BIAS -0.4 3.5 -4.6 -12.8 -18.2 -20.6 0.2 6.0

MAE 60.0 59.4 47.1 39.2 41.7 39.3 49.0 53.2

Wind speedWind speed 02HST 05HST 08HST 11HST 14HST 17HST 20HST 23HST

BIAS 0.4 0.3 0.7 -0.4 -0.6 0.2 0.5 0.3

MAE 1.9 1.9 2.0 1.7 1.8 1.8 2.0 2.0

One statistic index set for 13 Oahu hourly stations

Bias =mean( model –observation); MAE = mean( abs( model –observation)

Dew pointDew point 02HST 05HST 08HST 11HST 14HST 17HST 20HST 23HST

BIAS 0.7 0.6 -0.7 -1.1 -1.0 -0.5 0.5 0.5

MSE 1.6 1.6 1.7 2.0 2.1 1.9 1.6 1.9

MAE 1.4 1.4 1.5 1.8 1.8 1.6 1.4 1.4

Air tempAir temp 02HST 05HST 08HST 11HST 14HST 17HST 20HST 23HST

BIAS 0.4 0.4 0.0 -0.4 -0.3 -1.1 -0.3 -0.1

RMSE 1.4 1.5 1.5 2.1 2.2 1.9 1.2 1.2

MAE 1.1 1.2 1.2 1.8 1.8 1.6 0.9 1.0

6060

Simulation of mean temperatureSimulation of mean temperatureDiurnal cycle of temperature errors for 13 hourly Oahu stations


BIAS 0.4 0.4 0.0 -0.4 -0.3 -1.1 -0.3 -0.1 RMSE 1.4 1.5 1.5 2.1 2.2 1.9 1.2 1.2 MAE 1.1 1.2 1.2 1.8 1.8 1.6 0.9 1.0

Table 8.b. Diurnal cycle of dew point temperature at 13 hourly Oahu stations

6161

Simulation of temperature diurnal cycleSimulation of temperature diurnal cycle

+ At night,+ At night,temperature temperature decrease because decrease because of radiative cooling of radiative cooling and the isothermal and the isothermal contours are smoothcontours are smooth

+ With day time + With day time heating, the spatial heating, the spatial distribution of distribution of surface air surface air temperature is temperature is chaotic; much hotter chaotic; much hotter over Waianae than over Waianae than Ko’olauKo’olau

05 HST

14 HST

6262

Simulation of mean dew point temperatureSimulation of mean dew point temperatureDiurnal cycle of dew point temperature errors for 13 hourly Oahu stations


BIAS 0.7 0.6 -0.7 -1.1 -1.0 -0.5 0.5 0.5 MSE 1.6 1.6 1.7 2.0 2.1 1.9 1.6 1.9 MAE 1.4 1.4 1.5 1.8 1.8 1.6 1.4 1.4

6363

Temperature at 05 HST: No mountain case Temperature at 05 HST: No mountain case

No mountain makes temperature at 05 HST almost uniform with No mountain makes temperature at 05 HST almost uniform with open ocean value entire Oahuopen ocean value entire Oahu

6464

Temperature at 14 HST: No mountain case Temperature at 14 HST: No mountain case

No mountain makes temperature entire Oahu higher than open No mountain makes temperature entire Oahu higher than open ocean values 1-2 degree. Higher temperature over Waianaeocean values 1-2 degree. Higher temperature over Waianae

Numerical Simulations of Airflow and Weather over the Island of Oahu Yi-Leng Chen and Hiep Van Nguyen.

Documents

introduction island

island terrain of oahu

introduction effects

oahu rainfall

weather slide

islandscale airflow

orographic effects

islandscale flow