Geophysical Institute of Peru1, NASA Goddard, Hydrological Sciences Laboratory, USA2, Geophysical Institute of Peru3, National Meteorological and Hydrological Service - Peru4
Precipitation is a major component of the water cycle and key input for hydrological modeling applications. In developing countries, ground-based measurement networks (meteorological and hydrological) can be scarce or even nonexistent. However, in order to analyze extreme events such as droughts or floods by means hydrological models, reliable quantification of spatio-temporal precipitation in real or quasi real time is fundamental. Nonetheless, whether spatio-temporal rainfall variability in tropical regions is poorly represented, results from rainfall--runoff models are often unsatisfactory. Also, in case of the Western Amazon, the presence of the Andes contributes to high spatial rainfall variability (Espinoza et al., 2009). This problem is further aggravated by the very sparse rain gauging networks. Rain gauging density is low in the Amazon River basin, resulting in uncertain estimates of spatially distributed rates of precipitation reaching the land surface. This may cause difficulties while studying hydrological processes, hydroclimatic variability, biogeochemical analysis, drainage basin response and hydrological modeling in the basin (e.g., Collischonn et al., 2008; Paiva et al., 2011; Getirana et al., 2012). The limitations in estimating spatially-distributed precipitation restrict the management of water resources and hamper early flood warning systems resulting in massive socioeconomic damages (Behrangi et al., 2011). In the last decades, the proliferation of satellite-based precipitation products has tremendously improved the ability to estimate rainfall over much of the globe (Huffman et al., 2010), becoming a complementary alternative for hydrometeorological applications and climate studies emerging, in areas where data are missing. The main objective of this research is to assess the capability of three satellite-based rainfall datasets (TMPA, CMORPH, PERSIANN) in accurately representing precipitation fields and their impacts on the hydrological cycle over the Western Amazon basin. A ground-based precipitation dataset (HYBAM) was also considered allowing the evaluation of satellite estimates. Evaluating satellite-based precipitation datasets is particularly relevant in the Western Amazon, where hydrological modeling and hydrological forecast is yet poorly developed. Indeed, climate variability has severely impacted the Western Amazon during the last decades, for instance, significant rainfall diminution has been reported, particularly during dry seasons (Espinoza et al., 2011; Lavado et al., 2012). Moreover, this region has suffered severe extreme hydrological events, such as the intense droughts in 2005 and 2010 and floods in 2009 and 2012 (Espinoza et al., 2011; 2012; 2013). The evaluation is performed through the use of the MGB-IPH hydrological model (Collischonn et al., 2007), where such datasets are used as inputs. Simulated streamflows are compared against observations in order to determine better suited satellite-based precipitation estimates. The Amazon basin of Peru and Ecuador (PEAAB) is located in the Western part of the Amazon and covers parts of both the Northern and Southern hemispheres. It occupies 14% of the Amazon basin, with a drainage area of around 878,300 km2 and a mean discharge of ~35,500 m3/s, considering the 2003-2009 period. The hydrological modeling is performed using seven years of daily data, from 2003 to 2009 and observed streamflows at 18 gauging stations (HYBAM). Results using HYBAM ground-based dataset in this research, shows better results in model performance, this is mainly when drainage area is greater than 100,000 km2 (Nash Sutcliffe coefficient NS >0.64). This indicates that rainfall observed data are even more reliable source than estimates satellite (TMPA, CMORPH, PERSIANN). However, it is likely that remote sensing of rainfall can improve in the near future. In addition, is very important increase ground based measurements, in order to improve validation mechanisms estimated rainfall. Despite of this, results presented in this research also suggest that TMPA is better suited to simulate the hydrological processes over the Peruvian tropical basins. Simulated streamflows using TMPA provided better results in most basins located within the study area (NS 0.64-0.83), suggesting that its precipitation estimates are more reliable than those provided by CMORPH and PERSIANN. This can be a helpful alternative source of data for rainfall--runoff simulation, in areas where conventional rainfall data lacks. However, the performance analysis over the equatorial Peruvian Ecuadorian basins indicates that satellite precipitation (TMPA, CMORPH and PERSIANN) products not agree well with observed data, probably due to the lack of adequate rainfall estimates because it is consistent with estimated streamflows, In addition, the three satellite-based precipitation datasets (TMPA, CMORPH and PERSIANN) were compared with against a ground-based product (HYBAM) over the Andean and Amazon regions. The results indicate an overestimation mainly in north of the PEAAB when is compared with TMPA. This result corroborates the findings by Zulkafli et al. (2014) for this regiÃ³n. But, there is an important understimation in the total amount of precipitation, mainly in northern Amazonian and Andes when CMORPH and PERSIANN are analyzed. Besides, this work has shown that these underestimations occur mainly during dry season. Despite TMPA is better adjusted to HYBAM observed data, cannot be considered completely reliable, because seasonal variability of rain is not well represented in the MaraÃ±Ã³n basin against the Ucayali basin where averaged over PEAAB are very similar to HYBAM. This suggests that rainfall estimates still differ from groun-based measurements and depends of rainfall seasonal regime to represent. Generally, all of this has allowed knowing the potential for use in operational hydrology of three satellite-based rainfall datasets. 1. Behrangi, A. Khakbaz, B., Chun Jaw, T., AghaKouchak, A., Hsu, K., (2011). Hydrologic evaluation of satellite precipitation products over a mid-size basin. J. Hydrol. 397, 225Â–237. 2. Collischonn, W., Allasia, D.G., Silva, B.C., Tucci, C.E.M. (2007). The MGB-IPH model for large-scale rainfall-runoff modeling. J. Hydrol. Sci. 52, 878Â–895. 3. Collischonn, B., W. Collischonn, and C. E. M. Tucci. (2008). Daily hydrological modeling in the Amazon basin using TRMM rainfall estimates, J. Hydrol., 360(1Â–4), 207Â–216, doi:10.1016/j.jhydrol.2008.07.032. 4. Espinoza JC., J. Ronchail, J.L. Guyot, Cocheneau G., N Filizola, W. Lavado, E. de Oliveira, R. Pombosa and P. Vauchel. (2009). Spatio Â– Temporal rainfall variability in the Amazon Basin Countries (Brazil, Peru, Bolivia, Colombia and Ecuador). International Journal of Climatology, 29, 1574-1594. 5. Espinoza, JC., Ronchail, J., Guyot J.L., Junquas, C., Vauchel, P., Lavado, W.S., Drapeau, G., Pombosa, R. (2011). Climate variability and extremes drought in the upper SolimÃµes River (Western Amazon Basin): Understanding the exceptional 2010 drought. Geophys. Res. Lett., 38, L13406, doi:10.1029/2011GL047862. 6. Espinoza, J.C., Ronchail, J. Guyot, J. L., Junquas, C., Drapeau, G., Martinez, J.M., Santini, W., Vauchel, P., Lavado, W., OrdoÃ±ez, J., and Espinoza, R., (2012). From drought to flooding: understanding the abrupt 2010Â–11 hydrological annual cycle in the Amazonas River and tributaries, Environ. Res. Lett. 7 024008 doi:10.1088/1748-9326/7/2/024008. 7. Espinoza, JC., Ronchail, J., Frappart, F., Lavado, W., Santini, W., Guyot, JL. (2013). The major floods in the Amazonas River and tributaries (Western Amazon basin) during the 1970 Â– 2012 period: A focus on the 2012 flood. Journal of Hydrometeorology, 14, 1000-1008. doi: 10.1175/JHM-D-12-0100.1. 8. Getirana, A. C. V., Boone, A., Yamazaki, D., Decharme, B., Papa, F. and Mognard, N. (2012). The Hydrological Modeling and Analysis Platform (HyMAP). Evaluation in the Amazon basin, J. Hydrometeorol., 13, 1641Â–1665. 9. Huffman, G.J., Adler, R.F., Bolvin, D.T., Nelkin, E.J. (2010). The TRMM Multi-Satellite Precipitation Analysis (TMPA), Satellite Rainfall Applications for Surface Hydrology, DOI:10.100/978-90-481-2915-7_1. 10. Lavado C., W.S., Ronchail, J., Labat, D., Espinoza, J.C. and Guyot, J.L. (2012). Basin-scale analysis of rainfall and runoff in Peru (1969Â–2004): Pacific, Titicaca and Amazonas watersheds. Hydrological Sciences Journal, 57 (4), 625Â–642. 11. Paiva, R. C. D., D. Costa Buarque, R. T. Clarke, W. Collischonn, and D. G. Allasia. (2011). Reduced precipitation over large water bodies in the Brazilian Amazon shown from TRMM data, Geophys. Res. Lett.,38, L04406, doi:10.1029/2010GL045277. 12. Zulkafli, Z., Buytaert, W., Onof C, Manz, B., Tarnavsky, E., Lavado, W., Guyot, J.L. (2014). A Comparative Performance Analysis of TRMM 3B42 (TMPA) Versions 6 and 7 for Hydrological Applications over AndeanÂ–Amazon River Basins. J. Hydrometeor, 15, 581Â–592.