Search Results
You are looking at 1 - 10 of 21 items for
- Author or Editor: Noel E. Davidson x
- Refine by Access: All Content x
Abstract
Six-hourly analyses based on the special observational dataset from the Australian Monsoon Experiment are used to derive vorticity budget diagnostics for a number of tropical weather situations. The quality of the analyses is demonstrated by observation fitting statistics, comparison of digital satellite cloud imagery with diagnosed vertical motion, and by comparison of derived quantities with those obtained directly from the observations using line integral calculations.
For the mean of the entire experimental period, balance generally exists between stretching and horizontal advection, with some contribution from an apparent vorticity source at upper levels. The individual-day behavior, however, is often quite different. Three categories are evident: 1) For weak, low-level cyclonic flows at the time of maximum convection (disorganized, deep convection), an apparent source of cyclonic vorticity is evident at low and high levels, and a sink is indicated through midlevels. These situations are characterized by a midlevel convergence maximum and a related cyclonic vorticity maximum. 2) For strong, low-level vorticity regimes (circulation systems with organized convection), an apparent sink of vorticity is evident everywhere below the convective outflow level, with a source above. These situations are characterized by deep convergence and a low level of maximum vorticity. 3) The third category, which seems to be associated with mostly stratiform regimes, is similar to 1), but a source is diagnosed at mid- to high levels (possibly associated with the stratiform cloud deck) with a sink farther aloft. The author postulates that die vertical structure of apparent vorticity sources is determined by the ascent and descent motions of the dominant cloud forms, which redistribute the background vorticity, and by enhanced boundary layer convergence as circulation systems spin up.
The implications of the (a) midlevel convergence maximum and (b) apparent vorticity sources to the understanding and prediction of monsoon onset, midtropospheric cyclones, and tropical cyclone behavior are discussed. In a companion paper, a representation of the apparent heat and vorticity sources is implemented in a numerical model and used in simulations of the above weather phenomenon.
Abstract
Six-hourly analyses based on the special observational dataset from the Australian Monsoon Experiment are used to derive vorticity budget diagnostics for a number of tropical weather situations. The quality of the analyses is demonstrated by observation fitting statistics, comparison of digital satellite cloud imagery with diagnosed vertical motion, and by comparison of derived quantities with those obtained directly from the observations using line integral calculations.
For the mean of the entire experimental period, balance generally exists between stretching and horizontal advection, with some contribution from an apparent vorticity source at upper levels. The individual-day behavior, however, is often quite different. Three categories are evident: 1) For weak, low-level cyclonic flows at the time of maximum convection (disorganized, deep convection), an apparent source of cyclonic vorticity is evident at low and high levels, and a sink is indicated through midlevels. These situations are characterized by a midlevel convergence maximum and a related cyclonic vorticity maximum. 2) For strong, low-level vorticity regimes (circulation systems with organized convection), an apparent sink of vorticity is evident everywhere below the convective outflow level, with a source above. These situations are characterized by deep convergence and a low level of maximum vorticity. 3) The third category, which seems to be associated with mostly stratiform regimes, is similar to 1), but a source is diagnosed at mid- to high levels (possibly associated with the stratiform cloud deck) with a sink farther aloft. The author postulates that die vertical structure of apparent vorticity sources is determined by the ascent and descent motions of the dominant cloud forms, which redistribute the background vorticity, and by enhanced boundary layer convergence as circulation systems spin up.
The implications of the (a) midlevel convergence maximum and (b) apparent vorticity sources to the understanding and prediction of monsoon onset, midtropospheric cyclones, and tropical cyclone behavior are discussed. In a companion paper, a representation of the apparent heat and vorticity sources is implemented in a numerical model and used in simulations of the above weather phenomenon.
Abstract
Diagnostics from the observational dataset of the Australian Monsoon Experiment (AMEX) have revealed two interesting characteristics of convective systems over the Australian Tropics (Part I of this study). The first is a midlevel convergence maximum in situations of disorganized convection, which implies weak low-level and strong upper-level convective heating. The second is the presence of large apparent vorticity sources during deep convective and stratiform events. It is suggested that these characteristics are related to the ascent and descent motions in tropical cloud systems and to the way in which they redistribute mass and the background vorticity.
To investigate the importance of these features of convection on tropical prediction, a representation of the observed thermodynamic and kinematic effects of clouds has been implemented in the Bureau of Meteorology Research Centre's tropical limited-area model. To represent the observed structure of convective heating in numerical simulation experiments, an analytic heating function is constructed that is similar to the observed heating profile. Satellite cloud imagery is then used to trigger the function in space and time during the model integration. In this way the importance of the convective processes can be assessed, without the uncertainties of incorrect triggering of the parameterization. To represent the observed structure of convective heating in numerical prediction experiments, the Kuo heating profile is modified to reflect the (AMEX) observed profile, with weak low-level heating. To parameterize the kinematic effects, a potential function is introduced. Vorticity tendencies due to clouds are represented by the Laplacian of the function, and the momentum tendencies by the spatial derivatives. Vorticity tendencies are specified according to the observed structure of apparent sources and are dependent on the level of maximum convergence and the ambient vorticity and divergence profiles.
The representation of convection, including both thermodynamic and kinematic effects, has been run in a number of simulations of AMEX weather events. It is shown that, in a limited number of events, realistic forecasts of monsoon onset, midtropospheric lows, and tropical cyclogenesis can be obtained.
For tropical cyclogenesis, it is shown that upper-level convective heating produces the required midlevel convergence and eventually a midtropospheric circulation. The parameterized apparent vorticity sources then act to spin down the midlevels and spin up the upper and lower levels. This, together with the resultant enhanced boundary layer convergence, is sufficient to intensify the low-level circulation without the need for large low-level convective heating. Indeed, simulations with convective heating, and no representation of apparent vorticity sources, cannot accurately reproduce the observed four-dimensional structure of the developing tropical cyclone.
Abstract
Diagnostics from the observational dataset of the Australian Monsoon Experiment (AMEX) have revealed two interesting characteristics of convective systems over the Australian Tropics (Part I of this study). The first is a midlevel convergence maximum in situations of disorganized convection, which implies weak low-level and strong upper-level convective heating. The second is the presence of large apparent vorticity sources during deep convective and stratiform events. It is suggested that these characteristics are related to the ascent and descent motions in tropical cloud systems and to the way in which they redistribute mass and the background vorticity.
To investigate the importance of these features of convection on tropical prediction, a representation of the observed thermodynamic and kinematic effects of clouds has been implemented in the Bureau of Meteorology Research Centre's tropical limited-area model. To represent the observed structure of convective heating in numerical simulation experiments, an analytic heating function is constructed that is similar to the observed heating profile. Satellite cloud imagery is then used to trigger the function in space and time during the model integration. In this way the importance of the convective processes can be assessed, without the uncertainties of incorrect triggering of the parameterization. To represent the observed structure of convective heating in numerical prediction experiments, the Kuo heating profile is modified to reflect the (AMEX) observed profile, with weak low-level heating. To parameterize the kinematic effects, a potential function is introduced. Vorticity tendencies due to clouds are represented by the Laplacian of the function, and the momentum tendencies by the spatial derivatives. Vorticity tendencies are specified according to the observed structure of apparent sources and are dependent on the level of maximum convergence and the ambient vorticity and divergence profiles.
The representation of convection, including both thermodynamic and kinematic effects, has been run in a number of simulations of AMEX weather events. It is shown that, in a limited number of events, realistic forecasts of monsoon onset, midtropospheric lows, and tropical cyclogenesis can be obtained.
For tropical cyclogenesis, it is shown that upper-level convective heating produces the required midlevel convergence and eventually a midtropospheric circulation. The parameterized apparent vorticity sources then act to spin down the midlevels and spin up the upper and lower levels. This, together with the resultant enhanced boundary layer convergence, is sufficient to intensify the low-level circulation without the need for large low-level convective heating. Indeed, simulations with convective heating, and no representation of apparent vorticity sources, cannot accurately reproduce the observed four-dimensional structure of the developing tropical cyclone.
Abstract
Characteristics of 500 tropical cyclones (TCs) in the Australian region and its three individual basins are examined based on 40 yr of satellite-supported observations. While tropical cyclones exhibit highly individual behaviors resulting in significant standard deviations, there are some systematic behaviors, which are documented. Most TCs in the Australian region originate from December to April. About 13 are observed each season, with half occurring in the western basin. Generally, the lifetime of a TC is about 7½ days, during which time it covers over 2500 km at a mean speed of 15 km h−1. Around half of the storms reach a maximum intensity corresponding to category 3 or higher (<970 hPa), as classified using a modified Saffir–Simpson scale. Tropical cyclones in the western and eastern basins have around 25% chance of making landfall, while those in the northern basin have an 80% chance.
There appear to be preferred locations for TC genesis, close to the Australian coastline at around 120°, 135°, and 150°E. Genesis occurs near the mean position of the maximum low-level cyclonic vorticity and coincides with the monsoon trough from December to February, but occurs poleward of the trough in other months. The maximum intensity eventually achieved by TCs varies with genesis locations. For storms that reach category 3 or above, there are more corresponding origin points in the west than in both the gulf and the eastern basin.
Recurvature generally follows attainment of maximum intensity, suggesting the importance of trough interactions on this behavior. The likelihood of extratropical transition of TCs, in the mean, increases to a peak in March, although there is variability across the three basins of the Australian region. Final dissipation has no preferred latitude, with many storms dissipating over the warm tropical oceans equatorward of 20°S.
Abstract
Characteristics of 500 tropical cyclones (TCs) in the Australian region and its three individual basins are examined based on 40 yr of satellite-supported observations. While tropical cyclones exhibit highly individual behaviors resulting in significant standard deviations, there are some systematic behaviors, which are documented. Most TCs in the Australian region originate from December to April. About 13 are observed each season, with half occurring in the western basin. Generally, the lifetime of a TC is about 7½ days, during which time it covers over 2500 km at a mean speed of 15 km h−1. Around half of the storms reach a maximum intensity corresponding to category 3 or higher (<970 hPa), as classified using a modified Saffir–Simpson scale. Tropical cyclones in the western and eastern basins have around 25% chance of making landfall, while those in the northern basin have an 80% chance.
There appear to be preferred locations for TC genesis, close to the Australian coastline at around 120°, 135°, and 150°E. Genesis occurs near the mean position of the maximum low-level cyclonic vorticity and coincides with the monsoon trough from December to February, but occurs poleward of the trough in other months. The maximum intensity eventually achieved by TCs varies with genesis locations. For storms that reach category 3 or above, there are more corresponding origin points in the west than in both the gulf and the eastern basin.
Recurvature generally follows attainment of maximum intensity, suggesting the importance of trough interactions on this behavior. The likelihood of extratropical transition of TCs, in the mean, increases to a peak in March, although there is variability across the three basins of the Australian region. Final dissipation has no preferred latitude, with many storms dissipating over the warm tropical oceans equatorward of 20°S.
Abstract
A new Tropical Cyclone Limited Area Prediction System has been developed at the Australian Bureau of Meteorology Research Centre. The features of the new prediction system can be summarized as follows: first, a 12-h, coarse-resolution data assimilation is used to define the outer structure and environment of the storm and to provide initial conditions for coarse mesh prediction. Then, synthetic data are generated to define a storm’s circulation, consistent with observed location, size, intensity, and past motion. The method involves a definition of the environmental flow by filtering of the misplaced tropical cyclone circulation in the (old) objective analysis, the generation of a new, correctly located and intense symmetric vortex, and the construction of vortex asymmetries by requiring that the observed motion be the vector sum of the environmental flow and the asymmetric flow. The subsequent initialization for fine-mesh prediction is carried out using 24 h of diabatic, dynamical nudging through 6-hourly, high-resolution objective analyses, which include the synthetic vortex. During this phase the vorticity and surface pressure fields are largely preserved, while infrared satellite cloud imagery is used to reconstruct the vertical motion field. Finally, numerical prediction is carried out with a high-resolution version of the operational limited-area model of the Australian Bureau of Meteorology, which includes high-order numerics and advanced physical parameterizations.
The very encouraging quality of the forecasts is demonstrated in numerous case studies of tropical cyclone events, including improved prediction for some situations when the official forecasts were poor. Average track errors at 24 and 48 h are 115 and 259 km, respectively. These are significantly smaller than corresponding errors in the official and in climatology-persistence forecasts. Given the uncertainties in estimated central pressures, forecasts of intensity are also encouraging. General discussion focuses on system characteristics and some remaining, unresolved issues.
Abstract
A new Tropical Cyclone Limited Area Prediction System has been developed at the Australian Bureau of Meteorology Research Centre. The features of the new prediction system can be summarized as follows: first, a 12-h, coarse-resolution data assimilation is used to define the outer structure and environment of the storm and to provide initial conditions for coarse mesh prediction. Then, synthetic data are generated to define a storm’s circulation, consistent with observed location, size, intensity, and past motion. The method involves a definition of the environmental flow by filtering of the misplaced tropical cyclone circulation in the (old) objective analysis, the generation of a new, correctly located and intense symmetric vortex, and the construction of vortex asymmetries by requiring that the observed motion be the vector sum of the environmental flow and the asymmetric flow. The subsequent initialization for fine-mesh prediction is carried out using 24 h of diabatic, dynamical nudging through 6-hourly, high-resolution objective analyses, which include the synthetic vortex. During this phase the vorticity and surface pressure fields are largely preserved, while infrared satellite cloud imagery is used to reconstruct the vertical motion field. Finally, numerical prediction is carried out with a high-resolution version of the operational limited-area model of the Australian Bureau of Meteorology, which includes high-order numerics and advanced physical parameterizations.
The very encouraging quality of the forecasts is demonstrated in numerous case studies of tropical cyclone events, including improved prediction for some situations when the official forecasts were poor. Average track errors at 24 and 48 h are 115 and 259 km, respectively. These are significantly smaller than corresponding errors in the official and in climatology-persistence forecasts. Given the uncertainties in estimated central pressures, forecasts of intensity are also encouraging. General discussion focuses on system characteristics and some remaining, unresolved issues.
Abstract
Geostationary and polar-orbiting satellites can provide useful proxy sources of moisture data and diabatic heating. It is shown that the use of this information during data assimilation leads to improved precipitation in the tropics and has the potential to minimize spinup in the model. Furthermore, the use of moisture initialization leads to improved agreement between the model and observed precipitation during the early stages of model integration.
Abstract
Geostationary and polar-orbiting satellites can provide useful proxy sources of moisture data and diabatic heating. It is shown that the use of this information during data assimilation leads to improved precipitation in the tropics and has the potential to minimize spinup in the model. Furthermore, the use of moisture initialization leads to improved agreement between the model and observed precipitation during the early stages of model integration.
Abstract
Some notable problems in tropical prediction have been (i) the sensitivity to, and inaccuracies in, the four-dimensional structure of parameterized convective heating, (ii) the inability of conventional data networks to adequately define tropical cyclone structures, and (iii) the so-called spinup problem of numerical models. To help overcome some of these deficiencies, a diabatic nudging scheme has been developed for the Bureau of Meteorology Research Centre (BMRC) limited-area tropical prediction system.
A target analysis for the nudging is first obtained from statistical interpolation of all observational data, using, as first-guess field, output from a global assimilation and prediction system. Tropical cyclones are optionally inserted via bogus wind observations. From 12 or 24 h prior to the base time of the forecast, the prediction model is nudged toward the target analysis. During nudging the “observationally reliable” rotational wind component is preserved and the heating from the Kuo scheme is replaced by a heating function determined from 6-h satellite-observed cloud-top temperatures. The system introduces realistic tropical cyclone structures into the initial condition, defines a vertical-motion field consistent with the satellite cloud imagery, enhances rainfall rates during the early hours of the forecast, reduces the occurrence of spurious rainfall maxima, and improves mass-wind balance and retention of cyclone circulations during the model integration.
Examples of system performance from enhanced observational datasets and from real-time forecasting are presented. Encouraging results for short-term prediction of both tropical cyclone behavior and rainfall events are documented.
Abstract
Some notable problems in tropical prediction have been (i) the sensitivity to, and inaccuracies in, the four-dimensional structure of parameterized convective heating, (ii) the inability of conventional data networks to adequately define tropical cyclone structures, and (iii) the so-called spinup problem of numerical models. To help overcome some of these deficiencies, a diabatic nudging scheme has been developed for the Bureau of Meteorology Research Centre (BMRC) limited-area tropical prediction system.
A target analysis for the nudging is first obtained from statistical interpolation of all observational data, using, as first-guess field, output from a global assimilation and prediction system. Tropical cyclones are optionally inserted via bogus wind observations. From 12 or 24 h prior to the base time of the forecast, the prediction model is nudged toward the target analysis. During nudging the “observationally reliable” rotational wind component is preserved and the heating from the Kuo scheme is replaced by a heating function determined from 6-h satellite-observed cloud-top temperatures. The system introduces realistic tropical cyclone structures into the initial condition, defines a vertical-motion field consistent with the satellite cloud imagery, enhances rainfall rates during the early hours of the forecast, reduces the occurrence of spurious rainfall maxima, and improves mass-wind balance and retention of cyclone circulations during the model integration.
Examples of system performance from enhanced observational datasets and from real-time forecasting are presented. Encouraging results for short-term prediction of both tropical cyclone behavior and rainfall events are documented.
Abstract
Evidence is presented of a downstream development mechanism operating across the entire longitudinal span of the 1978/79 Southern Hemisphere monsoon. Observationally it is seen as progressive cyclonic and anticyclonic vorticity increases that develop eastward in the monsoon trough at a speed of approximately 5 m s−1. The process results in many tropical cyclone and tropical depression formations over northern Australia and the South Pacific.
It is shown that the downstream development process is generally consistent with linearized barotropic dynamics, and that the Southern Hemisphere monsoon, because of an intrinsic westerly basic state, is a particularly suitable region for downstream events.
It is also shown that some apparent contradictions in previous observational studies can be rationalized by the theory. The interactions between the regional components of the monsoon (Indonesian, Australian and South Pacific sectors) can also he better understood. We further suggest that the process has implications for other features of the monsoon circulation, namely onset and 40–50 day events.
Abstract
Evidence is presented of a downstream development mechanism operating across the entire longitudinal span of the 1978/79 Southern Hemisphere monsoon. Observationally it is seen as progressive cyclonic and anticyclonic vorticity increases that develop eastward in the monsoon trough at a speed of approximately 5 m s−1. The process results in many tropical cyclone and tropical depression formations over northern Australia and the South Pacific.
It is shown that the downstream development process is generally consistent with linearized barotropic dynamics, and that the Southern Hemisphere monsoon, because of an intrinsic westerly basic state, is a particularly suitable region for downstream events.
It is also shown that some apparent contradictions in previous observational studies can be rationalized by the theory. The interactions between the regional components of the monsoon (Indonesian, Australian and South Pacific sectors) can also he better understood. We further suggest that the process has implications for other features of the monsoon circulation, namely onset and 40–50 day events.
Abstract
High resolution observational data from the Australian Monsoon Experiment have been used to verify simulations of the development of tropical cyclone Irma.
From a small amplitude, prehurricane cloud cluster, the FSU high resolution regional prediction model quite skillfully simulates the temporal and spatial structure changes during development. The results are, however, sensitive to the initial windfield and somewhat sensitive to both the initial moisture field and the imposed boundary conditions.
Temporal changes in the symmetric and asymmetric structure of the observed and simulated disturbances are compared. Apart from other well-documented necessary conditions, changes in the storm's vertical structure, and the development of horizontal asymmetries appear to be crucial features of the simulated development. The associated physical processes are discussed.
Deficiencies in the simulation are lack of diurnal modulation of the vertical motion field, larger than diagnosed values of vertical motion over the genesis area, and an upper level flow that is too divergent and anticyclonic. We speculate that more accurate representation of diurnal effects, and a parameterization of momentum transports by cumulus clouds will reduce these deficiencies.
Abstract
High resolution observational data from the Australian Monsoon Experiment have been used to verify simulations of the development of tropical cyclone Irma.
From a small amplitude, prehurricane cloud cluster, the FSU high resolution regional prediction model quite skillfully simulates the temporal and spatial structure changes during development. The results are, however, sensitive to the initial windfield and somewhat sensitive to both the initial moisture field and the imposed boundary conditions.
Temporal changes in the symmetric and asymmetric structure of the observed and simulated disturbances are compared. Apart from other well-documented necessary conditions, changes in the storm's vertical structure, and the development of horizontal asymmetries appear to be crucial features of the simulated development. The associated physical processes are discussed.
Deficiencies in the simulation are lack of diurnal modulation of the vertical motion field, larger than diagnosed values of vertical motion over the genesis area, and an upper level flow that is too divergent and anticyclonic. We speculate that more accurate representation of diurnal effects, and a parameterization of momentum transports by cumulus clouds will reduce these deficiencies.
Abstract
Multiple secondary eyewall formations (SEFs) and eyewall replacement cycles (ERCs) are simulated with the fifth-generation Pennsylvania State University (PSU)–National Center for Atmospheric Research (NCAR) Mesoscale Model (MM5) at horizontal grid spacing of 0.67 km. The simulated hurricane is initialized from a weak, synthetic vortex in a quiescent environment on an f plane. After spinup and rapid intensification, the hurricane enters a mature phase during which the intensity change is relatively slow. Convective clouds then organize into a ring with a secondary tangential wind maximum at radii beyond the hurricane’s primary eyewall. This secondary eyewall (SE) then contracts and strengthens. The primary eyewall weakens and is eventually replaced by the SE. The hurricane grows in size and the radius of maximum wind (RMW) increases as similar ERCs repeat 5 times during the simulation.
Two existing hypotheses on SEF are evaluated using the simulation output. Then, model diagnostics are used to reveal that crucial linked components of SEF are (i) a broadening of the swirling flow, (ii) the structure of the evolving secondary circulation, and (iii) the structure of the net radial force (NRF) in the boundary layer (with largest contributions from the agradient and frictional forces). During SEF, there exists strong positive NRF in the region of the primary eyewall, a secondary positive maximum over the SEF region, and a minimum between the two. As a response of the boundary layer depth–integrated radial flow to the NRF, a secondary maximum convergence zone (SMCZ) in the boundary layer develops at the SEF radii. Eventually moist convection in the SMCZ becomes active as the SEF develops.
Abstract
Multiple secondary eyewall formations (SEFs) and eyewall replacement cycles (ERCs) are simulated with the fifth-generation Pennsylvania State University (PSU)–National Center for Atmospheric Research (NCAR) Mesoscale Model (MM5) at horizontal grid spacing of 0.67 km. The simulated hurricane is initialized from a weak, synthetic vortex in a quiescent environment on an f plane. After spinup and rapid intensification, the hurricane enters a mature phase during which the intensity change is relatively slow. Convective clouds then organize into a ring with a secondary tangential wind maximum at radii beyond the hurricane’s primary eyewall. This secondary eyewall (SE) then contracts and strengthens. The primary eyewall weakens and is eventually replaced by the SE. The hurricane grows in size and the radius of maximum wind (RMW) increases as similar ERCs repeat 5 times during the simulation.
Two existing hypotheses on SEF are evaluated using the simulation output. Then, model diagnostics are used to reveal that crucial linked components of SEF are (i) a broadening of the swirling flow, (ii) the structure of the evolving secondary circulation, and (iii) the structure of the net radial force (NRF) in the boundary layer (with largest contributions from the agradient and frictional forces). During SEF, there exists strong positive NRF in the region of the primary eyewall, a secondary positive maximum over the SEF region, and a minimum between the two. As a response of the boundary layer depth–integrated radial flow to the NRF, a secondary maximum convergence zone (SMCZ) in the boundary layer develops at the SEF radii. Eventually moist convection in the SMCZ becomes active as the SEF develops.
Abstract
Tropical cyclone (TC) rainfall over the Australian continent is studied using observations from 41 TC seasons 1969/70 to 2009/10. A total of 318 storms, whose centers either crossed the coastline or were located within 500 km of the coast, are considered in this study. Mean seasonal (November/April) contributions by TCs to the total rainfall are largest along the northern coastline from 120°–150°E. However, the percentage contributions by TCs are greatest west of 125°E, with mean coastal values of 20%–40% and inland values of approximately 20%. Farther east, percentages near the coast are only around 10%, and even lower inland. Inland penetration by TC rainfall is generally greatest over western portions of the continent, associated with greater inland penetration of TC tracks. During the peak of the TC season (January–March), TCs contribute around 40% to the rainfall total of coastal regions west of 120°E, while during December, TCs contribute approximately 60%–70% to the total rainfall west of 115°E. Rain from TCs varies sharply between TC seasons, with some longitude bands receiving no TC rain during some seasons. For the 110°–115°E longitude band the TC rain contribution is quite inconsistent, varying interannually from 0%–86%. This has an impact on water supplies, with storage dams falling to low levels during some years, while filling to capacity during TC-related flood events in other years. These large interannual variations and their impacts underline why it is important to understand TC rainfall characteristics over the Australian continent.
Abstract
Tropical cyclone (TC) rainfall over the Australian continent is studied using observations from 41 TC seasons 1969/70 to 2009/10. A total of 318 storms, whose centers either crossed the coastline or were located within 500 km of the coast, are considered in this study. Mean seasonal (November/April) contributions by TCs to the total rainfall are largest along the northern coastline from 120°–150°E. However, the percentage contributions by TCs are greatest west of 125°E, with mean coastal values of 20%–40% and inland values of approximately 20%. Farther east, percentages near the coast are only around 10%, and even lower inland. Inland penetration by TC rainfall is generally greatest over western portions of the continent, associated with greater inland penetration of TC tracks. During the peak of the TC season (January–March), TCs contribute around 40% to the rainfall total of coastal regions west of 120°E, while during December, TCs contribute approximately 60%–70% to the total rainfall west of 115°E. Rain from TCs varies sharply between TC seasons, with some longitude bands receiving no TC rain during some seasons. For the 110°–115°E longitude band the TC rain contribution is quite inconsistent, varying interannually from 0%–86%. This has an impact on water supplies, with storage dams falling to low levels during some years, while filling to capacity during TC-related flood events in other years. These large interannual variations and their impacts underline why it is important to understand TC rainfall characteristics over the Australian continent.
