1. Introduction
The zonal mean precipitation peaks in the tropics, the heating of which is the major driving force of the general circulation of Earth’s atmosphere. Inside the tropics, the horizontal distribution of precipitation is quite inhomogeneous. The inhomogeneity must be attributable to the geographical distribution of land and ocean. The differences in precipitation characteristics between over the land and over the ocean have been recognized (e.g., Takayabu 2006), and the water circulation between the land and the ocean has been discussed as an aspect of global water circulation (e.g., Hartmann 1994). We should recognize, however, that significant amounts of precipitation concentrate in the coastal region as demonstrated by the recent satellite precipitation measurement (see Fig. 1); this is a somewhat different viewpoint from distinguishing precipitation over the open ocean and that over the inland region.
Precipitation distribution based on TRMM 3A25 data over (top) Southeast Asia, (middle) South and Central America, and (bottom) Africa. The values are the means for the period between December 1997 and January 2011.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0484.1
The behaviors of such coastal precipitation have been studied for several remarkable regions, such as the western coast of Sumatera Island (Mori et al. 2004; Wu et al. 2009), northern Kalimantan (Borneo) Island (Houze et al. 1981; Ichikawa and Yasunari 2006), northwestern South America (Mapes et al. 2003a,b; Warner et al. 2003), the Bay of Bengal (Zuidema 2003; Xie et al. 2006), the western coast of the Indian subcontinent (Grossman and Durran 1984), and the eastern coast of the Asian winter monsoon regions (Chang et al. 2005) including the eastern coast of the Indochina Peninsula (Yokoi and Matsumoto 2008; Wu et al. 2011; Chen et al. 2012). However, these studies investigated the precipitation characteristics and mechanisms for the individual regions, and no research up to this point has quantified precipitation in coastal regions specifically, or compared it to precipitation in other regions.
In this study, we sought to accomplish the following with the overall objective of describing the climatological characteristics of precipitation in the coastal regions of the tropics: 1) to define “coastal regions” from the standpoint of precipitation distribution; 2) to quantify precipitation inside the coastal regions and to compare it to that outside, including both land and ocean sides; and 3) to provide a brief discussion on quantitative characteristics of precipitation in the coastal regions.
2. Data
Monthly-averaged precipitation data generated by the precipitation radar on the Tropical Rainfall Measuring Mission (TRMM) satellite (3A25; e.g., Kummerow et al. 2000) were used. The dataset has a horizontal resolution of 0.5° × 0.5° over ~13 years from December 1997 to January 2011. The area covered by the TRMM 3A25 data (37°S–37°N globally) was analyzed in this study. All data were averaged for each cell to generate a single 13-yr mean climatological value of annual precipitation.
For elevation, the Global Land One-km Base Elevation (GLOBE; Hastings et al. 1999) data were used. Since the original horizontal gridcell size of this dataset (30 s × 30 s) was too high for the 0.5° × 0.5° precipitation data, the elevation data resolution was reduced (averaged) to 4 min × 4 min. Note that we included in our analysis all the topography resolved by the GLOBE data of which horizontal resolution is about 2.83 times gridcell size (about 2.6 km at the equator).
Examining four neighboring GLOBE cells as a set each time, if a set contains both land and ocean cells, the center node of the four-cell set is designated as being on the coastline. Note that a lake cell is regarded as a land cell in this study. The “distance from the coastline” is defined for each TRMM data cell using the distance from the nearest point on the coastline. We define the distance from the coastline in the direction toward land as being positive and that toward the ocean as being negative. In the case when a TRMM data cell along the coastline contains both ocean and land GLOBE grid cells, we designate each TRMM data cell as land or ocean based on the proportion of the ocean or land GLOBE grid cells contained within each TRMM data cell. For example, the TRMM cells containing more GLOBE ocean cells than GLOBE land cells are designated to be ocean, and vice versa.
Using the above data, annual precipitation per unit area was calculated for each 50-km bin as a function of the distance from the coastline. The obtained annual precipitation for each bin is considered valid only when the bin contains 400 or more TRMM data cells.
3. Precipitation as a function of distance from the coastline
First, let us examine how precipitation amount changes with distance from the coastline (Fig. 2). Precipitation amount exhibits a sharp peak centered around the coastline and tends to decrease with distance away from the coastline. Precipitation drops sharply over a distance of ~300 km from the coastline, on both sides of the coastline, with the rate of decline being greater on land. Precipitation reaches a peak of around 1300 mm yr−1 at the coastline; at a distance of 300 km from the coastline, it falls to about 750 mm yr−1 over land but only to about 1050 mm yr−1 over the ocean. At greater distances from the coastline, precipitation increases both over land and ocean.
Relationship between precipitation amount and distance from the coastline.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0484.1
4. Precipitation in the coastal region
We define the coastal region as the area within a certain distance from the coastline, calculate the total precipitation amount in the coastal regions, and determine its percentage with respect to the total precipitation over the whole tropics analyzed here. When the coastal regions are defined as being within 300 km of the coastline, the precipitation in the coastal regions (referred to as coastal precipitation amount) is estimated to be 9.9 × 1016 mm yr−1. As described in section 3, precipitation in this region drops sharply with a distance from the coastline both over the land and over the ocean. Accordingly, it is reasonable to consider this region as being influenced by precipitation systems unique to the coastline. When we define coastal regions in this way, the proportion of coastal precipitation amount to the total precipitation for all regions (2.9 × 1017 mm yr−1) is 34% (referred to as coastal precipitation share). Outside the coastal regions, the precipitation amount accounts for 52% and 14% on the ocean and the land, respectively. Since the coastal regions account for 29% of the total area (referred to as coastal area share), it is evident that the precipitation amount per unit area in the coastal regions is higher than that in the other regions. If we define the precipitation-to-area ratio as precipitation share divided by area share, a ratio of 1 means that the amount of precipitation over the coastal region is commensurate with the area share. When the coastal region is defined as being within 300 km of the coastline, the precipitation–area ratio is 1.17, meaning that the precipitation amount over the coastal region is 1.17 times that of the all area mean.
The above results are estimated when the coastal region is defined as being within 300 km of the coastline. The results will change when this threshold distance is varied between 50 and 2000 km, as presented in Fig. 3. Note that the whole Maritime Continent (including Indonesia and the Philippine Islands) becomes the coastal region if a threshold distance larger than 400 km is used. Coastal precipitation share (coastal area share) changes from 9.3% (6.8%) when the threshold is set at 50 km from the coastline to 98% (98%) when the threshold is extended to 2000 km (Fig. 3a). While it is expected that the coastal precipitation share should increase with threshold (distance) since the coastal area share increases, it is worth noting that the coastal precipitation share remains greater than the coastal area share, regardless of how the coastal region is defined. In other words, the precipitation–area ratio remains greater than 1.0, regardless of the definition of coastal region (Fig. 3b). The precipitation–area ratio is the highest (1.36) when the coastal region is defined as being within 50 km of the coastline, and the ratio declines drastically as the threshold distance is increased to 300 km (approximately 1.17) and 1000 km (approximately 1.06).
Changes in (a) coastal area share (dotted line) and coastal precipitation share (solid line), and (b) precipitation–area ratio, when the definition of the coastal region is changed (see the text for details). The vertical solid line denotes a distance from the coastline of 300 km.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0484.1
5. Highest annual precipitation at the coast
Let us examine the grid number variation with respect to the annual precipitation amount and compare it between the coastal regions and all regions (Fig. 4). In both coastal regions and all regions, the number of grid cells tends to decrease with increasing amount of annual precipitation (Fig. 4a). To better visualize the differences in grid number variation between coastal regions and all regions, Fig. 4b presents the grid number variation normalized by the total grid number in each of the two respective regions. It can be seen that the grid number variation in the coastal regions is skewed toward the higher amount of precipitation.
(a) Grid number distributions for coastal regions (red line) and for all regions (black line). (b) As in (a), but for those normalized by the total grid number for the corresponding regions. (c) Coastal share of grid number with respect to the total precipitation for each precipitation range.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0484.1
When we divide the grid number variation for the coastal regions (red line in Fig. 4a) by that of all regions (black line in Fig. 4a), we can see the coastal share of grid number for each precipitation range (Fig. 4c). The grid number in coastal regions accounts for a fairly consistent 25%–35% of the total grid number of all regions for an amount up to 3000 mm yr−1. However, the share of grid number attributable to coastal regions increases dramatically to over 90% above 3500 mm yr−1. In other words, heavy precipitation in excess of 3500 mm yr−1 occurs almost exclusively in the coastal regions and rarely in the other regions. This suggests that there must be precipitation systems unique to the coastal regions that are capable of producing higher annual precipitation on Earth. Further research is needed to elucidate the precipitation systems that produce such heavy precipitation in the coastal regions and to assess the physical significance of the 3500 mm yr−1 amount.
6. Summary and discussion
In this study, we quantified the annual mean precipitation amount in the coastal regions. The coastal regions have not been treated up to this point as unique regions from the perspective of precipitation distribution. We also compared the precipitation amount in the coastal regions with the amount in other (land and ocean) regions. Using the TRMM 3A25 precipitation data for ~13 years (from December 1997 to January 2011) and the GLOBE data, we defined the coastline and investigated the precipitation amount as a function of distance from the coastline.
Precipitation amount peaks at the coastline and decreases with distance away from the coastline. Precipitation amount decreases rapidly up to a distance of 300 km from the coastline over both the land and the ocean, although the rate of decline is somewhat lower over the ocean.
By defining the coastal region as an area within 300 km of the coastline, it is found that the precipitation in the coastal regions accounts for approximately 34% of the total precipitation, while the precipitation in noncoastal maritime and terrestrial regions accounts for 52% and 14% of the total precipitation, respectively. Meanwhile, the share of the total area occupied by the coastal regions is 29% (using 300 km as the threshold), indicating that the precipitation per unit area is higher in the coastal regions than that in other regions.
Considering the variation in the coastal share of grid number with respect to the annual precipitation amount, we found that more than 90% of annual precipitation with the amount of 3500 mm yr−1 or more occurs exclusively in the coastal regions and hardly in the other regions. Precipitation systems unique to the coastal regions are needed for producing the highest annual precipitation on Earth.
The precipitation over the coastal region should be distinguished from that over the ocean and the land in terms of not only amount but also mechanism. The principal and fundamental cause of producing the coastal precipitation must be the diurnal land–sea circulation driven by the heat contrast between them, as suggested by the previous studies of the western coast of Sumatera Island (Mori et al. 2004; Wu et al. 2009), northern Kalimantan (Borneo) Island (Houze et al. 1981), northwestern South America (Mapes et al. 2003a,b; Warner et al. 2003), and the Bay of Bengal (Zuidema 2003). The land–sea circulation over the coastal area most systematically generates convections and precipitation in the tropics where cyclones and frontal systems are not active unlike those in the midlatitude, and large-scale cloud systems, such as the Madden–Julian oscillation and other intraseasonal oscillations, are mainly developed over the open ocean as a result of ocean–atmosphere interaction. Recently, Yamanaka (2016) suggested that the long coastline is responsible for the world’s largest regional rainfall over the Maritime Continent.
Other processes have been suggested for producing the coastal precipitation, which may modulate or intensify the diurnal precipitation over the coastal region. Grossman and Durran (1984) suggested that the topographic blocking by the Western Ghats in India affects the precipitation over the western coast of the Indian subcontinent. An interaction between the topography and the monsoon flow is suggested for the precipitation mechanism of the Asian winter monsoon regions, such as the eastern coasts of the Indochina Peninsula and the Philippines (Chang et al. 2005). An importance of the arrival of westward-propagating disturbances is further suggested for autumn–winter precipitation over the eastern coast of the Indochina Peninsula (Yokoi and Matsumoto 2008; Wu et al. 2011; Chen et al. 2012). It is suggested that the surface roughness contrast between the ocean and land has some effect on the development of convection near the coastline when the near-surface flow is perpendicular to the coastline (Fernández et al. 1997). One of these processes may not fully explain all the coastal precipitation, and some of these processes are complexly involved in producing the coastal precipitation. The general explanation behind these processes should be clarified, which would be obtained both from investigating the nature of coastal precipitation and from performing the numerical experiments. Further investigation of the coastal precipitation would lead to deepening our understanding on the tropical climate.
Acknowledgments
The authors thank Prof. Kazuaki Yasunaga, the University of Toyama, for providing his GrADS-format TRMM data converted from the original data. The authors also thank the two anonymous reviewers for their constructive comments.
REFERENCES
Chang, C.-P., Z. Wang, J. McBride, and C.-H. Liu, 2005: Annual cycle of Southeast Asia—Maritime Continent rainfall and the asymmetric monsoon transition. J. Climate, 18, 287–301, doi:10.1175/JCLI-3257.1.
Chen, T.-C., J.-D. Tsay, M.-C. Yen, and J. Matsumoto, 2012: Interannual variation of the late fall rainfall in central Vietnam. J. Climate, 25, 392–413, doi:10.1175/JCLI-D-11-00068.1.
Fernández, C., M. A. Gaertner, C. Gallardo, and M. Castro, 1997: Simulation of a long-lived meso-β scale convective system over the Mediterranean coast of Spain. Part II: Sensitivity to external forcings. Meteor. Atmos. Phys., 62, 179–200, doi:10.1007/BF01029701.
Grossman, R. L., and D. R. Durran, 1984: Interaction of low-level flow with the Western Ghat Mountains and offshore convection in the summer monsoon. Mon. Wea. Rev., 112, 652–672, doi:10.1175/1520-0493(1984)112<0652:IOLLFW>2.0.CO;2.
Hartmann, D. L., 1994: Global Physical Climatology. Academic Press, 411 pp.
Hastings, D. A., and Coauthors, Eds., 1999: The Global Land One-kilometer Base Elevation (GLOBE) Digital Elevation Model, version 1.0. NOAA, National Geophysical Data Center. [Available online at http://www.ngdc.noaa.gov/mgg/topo/globe.html.]
Houze, R. A., S. G. Geotis, F. D. Marks, and A. K. West, 1981: Winter monsoon convection in the vicinity of North Borneo. Part I: Structure and time variation of the clouds and precipitation. Mon. Wea. Rev., 109, 1595–1614, doi:10.1175/1520-0493(1981)109<1595:WMCITV>2.0.CO;2.
Ichikawa, H., and T. Yasunari, 2006: Time–space characteristics of diurnal rainfall over Borneo and surrounding oceans as observed by TRMM-PR. J. Climate, 19, 1238–1260, doi:10.1175/JCLI3714.1.
Kummerow, C., and Coauthors, 2000: The status of the Tropical Rainfall Measuring Mission (TRMM) after two years in orbit. J. Appl. Meteor., 39, 1965–1982, doi:10.1175/1520-0450(2001)040<1965:TSOTTR>2.0.CO;2.
Mapes, B. E., T. T. Warner, M. Xu, and A. J. Negri, 2003a: Diurnal patterns of rainfall in northwestern South America. Part I: Observations and context. Mon. Wea. Rev., 131, 799–812, doi:10.1175/1520-0493(2003)131<0799:DPORIN>2.0.CO;2.
Mapes, B. E., T. T. Warner, and M. Xu, 2003b: Diurnal patterns of rainfall in northwestern South America. Part III: Diurnal gravity waves and nocturnal convection offshore. Mon. Wea. Rev., 131, 830–844, doi:10.1175/1520-0493(2003)131<0830:DPORIN>2.0.CO;2.
Mori, S., and Coauthors, 2004: Diurnal land–sea rainfall peak migration over Sumatera Island, Indonesian Maritime Continent, observed by TRMM satellite and intensive rawinsonde soundings. Mon. Wea. Rev., 132, 2021–2039, doi:10.1175/1520-0493(2004)132<2021:DLRPMO>2.0.CO;2.
Takayabu, Y. N., 2006: Rain-yield per flash calculated from TRMM PR and LIS data and its relationship to the contribution of tall convective rain. Geophys. Res. Lett., 33, L18705, doi:10.1029/2006GL027531.
Warner, T. T., B. E. Mapes, and M. Xu, 2003: Diurnal patterns of rainfall in northwestern South America. Part II: Model simulations. Mon. Wea. Rev., 131, 813–829, doi:10.1175/1520-0493(2003)131<0813:DPORIN>2.0.CO;2.
Wu, P., M. Hara, J. Hamada, M. D. Yamanaka, and F. Kimura, 2009: Why a large amount of rain falls over the sea in the vicinity of western Sumatra Island during nighttime. J. Appl. Meteor. Climatol., 48, 1345–1361, doi:10.1175/2009JAMC2052.1.
Wu, P., Y. Fukutomi, and J. Matsumoto, 2011: An observational study of the extremely heavy rain event in northern Vietnam during 30 October–1 November 2008. J. Meteor. Soc. Japan, 89A, 331–344, doi:10.2151/jmsj.2011-A23.
Xie, S.-P., H. Xu, N. H. Saji, Y. Wang, and W. T. Liu, 2006: Role of narrow mountains in large-scale organization of Asian monsoon convection. J. Climate, 19, 3420–3429, doi:10.1175/JCLI3777.1.
Yamanaka, M. D., 2016: Physical climatology of Indonesian Maritime Continent: An outline to comprehend observational studies. Atmos. Res., in press.
Yokoi, S., and J. Matsumoto, 2008: Collaborative effects of cold surge and tropical depression–type disturbance on heavy rainfall in central Vietnam. Mon. Wea. Rev., 136, 3275–3287, doi:10.1175/2008MWR2456.1.
Zuidema, P., 2003: Convective clouds over the Bay of Bengal. Mon. Wea. Rev., 131, 780–798, doi:10.1175/1520-0493(2003)131<0780:CCOTBO>2.0.CO;2.
