• Beven, J. L., and Coauthors, 2008: Atlantic hurricane season of 2005. Mon. Wea. Rev., 136 , 11091173.

  • Burgan, R. E., , and R. A. Hartford, 1993: Monitoring vegetation greenness with satellite data. General Tech. Rep. INT-297, Ogden, UT, U.S. Department of Agriculture, Forest Service, Intermountain Research Station, 13 pp.

    • Search Google Scholar
    • Export Citation
  • Curtis, S., 2002: Interannual variability of the bimodal distribution of summertime rainfall over Central America and tropical storm activity in the far-eastern Pacific. Climate Res., 22 , 141146.

    • Search Google Scholar
    • Export Citation
  • Curtis, S., , and D. W. Gamble, 2008: Regional variations of the Caribbean mid summer drought. Theor. Appl. Climatol., 94 , 2534.

  • Erasmi, S., , P. Propastin, , M. Kappas, , and O. Panferov, 2009: Spatial patterns of NDVI variation over Indonesia and their relationship to ENSO warm events during the period 1982–2006. J. Climate, 22 , 66126623.

    • Search Google Scholar
    • Export Citation
  • Gamble, D. W., , and S. Curtis, 2008: Caribbean precipitation: Review, model, and prospect. Prog. Phys. Geogr., 32 , 265276.

  • Gamble, D. W., , D. Parnell, , and S. Curtis, 2007: Spatial variability of the Caribbean mid-summer drought and relation to North Atlantic high circulation. Int. J. Climatol., 28 , 343350.

    • Search Google Scholar
    • Export Citation
  • Gamble, D. W., , D. Campbell, , T. L. Allen, , D. Barker, , S. Curtis, , D. F. M. McGregor, , and E. J. Popke, 2010: Climate change, drought, and Jamaican agriculture: Local knowledge and the climate record. Ann. Assoc. Amer. Geogr., in press.

    • Search Google Scholar
    • Export Citation
  • Giannini, A., , M. Cane, , and Y. Kushnir, 2001: Interdecadal changes in the ENSO teleconnections to the Caribbean region and the North Atlantic Oscillation. J. Climate, 14 , 28672879.

    • Search Google Scholar
    • Export Citation
  • Hastenrath, S., 1966: The flux of atmospheric water vapor over the Caribbean Sea and the Gulf of Mexico. J. Appl. Meteor., 5 , 778788.

    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., and Coauthors, 2007: The TRMM Multisatellite Precipitation Analysis (TMPA): Quasi-global, multiyear, combined-sensor precipitation estimates at fine scales. J. Hydrometeor., 8 , 3855.

    • Search Google Scholar
    • Export Citation
  • Inoue, M., , I. Handoh, , and G. Bigg, 2002: Bimodal distribution of tropical cyclogenesis in the Caribbean: Characteristics and environmental factors. J. Climate, 15 , 28972905.

    • Search Google Scholar
    • Export Citation
  • Kogan, F., 2001: Operational space technology for global vegetation assessment. Bull. Amer. Meteor. Soc., 82 , 19491964.

  • Kogan, F., 2002: World droughts in the new millenium from AVHRR-based vegetation health indices. Eos, Trans. Amer. Geophys. Union, 83 , 557564.

    • Search Google Scholar
    • Export Citation
  • Li, Z., , and M. Kafatos, 2000: Interannual variability of vegetation in the United States and its relation to El Niño/Southern Oscillation. Remote Sens. Environ., 71 , 239247.

    • Search Google Scholar
    • Export Citation
  • Magaña, V., , J. Amador, , and S. Medina, 1999: The midsummer drought over Mexico and Central America. J. Climate, 12 , 15771588.

  • Mapes, B. E., , P. Liu, , and N. Buenning, 2005: Indian monsoon onset and the Americas midsummer drought: Out-of-equilibrium responses to smooth seasonal forcing. J. Climate, 18 , 11091115.

    • Search Google Scholar
    • Export Citation
  • McGregor, D. F. M., , D. Barker, , and D. Campbell, 2009: Environmental change and Caribbean food security: Recent hazard impacts and domestic food production in Jamaica. Global Change and Caribbean Vulnerability: Environment, Economy, and Society at Risk? D. F. M. McGregor and D. Barker, Eds., University of West Indies Press, 197–217.

    • Search Google Scholar
    • Export Citation
  • Muñoz, E., , A. J. Busalacchi, , S. Nigam, , and A. Ruiz-Barradas, 2008: Winter and summer structure of the Caribbean low-level jet. J. Climate, 21 , 12601276.

    • Search Google Scholar
    • Export Citation
  • Peters, A., , E. Walter-Shea, , L. Ji, , A. Vina, , M. Hayes, , and M. Svoboda, 2002: Drought monitoring with NDVI-based standardized vegetation index. Photogramm. Eng. Remote Sens., 68 , 7175.

    • Search Google Scholar
    • Export Citation
  • Poveda, G., , A. Jaramillo, , M. M. Gil, , N. Quiceno, , and R. I. Mantilla, 2001: Seasonality in ENSO-related precipitation, river discharges, soil moisture, and vegetation index in Columbia. Water Resour. Res., 37 , 21692178.

    • Search Google Scholar
    • Export Citation
  • Rouse, J. W., , R. H. Haas Jr., , J. A. Schell, , and D. W. Deering, 1974: Monitoring vegetation systems in the Great Plains with ERTS. Third ERTS-1 Symposium, S. C. Freden, E. P. Mercanti, and M. A. Becker, Eds., Vol. 1, Special Publ. NASA SP-351, NASA, 309–317.

    • Search Google Scholar
    • Export Citation
  • Small, R. J. O., , S. P. de Szoeke, , and S-P. Xie, 2007: The Central American mid-summer drought: Regional aspects and large-scale forcing. J. Climate, 20 , 48534873.

    • Search Google Scholar
    • Export Citation
  • Wang, C. Z., 2007: Variability of the Caribbean low-level jet and its relation to climate. Climate Dyn., 29 , 411422.

  • Wang, C. Z., , S. K. Lee, , and D. B. Enfield, 2008: Climate response to anomalously large and small Atlantic warm pools during the summer. J. Climate, 21 , 24372450.

    • Search Google Scholar
    • Export Citation
  • Wang, J., , P. M. Rich, , and K. P. Price, 2003: Temporal responses of NDVI to precipitation and temperature in the central Great Plains, USA. Int. J. Remote Sens., 24 , 23452364.

    • Search Google Scholar
    • Export Citation
  • Yang, W., , L. Yang, , and J. W. Merchant, 1997: An assessment of AVHRR/NDVI ecoclimatical relations in Nebraska, USA. Int. J. Remote Sens., 18 , 21612180.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Map of study area in Jamaica with St. Elizabeth parish (shaded in gray) and related spatial coverage of TRMM rainfall estimates (solid line) and Terra MODIS NDVI values (hatched area).

  • View in gallery

    Annual climatological average values of NDVI (solid line) and precipitation estimates (bar graphs; mm day−1) for the study areas in the 2001–07 period. Annual data from 2005 are omitted because of the nonpresence of the midsummer drought.

  • View in gallery

    Mean rainfall (bar graph; mm day−1) and NDVI (line plot) for the study area for each year from 2001 to 2007.

  • View in gallery

    Percentage change in NDVI values across Jamaica with the St. Elizabeth parish outlined in black. Percent change is computed by dividing the difference of the 10–25 Jun 2001–07 (excluding 2005) average NDVI and 28 Jul–12 Aug 2001–07 (excluding 2005) NDVI by the 10–25 Jun NDVI (see Fig. 2).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 21 21 5
PDF Downloads 13 13 3

The Midsummer Dry Spell’s Impact on Vegetation in Jamaica

View More View Less
  • 1 Department of Geography, East Carolina University, Greenville, North Carolina
  • 2 Department of Geography and Geology, University of North Carolina at Wilmington, Wilmington, North Carolina
© Get Permissions
Full access

Abstract

The annual rainfall pattern of the intra-Americas sea reveals a bimodal feature with a minimum during the midsummer known as the midsummer dry spell (MSD). A first attempt is made to examine the impact of the MSD on vegetation through a normalized difference vegetation index (NDVI) analysis in Jamaica. Tropical Rainfall Measuring Mission rainfall estimates and NDVI derived from the Terra Moderate Resolution Imaging Spectroradiometer highlight a consistent MSD feature in both rainfall and vegetative vigor. Spatial variation of this MSD NDVI response is evident throughout Jamaica, with the strongest relationship between the rainfall reduction and NDVI decline throughout the southern portions of Jamaica including the area of major domestic food production. In all years except 2005 there is a notable reduction from early-summer NDVI to midsummer NDVI in this agricultural region. However, the lagged vegetative response undergoes clear interannual variation and is affected by other forcings besides rainfall, such as brush fires and extreme wind.

* Current affiliation: Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami, Florida

Corresponding author address: Scott Curtis, Dept. of Geography, East Carolina University, Brewster A232, Greenville, NC 27858. Email: curtisw@ecu.edu

Abstract

The annual rainfall pattern of the intra-Americas sea reveals a bimodal feature with a minimum during the midsummer known as the midsummer dry spell (MSD). A first attempt is made to examine the impact of the MSD on vegetation through a normalized difference vegetation index (NDVI) analysis in Jamaica. Tropical Rainfall Measuring Mission rainfall estimates and NDVI derived from the Terra Moderate Resolution Imaging Spectroradiometer highlight a consistent MSD feature in both rainfall and vegetative vigor. Spatial variation of this MSD NDVI response is evident throughout Jamaica, with the strongest relationship between the rainfall reduction and NDVI decline throughout the southern portions of Jamaica including the area of major domestic food production. In all years except 2005 there is a notable reduction from early-summer NDVI to midsummer NDVI in this agricultural region. However, the lagged vegetative response undergoes clear interannual variation and is affected by other forcings besides rainfall, such as brush fires and extreme wind.

* Current affiliation: Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami, Florida

Corresponding author address: Scott Curtis, Dept. of Geography, East Carolina University, Brewster A232, Greenville, NC 27858. Email: curtisw@ecu.edu

1. Introduction

The Gulf of Mexico and the Caribbean Sea, collectively referred to here as the intra-Americas sea (IAS), experience an annual bimodal precipitation pattern with peaks observed in the late-spring and late-summer months (Hastenrath 1966). The relative precipitation minimum occurring between the bimodal peaks is known as the midsummer drought or the midsummer dry spell (MSD) (Magaña et al. 1999; Gamble and Curtis 2008). The MSD occurs annually, but its pattern in terms of onset, duration, and magnitude exhibits interannual variability spatially (throughout the region) and temporally (during the boreal summer) (Curtis and Gamble 2008) because of the El Niño–Southern Oscillation (Curtis 2002), the North Atlantic Oscillation (Giannini et al. 2001), and other factors. The reduction of precipitation, combined with associated reduced cloud coverage and related increased surface heating during the MSD, can negatively affect vegetation within the IAS region, resulting in crop failure and agricultural stress. Interviews of Jamaican farmers reveal that they are more concerned about the timing of the MSD onset rather than duration or severity (McGregor et al. 2009; Gamble et al. 2010). Further, those surveyed state that midsummer dryness has the largest impact on agricultural yields.

Research to date has focused on the physical climate mechanisms controlling the development and evolution of the MSD (Mapes et al. 2005; Small et al. 2007; Gamble and Curtis 2008; etc.)—in particular, sea surface temperature (Wang et al. 2008) and the Caribbean low-level jet (Wang 2007; Muñoz et al. 2008). However, there is a paucity of research on the biophysical response to the MSD. Incorporating vegetative response with other current atmospheric/oceanic research provides a more complete understanding through an earth integrated-systems approach to the MSD and can be connected to societal effects as experienced by farmers.

The normalized difference vegetation index (NDVI), defined as the ratio between the difference of the reflectance in the near-infrared and red wavelengths and the sum of the two, is commonly used to assess vegetation vigor (Rouse et al. 1974; Yang et al. 1997). Previous studies have concluded that a strong, linear relationship exists between precipitation and NDVI at both concurrent and lagged time scales (Wang et al. 2003), and NDVI has been used for near-real-time drought monitoring, for drought forecasting, and as a method to catalog drought intensity (Burgan and Hartford 1993; Kogan 2001, 2002; Peters et al. 2002). By doing so, the timing and extent of drought can be discerned by comparing vegetation greenness during a particular year with that of previous years. In this case, NDVI is used to assess the impact of the MSD and to catalog its intensity as seen through vegetative response within the “breadbasket” region of an IAS island nation.

The island of Jamaica is located toward the western section of the Caribbean Sea in a region that experiences a strong MSD signature in July while also supporting a large agricultural sector. Thus, the development of the MSD often provides adverse effects for farmers, who consequently recognize the climatological persistence of the MSD during the midsummer. This is especially true in the St. Elizabeth parish, along the southwestern coast of Jamaica, which supports the majority of domestic agricultural production.

This note is a first attempt to relate the MSD to spatiotemporal patterns of vegetation vigor. The lack of submonthly in situ observations in Jamaica necessitates a satellite-based analysis. The data and methods are described in the next section, followed by results. Conclusions are offered in section 4.

2. Data and methods

The connection between the MSD rainfall and vegetation in St. Elizabeth is detected by relating a single grid cell from the 0.25° daily precipitation estimate of the Tropical Rainfall Measuring Mission (TRMM; Huffman et al. 2007) and an area-averaged, remotely sensed, 250-m NDVI product derived from the Terra Moderate Resolution Imaging Spectroradiometer (MODIS) from 2001 to 2007. NDVI is used to monitor vegetation conditions because in healthy green vegetation the near-infrared is highly reflective while the red is poorly reflective within the electromagnetic spectrum. NDVI values range from −1 (stressed) to +1 (healthy), indicating the vegetative vigor or greenness of the vegetation. NDVI is determined within the area bounded by 18.08°–17.92°N, 77.79°–77.48°W, and the closest TRMM node, covering the region defined with a center point of 18.125°N, 77.625°W, is used for precipitation estimation purposes (Fig. 1). A single TRMM grid box provides adequate spatial coverage for this investigation, whereas almost 9000 MODIS pixels are needed to represent the region. TRMM was averaged over 16 days to match one 16-day granule of the NDVI MODIS product.

Both TRMM rainfall estimates and MODIS NDVI data from 2001 to 2007, excluding 2005, provide evidence of an unmistakable bimodal rainfall and vegetation signal with a midsummer reduction (Fig. 2). The year 2005 lacked an MSD (Fig. 3) because of the unusual approach of both Hurricanes Dennis (category 4) and Emily (category 4) to Jamaica in July (Beven et al. 2008) and is excluded from the analysis. Tropical cyclogenesis in the Caribbean is typically bimodal, with a minimum in July (Inoue et al. 2002).

The TRMM product employed here is a widely used and well-validated multisatellite precipitation analysis, extending from 1998 to present. However, our period of record is further restricted by the MODIS observations being available only since 2001. A monthly index of precipitation over St. Elizabeth—an average of all available rain gauge data (15–18 stations)—from 2001 to 2006 was constructed to validate the TRMM node estimates. A monthly average of TRMM was correlated in the overlapping period at 0.86 (p < 0.001). The correlation between TRMM and a high-quality single station (Appleton) also shows high significance (0.80; p < 0.001). Thus, we have confidence in the reliability of the satellite product to represent rainfall in the parish.

3. Results

Annual correlation between rainfall and NDVI is positive and in most cases is strongest at a lagged response. Often vegetation change lags precipitation by up to 4 weeks but can be determined by preceding moisture conditions (Wang et al. 2003). Wang et al. (2003) state that a shorter lag period, or none at all, is commonly detected during very dry periods prior to rainfall. The peaks in rainfall and NDVI are nearly simultaneous during the early summer after low winter rainfall (Fig. 2). The vegetation response lags the MSD by two granules (32 days) and the late-summer rains by three granules (48 days) (Fig. 2). Preseason and early-season peak rainfall, typically occurring before June, act to condition the summer NDVI amplitude. An examination of maxima and minima for the six years individually shows strong interannual variability but provides a consistent picture in which the vegetation response lag increases over the course of the year (Table 1).

It is important to note that lack of precipitation in July is only one factor that leads to a decrease in vegetation vigor. The lowest recorded NDVI values during midsummer occurred during 2003 (NDVI = 0.301) and 2004 (NDVI = 0.376) (Fig. 3; Table 2). Brushfires were reported within the region during 2004 (D. Campbell 2008, personal communication), and the very low NDVI value during the 2003 summer may be associated with a significant weather event on 3 July, described by the Jamaican meteorological office as a “freak storm (trough) reported in St. Elizabeth that produced damage to both crops and roofs.” Higher midsummer NDVI values occurred in 2002 (NDVI = 0.42) and 2006 (NDVI = 0.456) (Fig. 3; Table 2). June 2002 rainfall was the highest during the record, suggesting that high soil moisture overwhelmed the impact of the MSD. The year 2006 experienced consistent, albeit light, rains from early April through the end of July (Fig. 3).

For this study, correlations between rainfall and NDVI were computed from April to November, consistent with Gamble et al.’s (2007) defined bimodal season, to focus upon MSD-related vegetation changes and to avoid vegetation signals originating from the dry winter season (December–February). This gives an overall indication of seasonal vegetation lag, but, as noted above, the lag appears to increase from April to November. No concurrent correlation between rainfall and NDVI is significant, but the largest value (r = +0.35) occurred during 2004 (Table 3). The 2003 rainfall and NDVI correlation is strongest (r = +0.48) at a lag of two granules (32 days), suggesting that leaf loss due to the 2003 significant weather event may have contributed toward the NDVI decrease in addition to the preceding anomalously low precipitation (Fig. 3). The 2006 season and overall “climatology” also experience their strongest correlations at a two-granule (32 day) lag, but the remaining years experience their strongest correlations at a one-granule (16 day) lag (Table 3).

Spatial variation of MSD-related NDVI within Jamaica is detected by computing an NDVI percent difference composite between points that represent the decline in vegetative vigor in midsummer (Fig. 2). This difference provides a general concept of both the magnitude of NDVI response to the MSD as well as the spatial variation across Jamaica. In this case, the peak NDVI occurs on 10–25 June, and the NDVI minimum is 28–12 August. The composite NDVI image (Fig. 4) highlights the southern regions of Jamaica that are vulnerable to NDVI decline during the midsummer from 2001 to 2007 (excluding 2005). The “domestic breadbasket” region of St. Elizabeth exhibits some of the strongest levels of NDVI decline (Fig. 4).

Understanding the variability of the midsummer NDVI can prove to be valuable when attempting to create an MSD classification system. There are numerous methods to describe MSD intensity: cloud cover reduction, various precipitation relationships, and outgoing longwave radiation, for example, but perhaps a direct land surface signal may prove to be just as useful. The percent change between early-summer NDVI maximum and midsummer minimum represents one possible MSD-NDVI classification technique. In our study region during the 2001–07 period (excluding 2005), there was an average 17% reduction in NDVI (Table 2). The highest overall percent change (−43%) occurred in 2003 and the lowest overall percent change occurred in 2006 (−25%). It is important to note that 2004 and 2006, which demonstrated little decrease in NDVI from early to midsummer (Table 2), also had the lowest correlations with precipitation (Table 3).

4. Conclusions

The inclusion of vegetative vigor quantitatively adds another valuable geographic dimension toward assessing the impact of the MSD upon the IAS region. Previous work has examined NDVI variation related to interannual events such as ENSO (Poveda et al. 2001; Li and Kafatos 2000; Erasmi et al. 2009), but none has been directed toward the study of an intraseasonal climate feature such as the MSD. The MSD has a strong impact on the southern St. Elizabeth region (a decrease of over 50%), which is consistent with farmer testimonies. The reasons contributing to this spatial variation reveal a possible mixture of local atmospheric mechanisms, land cover, and irrigation techniques. St. Elizabeth sits in the rain shadow of Jamaica and is relatively dry in comparison with the rest of the country. Further, the less vulnerable area of northern St. Elizabeth is located in a fertile region dominated by large-scale commercial agriculture. In southern St. Elizabeth, small-scale farming tends to exist on steep slopes with less favorable soils and limited irrigation, which compounds an already vegetative-stressful MSD.

This note is also provided as motivation for further work relating precipitation and agriculture at subseasonal time scales. With regard to Jamaica, we intend to collaborate with the Jamaican meteorological service and University of West Indies (Mona) to collect more in situ precipitation and crop data for validation and refinement of our results from satellites. These additional steps are important because vegetation variation can have significant impacts on hydrologic modeling and monthly and seasonal climate simulations of evapotranspiration, surface air temperature, and precipitation.

This study reminds us of the elusive and complex nature of vegetation response to external forcings but clearly shows a decrease in midsummer for the years sampled, which can at least be partially explained by the MSD. Further, mapping the spatial variability of intraseasonal NDVI change can potentially reduce farming vulnerability as progress toward midsummer forecasting is improved within Jamaica.

Acknowledgments

We thank two helpful reviewers and the editor for their comments, which improved the manuscript. Also, we acknowledge the support of NSF Grants 0718279 and 0718257.

REFERENCES

  • Beven, J. L., and Coauthors, 2008: Atlantic hurricane season of 2005. Mon. Wea. Rev., 136 , 11091173.

  • Burgan, R. E., , and R. A. Hartford, 1993: Monitoring vegetation greenness with satellite data. General Tech. Rep. INT-297, Ogden, UT, U.S. Department of Agriculture, Forest Service, Intermountain Research Station, 13 pp.

    • Search Google Scholar
    • Export Citation
  • Curtis, S., 2002: Interannual variability of the bimodal distribution of summertime rainfall over Central America and tropical storm activity in the far-eastern Pacific. Climate Res., 22 , 141146.

    • Search Google Scholar
    • Export Citation
  • Curtis, S., , and D. W. Gamble, 2008: Regional variations of the Caribbean mid summer drought. Theor. Appl. Climatol., 94 , 2534.

  • Erasmi, S., , P. Propastin, , M. Kappas, , and O. Panferov, 2009: Spatial patterns of NDVI variation over Indonesia and their relationship to ENSO warm events during the period 1982–2006. J. Climate, 22 , 66126623.

    • Search Google Scholar
    • Export Citation
  • Gamble, D. W., , and S. Curtis, 2008: Caribbean precipitation: Review, model, and prospect. Prog. Phys. Geogr., 32 , 265276.

  • Gamble, D. W., , D. Parnell, , and S. Curtis, 2007: Spatial variability of the Caribbean mid-summer drought and relation to North Atlantic high circulation. Int. J. Climatol., 28 , 343350.

    • Search Google Scholar
    • Export Citation
  • Gamble, D. W., , D. Campbell, , T. L. Allen, , D. Barker, , S. Curtis, , D. F. M. McGregor, , and E. J. Popke, 2010: Climate change, drought, and Jamaican agriculture: Local knowledge and the climate record. Ann. Assoc. Amer. Geogr., in press.

    • Search Google Scholar
    • Export Citation
  • Giannini, A., , M. Cane, , and Y. Kushnir, 2001: Interdecadal changes in the ENSO teleconnections to the Caribbean region and the North Atlantic Oscillation. J. Climate, 14 , 28672879.

    • Search Google Scholar
    • Export Citation
  • Hastenrath, S., 1966: The flux of atmospheric water vapor over the Caribbean Sea and the Gulf of Mexico. J. Appl. Meteor., 5 , 778788.

    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., and Coauthors, 2007: The TRMM Multisatellite Precipitation Analysis (TMPA): Quasi-global, multiyear, combined-sensor precipitation estimates at fine scales. J. Hydrometeor., 8 , 3855.

    • Search Google Scholar
    • Export Citation
  • Inoue, M., , I. Handoh, , and G. Bigg, 2002: Bimodal distribution of tropical cyclogenesis in the Caribbean: Characteristics and environmental factors. J. Climate, 15 , 28972905.

    • Search Google Scholar
    • Export Citation
  • Kogan, F., 2001: Operational space technology for global vegetation assessment. Bull. Amer. Meteor. Soc., 82 , 19491964.

  • Kogan, F., 2002: World droughts in the new millenium from AVHRR-based vegetation health indices. Eos, Trans. Amer. Geophys. Union, 83 , 557564.

    • Search Google Scholar
    • Export Citation
  • Li, Z., , and M. Kafatos, 2000: Interannual variability of vegetation in the United States and its relation to El Niño/Southern Oscillation. Remote Sens. Environ., 71 , 239247.

    • Search Google Scholar
    • Export Citation
  • Magaña, V., , J. Amador, , and S. Medina, 1999: The midsummer drought over Mexico and Central America. J. Climate, 12 , 15771588.

  • Mapes, B. E., , P. Liu, , and N. Buenning, 2005: Indian monsoon onset and the Americas midsummer drought: Out-of-equilibrium responses to smooth seasonal forcing. J. Climate, 18 , 11091115.

    • Search Google Scholar
    • Export Citation
  • McGregor, D. F. M., , D. Barker, , and D. Campbell, 2009: Environmental change and Caribbean food security: Recent hazard impacts and domestic food production in Jamaica. Global Change and Caribbean Vulnerability: Environment, Economy, and Society at Risk? D. F. M. McGregor and D. Barker, Eds., University of West Indies Press, 197–217.

    • Search Google Scholar
    • Export Citation
  • Muñoz, E., , A. J. Busalacchi, , S. Nigam, , and A. Ruiz-Barradas, 2008: Winter and summer structure of the Caribbean low-level jet. J. Climate, 21 , 12601276.

    • Search Google Scholar
    • Export Citation
  • Peters, A., , E. Walter-Shea, , L. Ji, , A. Vina, , M. Hayes, , and M. Svoboda, 2002: Drought monitoring with NDVI-based standardized vegetation index. Photogramm. Eng. Remote Sens., 68 , 7175.

    • Search Google Scholar
    • Export Citation
  • Poveda, G., , A. Jaramillo, , M. M. Gil, , N. Quiceno, , and R. I. Mantilla, 2001: Seasonality in ENSO-related precipitation, river discharges, soil moisture, and vegetation index in Columbia. Water Resour. Res., 37 , 21692178.

    • Search Google Scholar
    • Export Citation
  • Rouse, J. W., , R. H. Haas Jr., , J. A. Schell, , and D. W. Deering, 1974: Monitoring vegetation systems in the Great Plains with ERTS. Third ERTS-1 Symposium, S. C. Freden, E. P. Mercanti, and M. A. Becker, Eds., Vol. 1, Special Publ. NASA SP-351, NASA, 309–317.

    • Search Google Scholar
    • Export Citation
  • Small, R. J. O., , S. P. de Szoeke, , and S-P. Xie, 2007: The Central American mid-summer drought: Regional aspects and large-scale forcing. J. Climate, 20 , 48534873.

    • Search Google Scholar
    • Export Citation
  • Wang, C. Z., 2007: Variability of the Caribbean low-level jet and its relation to climate. Climate Dyn., 29 , 411422.

  • Wang, C. Z., , S. K. Lee, , and D. B. Enfield, 2008: Climate response to anomalously large and small Atlantic warm pools during the summer. J. Climate, 21 , 24372450.

    • Search Google Scholar
    • Export Citation
  • Wang, J., , P. M. Rich, , and K. P. Price, 2003: Temporal responses of NDVI to precipitation and temperature in the central Great Plains, USA. Int. J. Remote Sens., 24 , 23452364.

    • Search Google Scholar
    • Export Citation
  • Yang, W., , L. Yang, , and J. W. Merchant, 1997: An assessment of AVHRR/NDVI ecoclimatical relations in Nebraska, USA. Int. J. Remote Sens., 18 , 21612180.

    • Search Google Scholar
    • Export Citation
Fig. 1.
Fig. 1.

Map of study area in Jamaica with St. Elizabeth parish (shaded in gray) and related spatial coverage of TRMM rainfall estimates (solid line) and Terra MODIS NDVI values (hatched area).

Citation: Journal of Applied Meteorology and Climatology 49, 7; 10.1175/2010JAMC2422.1

Fig. 2.
Fig. 2.

Annual climatological average values of NDVI (solid line) and precipitation estimates (bar graphs; mm day−1) for the study areas in the 2001–07 period. Annual data from 2005 are omitted because of the nonpresence of the midsummer drought.

Citation: Journal of Applied Meteorology and Climatology 49, 7; 10.1175/2010JAMC2422.1

Fig. 3.
Fig. 3.

Mean rainfall (bar graph; mm day−1) and NDVI (line plot) for the study area for each year from 2001 to 2007.

Citation: Journal of Applied Meteorology and Climatology 49, 7; 10.1175/2010JAMC2422.1

Fig. 4.
Fig. 4.

Percentage change in NDVI values across Jamaica with the St. Elizabeth parish outlined in black. Percent change is computed by dividing the difference of the 10–25 Jun 2001–07 (excluding 2005) average NDVI and 28 Jul–12 Aug 2001–07 (excluding 2005) NDVI by the 10–25 Jun NDVI (see Fig. 2).

Citation: Journal of Applied Meteorology and Climatology 49, 7; 10.1175/2010JAMC2422.1

Table 1.

Time lag in 16-day granules between maxima and minima in precipitation and NDVI for the months of April–November from 2001 to 2007. Average condition is reported in the last row. Negative values indicate that precipitation lags NDVI.

Table 1.
Table 2.

Annual percent change between post-early-season rainfall NDVI maximum and post-MSD NDVI minimum for 2001–07.

Table 2.
Table 3.

Correlations between NDVI and rainfall from concurrent to a two-granule (32 day) lag during April–November for each year. Years with asterisks show a significant relationship (at 0.1 level) at the 90% confidence interval.

Table 3.
Save