Sub-hourly rainfall patterns by hyetograph type under distinct climate conditions in Northeast of Brazil: a comparative inference of their key properties

Abstract

The lack of process-based classification procedures may lead to unrealistic hyetograph design due to complex oscillation of rainfall depths when assimilated at high temporal resolutions. Four consecutive years of sub-hourly rainfall data were assimilated in three study areas (Guaraíra, GEB, São João do Cariri, CEB, and Aiuaba, AEB) under distinct climates (very hot semi-arid and tropical wet). This study aimed to define rainfall events (for Minimum Inter-event Time, MIT, and Minimum Rainfall Depth, MRD, equal to 30 min and 1.016 mm, respectively), classify their hyetograph types (rectangular, R, unimodal with left-skewed, UL, right-skewed, UR, and centred peaks, UC, bimodal, B, and shapeless, SL), and compare their key rainfall properties (frequency, duration, depth, rate and peak). A rain pulse aggregation process allowed for reshaping SL-events for six different time spans varying from 2 to 30 min. The results revealed that the coastal area held predominantly R-events (64% events and 49% rainfall depth), in western semi-arid prevailed UL-events (57% events and 63% rainfall depth), whereas in eastern semi-arid mostly were R-events (61% events and 30% rainfall depth) similar to coastal area. It is concluded that each cloud formation type had important effects on hyetograph properties, differentiating them even within the same climate.

The performance of the IMERG satellite-based product in identifying sub-daily rainfall events and their properties

“…Both the USLE and RUSLE equations are based on the so-called rainfall erosivity index, which is calculated with the maximum rainfall intensity during the 30-min period. The delineation of individual rain events is also the first step to generate the hyetographs for analysing the spatiotemporal variability of rainfall (Barbosa et al, 2018;Coutinho et al, 2014;Dolšak et al, 2016). Moreover, the development of robust Intensity-Duration-Frequency (IDF) curves for extreme rainfall studies, which are the key to many engineering designs, is sometimes interfered with by sparse, infrequent or short-record sub-daily observations in many locations (Courty et al, 2019;Lumbroso et al, 2011).…”

The big picture of field hydrology studies in Brazil

“…Monitored basins have been used to assess rainfall spatial homogeneity over small areas. Barbosa et al (2018) observed rainfall homogeneity at four rain gauges analysed in the Guaraíra experimental basin (ID-16, AOR hydrographic region). Marques et al (2017) detected a similar seasonal distribution of rainfall in an area of 2065 km 2 in the Piabanha River Basin (ID-54, ASD hydrographic region); however, differences from 100 to 1000 mm were found among three rain gauge measurements in the same year.…”

Stability of Soil Moisture Patterns Retrieved at Different Temporal Resolutions in a Tropical Watershed

“…The minimum SM values for Espodossolo Humilúvico (Carbic Podzol) and Argissolo Vermelho-Amarelo (Haplic Acrisol) soil type plots (Table 3) were higher than those obtained by Costa et al (2016) for Argissolo Eutrófico (Haplic Lixisol), Luvissolo Crômico (Chromic Luvisol) or Neossolo Litólico (Lithic Leptsol) in the same region. Most likely, this is because coastal watersheds are susceptible to a tropical regime with high rainfall rates in the rainy season, differently from the Caatinga biome over a very hot semi-arid zone (Barbosa et al, 2018;Rozante et al, 2018;Gadelha et al, 2019), where 70 % of the SM drop below the wilting point (Costa et al, 2016).…”

“…Mean annual rainfall is ~1,700 mm, most of which (~70 %) occurs from March to July, i.e. in autumn-winter seasons (Barbosa et al, 2018). Accordingly, mean annual class-A pan evaporation reaches ~1,300 mm in the GEW (Silva et al, 2000).…”