RESUMO
Lakes in arid and semi-arid regions are experiencing dramatic variations in water level and volume, which has caused severe ecological and social problems. Long-term study of the lake dynamics in arid/semi-arid regions could provide particular insights into the mechanisms driving lake variations, while hydro-meteorological data were usually limited in these regions, especially before the instrumental period. In the present study, we focused on a typical great lake - Hulun Lake in semi-arid region in northern China, simulated the hydrological processes from 1904 to 2016 using SWAT model, CRUNCEP7 reanalysis data, and sparse records of lake level during 1900s-1950s, and investigated the mechanisms driving the dramatic variations of the lake at the hundred-year time scale. Results illustrated that the simplified Penman equation by Valiantzas (2006) could reproduce the evaporation dynamics of Hulun Lake, with monthly R2 being 0.93-0.95. The long-term simulation since 1904 reproduced runoff dynamics, which were consistent with the dramatic variations of lake level over hundred years. The largest water level increase (~5.0 m in 1950s) and decrease (~4.5 m in 2000s) during 1904-2016 were jointly affected by river runoff, lake evaporation, and precipitation into the lake. Both the positive/negative phase and the multi-decadal trend of PDO clearly influenced the hydrological cycle of Hunlun Lake, especially for the period of 1904-1950 with low lake levels. Overall, the present study provided a methodology for investigating the hundred-year hydrological processes for lakes in semi-arid regions in northeastern Asia.
RESUMO
Observations of water levels in coastal aquifers and corresponding tides coupled with meteorological variances near the Ariake Sea show that groundwater in this area mainly fluctuates with atmospheric and tidal variations. Tidal effects occur with semi-monthly, diurnal, or semi-diurnal periodicity, whereas the barometric influences commonly act in the low-frequency domain. Tidal and barometric effects in water levels are separable using wavelet techniques and can be evaluated statistically. Results show the following. (1) The tidal coefficients are 0.002-0.154, attenuating roughly exponentially from the seashore. The time lags in water levels increase linearly approximately with increasing inshore distance. Relations between tidal coefficients and time lags and the inshore distance indicate higher hydraulic diffusivity in the south aquifer, which was confirmed by the hydraulic property calibrations in analytical simulations. (2) Water levels related to meteorological phenomena fluctuate inversely according to barometric loading variation with time lags of 2-3 h. The effective barometric efficiencies are 0.022-0.12. Lower barometric influences were found in the south aquifer.