Evaluation of operational tropical cyclone intensity forecasts over north Indian Ocean issued by India Meteorological Department

India Meteorological Department (IMD) introduced the objective tropical cyclone (TC) intensity forecast valid for next 24 h over the north Indian Ocean (NIO) in 2003 and extended up to 72 h in 2009. In this study, an attempt is made to evaluate the TC intensity forecast issued by IMD during 2005–2011 (7 years) by calculating the absolute error (AE), root mean square error (RMSE) and skill in intensity forecast in terms of maximum sustained surface wind (MSW). The accuracy of TC intensity forecast has been analysed with respect to basin of formation (Bay of Bengal, Arabian Sea and NIO as whole), season of formation (pre-monsoon and post-monsoon seasons), intensity of TCs (cyclonic storm and severe cyclonic storm or higher intensities) and type of track of TCs (climatological/straight moving and recurving/looping type). The study shows that the average AE (RMSE) in intensity forecast is about 11(14), 14(19) and 20(26) knots, respectively, for 24-, 48- and 72-h forecasts over the NIO as a whole during 2009–2011. The skill of intensity forecast is about 44 %(48 %), 60 %(58 %) and 60 %(65 %) for 24-, 48- and 72-h forecasts during 2009–2011 with respect to AE (RMSE). There is no significant improvement in terms of reduction in AE and RMSE of MSW forecast over the NIO like that over the northwest Pacific and northern Atlantic Oceans during 2005–2011. However, the skill in intensity forecast compared to persistence method has significantly improved by about 6 %(10 %) and 9 %(8 %) per year, respectively, for 12- and 24-h forecasts considering the AE (RMSE) during 2005–2011. There is also significant increasing trend in percentage of 24-h intensity forecasts with error of 10 knots or less during 2005–2011.     

A fast intensity simulator for tropical cyclone risk analysis

Robust estimates of tropical cyclone risk can be made using large sets of storm events synthesized from historical data or from physics-based algorithms. While storm tracks can be synthesized very rapidly from statistical algorithms or simple dynamical models (such as the beta-and-advection model), estimation of storm intensity by using full-physics models is generally too expensive to be practical. Although purely statistical intensity algorithms are fast, they may not be general enough to encompass the effects of natural or anthropogenic climate change. Here we present a fast, physically motivated intensity algorithm consisting of two coupled ordinary differential equations predicting the evolution of a wind speed and an inner core moisture variable. The algorithm includes the effects of ocean coupling and environmental wind shear but does not explicitly simulate spatial structure, which must be handled parametrically. We evaluate this algorithm by using it to simulate several historical events and by comparing a risk analysis based on it to an existing method for assessing long-term tropical cyclone risk. For simulations based on the recent climate, the two techniques perform comparably well, though the new technique does better with interannual variability in the Atlantic. Compared to the existing method, the new method produces a smaller increase in global tropical cyclone frequency in response to global warming, but a comparable increase in power dissipation.     

Impact of bogus vortex for track and intensity prediction of tropical cyclone

The initialization scheme designed to improve the representation of a tropical cyclone in the initial condition is tested during Orissa super cyclone (1999) over Bay of Bengal using the fifth-generation Pennsylvania State University — National Center for Atmospheric Research (Penn State — NCAR) Mesoscale Model (MM5). A series of numerical experiments are conducted to generate initial vortices by assimilating the bogus wind information into MM5. Wind speed and location of the tropical cyclone obtained from best track data are used to define maximum wind speed, and centre of the storm respectively, in the initial vortex. The initialization scheme produced an initial vortex that was well adapted to the forecast model and was much more realistic in size and intensity than the storm structure obtained from the NCEP analysis. Using this scheme, the 24-h, 48-h, and 72-h forecast errors for this case was 63, 58, and 46 km, respectively, compared with 120, 335, and 550 km for the non-vortex initialized case starting from the NCEP global analysis. When bogus vortices are introduced into initial conditions, the significant improvements in the storm intensity predictions are also seen. The impact of the vortex size on the structure of the initial vortex is also evaluated. We found that when the radius of maximum wind (RMW) of the specified vortex is smaller than that of which can be resolved by the model, the specified vortex is not well adapted by the model. In contrast, when the vortex is sufficiently large for it to be resolved on horizontal grid, but not so large to be unrealistic, more accurate storm structure is obtained.     

A parametric Markov renewal model for predicting tropical cyclones in Bangladesh

In this paper, we consider a Markov renewal process (MRP) to model tropical cyclones occurred in Bangladesh during 1877–2009. The model takes into account both the occurrence history and some physical constraints to capture the main physical characteristics of the storm surge process. We assume that the sequence of cyclones constitutes a Markov chain, and sojourn times follow a Weibull distribution. The parameters of the Weibull MRP jointly with transition probabilities are estimated using the maximum likelihood method. The model shows a good fit with the real events, and probabilities of occurrence of different types of cyclones are calculated for various lengths of time interval using the model. Stationary probabilities and mean recurrence times are also calculated. A brief comparison with a Poisson model and a marked Poisson model has also been demonstrated.     

Sensitivity of high-resolution tropical cyclone intensity forecasts to surface flux parameterization

Surface flux parameterization schemes used in current dynamic models are primarily based upon measurements at low and moderate wind speeds. Recent studies show that these parameterization schemes may be incorrect at high wind speeds (e.g., tropical cyclone forecasts). Five high-resolution numerical model experiments are designed to assess the sensitivity of tropical cyclone intensity forecasts to changes in the surface flux parameterization. The sensitivity experiments are conducted by running 48 h forecasts of the Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) for six selected tropical cyclones with individual modifications to surface flux calculation that include: (1) limiting the surface stress for wind speeds greater than 33 m s⁻¹, or 64 knots (kt); (2) computing the stress at the top of the model bottom grid layer (MBGL) by averaging results from surface layer similarity and turbulence mixing parameterization for wind speeds greater than 33 m s⁻¹; (3) increasing the roughness lengths for heat and moisture transfer by a factor of ten; (4) setting the roughness lengths for heat and moisture transfer to 1/10 of the momentum roughness length; and (5) cooling the sea surface temperature (SST) by a prescribed rate at high winds. Averaged responses for the six storms to these sensitivity tests show that: (i) the limit on surface stress at high winds significantly increases the cyclone intensity in 48 h forecasts; (ii) the averaged surface layer stress at high winds increases the cyclone intensity but to a much lesser degree than limiting the surface stress; (iii) large increases in the roughness lengths for heat and moisture transfer are needed to significantly impact the intensity forecast; (iv) the different roughness length formula for surface transfer coefficients notably increases C _h/C _d ratio from 0.59 to 0.79 for 25 m s⁻¹ and 0.41 to 0.75 for 50 m s⁻¹ that significantly increases the predicted cyclone intensity; and (v) cooling of the SST by −5.8°C in 48 h reduces the maximum surface wind speed by −32 kt, or 16.5 m s⁻¹, at 48 h forecast. These results suggest that a surface flux parameterization scheme suitable for tropical cyclone intensity forecast must correctly model the leveling-off character of surface stress and C _h/C _d ratio at high winds. All modifications to surface flux calculation have little influence on 48 h track forecasts, even though they may significantly impact the intensity forecasts.

A statistical cyclone intensity prediction (SCIP) model for the Bay of Bengal

A statistical model for predicting the intensity of tropical cyclones in the Bay of Bengal has been proposed. The model is developed applying multiple linear regression technique. The model parameters are determined from the database of 62 cyclones that developed over the Bay of Bengal during the period 1981–2000. The parameters selected as predictors are: initial storm intensity, intensity changes during past 12 hours, storm motion speed, initial storm latitude position, vertical wind shear averaged along the storm track, vorticity at 850 hPa, Divergence at 200 hPa and sea surface temperature (SST). When the model is tested with the dependent samples of 62 cyclones, the forecast skill of the model for forecasts up to 72 hours is found to be reasonably good. The average absolute errors (AAE) are less than 10 knots for forecasts up to 36 hours and maximum forecast error of order 14 knots occurs at 60 hours and 72 hours. When the model is tested with the independent samples of 15 cyclones (during 2000 to 2007), the AAE is found to be less than 13 knots (ranging from 5.1 to 12.5 knots) for forecast up to 72 hours. The model is found to be superior to the empirical model proposed by Roy Bhowmik et al (2007) for the Bay of Bengal.     

Stochastic simulation model for tropical cyclone tracks,with special emphasis on landfall behavior

We consider a spatial stochastic model for the simulation of tropical cyclone tracks, which has recently been introduced. Cyclone tracks are represented as labeled polygonal lines, which are described by the movement directions, translational speeds, and wind speeds of the cyclones in regular 6-h intervals. In the present paper, we compare return levels for wind speeds of historically observed cyclone tracks with those generated by the simulator, where a mismatch is shown for most of the considered coastal regions. To adjust this discrepancy, we develop a stochastic algorithm for acceptance and rejection of simulated cyclone tracks with landfall. It is based on the fact that the locations, translational speeds, and wind speeds of cyclones at landfall constitute three-dimensional Poisson point processes, which are a basic model type in stochastic geometry. Due to that, a well-known thinning property of Poisson processes can be applied. This means that to each simulated cyclone, an acceptance probability is assigned, which is higher for cyclones with suitable landfall characteristics and lower for implausible ones. More intuitively, the algorithm comprises the simulation of a more comprehensive cyclone event set than needed and the random selection of those tracks that best match historical observations at landfall. A particular advantage of our algorithm is its applicability to multiple landfalls, i.e., to cyclones that successively make landfall at two geographically distinct coastlines, which is the most relevant case in applications. It turns out that the extended simulator provides a much better accordance between landfall characteristics of historical and simulated cyclone tracks.     

Modern and historical tropical cyclone and tsunami deposits at the coast of Myanmar: Implications for their identification and preservation in the geological record

The catastrophic storm surge of tropical cyclone Nargis in May 2008 demonstrated Myanmar's exposure to coastal flooding. The investigation of sediments left by tropical cyclone Nargis and its predecessors is an important contribution to prepare for the impact of future tropical cyclones and tsunamis in the region, because they may extend the database for long-term hazard assessment beyond the relatively short instrumental and historical record. This study, for the first time, presents deposits of modern and historical tropical cyclones and tsunamis from the coast of Myanmar. The aim is to establish regional sedimentary characteristics that may help to identify and discriminate cyclones and tsunamis in the geological record, and to document post-depositional changes due to tropical weathering in the first years after deposition. These findings if used to interpret older deposits will extend the existing instrumental record of flooding events in Myanmar. Evaluating deposits that can be related to specific events, such as the 2006 tropical cyclone Mala and the 2004 Indian Ocean tsunami, indicates similar sedimentary characteristics for both types of sediments. Landward thinning and fining trends, littoral sediment sources and sharp lower contacts allow for the differentiation from underlying deposits, while discrimination between tropical cyclone and tsunami origin is challenging based on the applied methods. The modern analogues also demonstrate a rather low preservation potential of the sand sheets due to carbonate dissolution, formation of organic top soils, and coastal erosion. However, in coastal depressions sand sheets of sufficient thickness (>10 cm) may be preserved where the shoreline is prograding or stable. In the most seaward swale of a beach-ridge plain at the Rakhine coast, two sand sheets have been identified in addition to the deposits of 2006 tropical cyclone Mala. Based on a combination of optically stimulated luminescence, radiocarbon and ¹³⁷Cs dating, the younger sand layer is related to 1982 tropical cyclone Gwa, while the older sand layer is most probably the result of an event that took place prior to 1950. Comparison with historical records indicates that the archive is only sensitive to tropical cyclones of category 4 (or higher) with landfall directly in or a few tens of kilometres north of the study area. While the presented tropical cyclone records are restricted to the last 100 years, optically stimulated luminescence ages of the beach ridges indicate that the swales landward of the one investigated in this study might provide tropical cyclone information for at least the past 700 years.     

Evaluation of the Hurricane Weather Research and Forecasting (HWRF) model for tropical cyclone forecasts over the North Indian Ocean (NIO)

Natural Hazards - The Hurricane Weather Research and Forecast (HWRF) model, which was operational at the US National Centers for Environmental Prediction, was ported in India Meteorological...     

Intensity of tropical cyclones during pre- and post-monsoon seasons in relation to accumulated tropical cyclone heat potential over Bay of Bengal

The aim of the present study is to understand the impact of oceanic heat potential in relation to the intensity of tropical cyclones (TC) in the Bay of Bengal during the pre-monsoon (April–May) and post-monsoon (October–November) cyclones for the period 2006–2010. To accomplish this, the two-layer gravity model (TLGM) is employed to estimate daily tropical cyclone heat potential (TCHP) utilizing satellite altimeter data, satellite sea surface temperature (SST), and a high-resolution comprehensive ocean atlas developed for Indian Ocean, subsequently validated with in situ ARGO profiles. Accumulated TCHP (ATCHP) is estimated from genesis to the maximum intensity of cyclone in terms of minimum central pressure along their track of all the cyclones for the study period using TLGM generated TCHP and six-hourly National Centre for Environmental Prediction Climate Forecast System Reanalysis data. Similarly, accumulated sea surface heat content (ASSHC) is estimated using satellite SST. In this study, the relationship between ATCHP and ASSHC with the central pressure (CP) which is a function of TC intensity is developed. Results reveal a distinct relationship between ATCHP and CP during both the seasons. Interestingly, it is seen that requirement of higher ATCHP during pre-monsoon cyclones is required to attain higher intensity compared to post-monsoon cyclones. It is mainly attributed to the presence of thick barrier layer (BL) resulting in higher enthalpy fluxes during post-monsoon period, where as such BL is non-existent during pre-monsoon period.     

A simple Monte Carlo method for estimating the chance of a cyclone impact

Natural Hazards - Cyclones endanger life and cause great financial impact on interior and coastal regions through the destruction of buildings and land. Governments need to have a way of estimating...     

Modeling of tropical cyclone activity over the North Indian Ocean using generalised additive model and machine learning techniques: role of Boreal summer intraseasonal oscillation

Natural Hazards - This study investigates the contribution of Boreal Summer Intraseasonal Oscillation (BSISO) to the tropical cyclone (TC) activity over the North Indian Ocean (NIO) and assesses...     

Holocene changes in tropical wind intensity and rainfall: Evidence from southeast Ghana

Remnants of a fixed aeolian dune ridge occur along the southeast coast of Ghana, just behind the present shoreline. Aeolian sands also cover extensive areas of the Accra Plains. No dunes are present here, the sands mainly occurring as sheets which blanket an early Holocene landscape. The sediments are of mid-Holocene age and were deposited during the interval 4500 B.P.–3800 yr B.P., when the southwesterly winds were stronger than they are at present and much of tropical Africa seems to have been subject to marked aridity. The onset of drier, windier conditions around 4500 yr B.P. brought to an end the more equable climates than had characterized much of West Africa during the earlier Holocene. Aridity, intensified winds, and desert expansion between 4500 and 3800 yr B.P. parallel environmental conditions in tropical continental areas at the height of the Late Pleistocene glaciation.     

About the observed and future changes in temperature extremes over India

An attempt is made in the present study to analyse observed and model simulated temperature extremes over Indian region. Daily maximum and minimum temperature data at 121 well-distributed stations for the period 1970–2003 have been used to study the observed changes in objectively defined values of temperature extremes. In addition, an assessment of future scenarios of temperature extremes associated with increase in the concentration of atmospheric greenhouse gases is done using simulations of a state-of-the-art regional climate modelling system known as PRECIS (Providing Regional Climate for Impact Studies) performed to generate the climate for the present (1961–1990) and future projections for the period 2071–2100. Observational analysis done with 121 stations suggests the widespread warming through increase in intensity and frequency of hot events and also with decrease in frequency of cold events. More than 75% stations show decreasing trend in number of cold events and about 70% stations show increasing trend in hot events. Percentage of stations towards the warming through intensity indices of highest maximum temperature, lowest minimum temperature is 78 and 71% stations, respectively. Remaining stations show opposite trends, however, most of them are statistically insignificant. Observational analysis for India as a whole also shows similar results. Composite anomalies for monthly temperature extremes over two equal parts of the data period show increase (decrease) in the frequency of hot (cold) events for all months. In general, PRECIS simulations under both A2 and B2 scenarios indicate increase (decrease) in hot (cold) extremes towards the end of twenty-first century. Both show similar patterns, but the B2 scenario shows slightly lower magnitudes of the projected changes. Temperatures are likely to increase in entire calendar year, but the changes in winter season are expected to be prominent. Diurnal temperature range is expected to decrease in winter (JF) and pre-monsoon (MAM) months.