The study gives an estimate of the effect of four methods most often used in hydrological models to calculate the potential evapotranspiration—Penman–Monteith, Priestley–Taylor, Oudin, and Hargreaves, which differ in complexity and meteorological data requirements—on the efficiency and the results of simulating the dynamics of water balance components. The methodological base of the hydrological simulation was the widely known HBV conceptual model, in which the actual evapotranspiration is associated with the current state of a storage, which simulates the dynamics of soil moisture content. The objects of the study were 18 catchments with areas from 2.4 to 755 km2, located within the former Primorskaya Water-Balance station, the data of which were used in calculations of potential evapotranspiration and simulations of runoff. The comparison of simulation results with observations showed that the model values of evapotranspiration at the use of physically sound methods are in better agreement with the Budyko’s procedure of determination of potential evapotranspiration. The analysis of model sensitivity to the input potential evaportranspiration showed that the use of physically based methods to calculate the potential evapotranspiration has no general effect on the efficiency of runoff hydrograph simulation of the studied objects. At the same time, in the case when empirical models of evapotranspiration are used, the uncertainty in the results of calculation of potential evapotranspiration should be compensated for by model parameters that are less physically substantiated, which, in turn, can lead to distortion of the dynamics of water balance components and runoff genetic components.

Numerical experiments have been carried out to model hydrodynamic and morphodynamic processes to explain the nature of the spatial distribution of Zostera marina (eelgrass) in the inner bays of Posyet Bay. Hydrodynamic modeling has been performed using the Delft3D flow model. The Simulating WAves Near shore (SWAN) spectral wave model has been used to simulate wind wave parameters (direction of propagation, wave length and height, and near-bottom orbital velocities). The restructuring of the bottom topography under the influence of wind waves and the hydrodynamic influence of tidal and wind currents has been calculated with the Delft3D software package. The results of numerical experiments showed that the dynamics of the coastal bottom topography is determined by the nature of sediment motion in the coastal zone under the combined influence of waves and currents. A comparison of the modeling results with data on the distribution of Zostera showed a high correlation with model simulations of erosion and accumulation in the coastal zone.

In order to explain the spatial nature of the distribution of eelgrass in the inner bays of Posyet Bay, numerical experiments were performed to simulate hydrodynamics and morphodynamic processes. The hydrodynamics was modeled using the Delft3D Flow model. The spectral wave model SWAN (Simulating WAves Near shore) was used to calculate the wind wave parameters (propagation direction, wave length and height, near-bottom orbital velocities). Alteration of coastal zone profile under the impact waves and the hydrodynamic of tidal and wind currents was performed using the Delft3-MOR morphodynamics module included in the Delft3D software package. The results of numerical experiments have shown that the dynamics of the coastal bottom topography is determined by sediment transport forced by combined action of waves and currents. Comparison of the modeling results with information on the distribution of Zostera of previous studies and field surveys data showed a high agreement with model data on the spatial distribution of erosion and accumulation zones in the coastal zone.

Experience of using meteorological observations and the ERA5 reanalysis for runoff modeling using the GR4J conceptual model is outlined. The study objects are catchments within the Ussuri River basin (Kirovskii, the Russian Far East). The results of the comparison of ground-based observations and reanalysis data are presented. The hydrological model has been calibrated and verified on the basis of various data sources. The traditional scores NSE, logNSE, and BIAS have been used to evaluate the modeling efficiency. According to the scores, the modeling efficiency is generally "satisfactory" and better. It is shown that for simulations, it is better to use observation network data in case of floods and the reanalysis data in case of spring high water and low flow periods. It is concluded that the effective resolution of the ERA5 data for daily precipitation and air temperature for hydrological modeling in the study area is 0.75^∘-1.0^∘ ( ∼ 90–120 km).

The effect of time aggregation (1, 3, 6, 12, and 24 h) of input data on the efficiency and parameter values of the conceptual HBV model was studied based on special field observations, carried out on four small experimental catchments (3.1–26.3 km2) in the basin of the Pravaya Sokolovka R. (Verkhneussuriiskii gage, upper Ussuri R., southern Primorskii Krai) in 2012–2022. It was shown that a decrease in the calculation step leads to a gradual decrease in the simulation quality, calculated with the use of Nash–Sutcliff coefficient. The model demonstrates high efficiency in calculating the hydrographs of high floods for all examined catchment areas, but does not reproduce diurnal variations of water discharges (0.1–1.0 mm layer) during low-water season. The model proportions of runoff genetic components, at which the system gradually passes to the predominance of some genetic component, do not depend on the calculation step and have relatively constant values. At the same time, an increase in the calculation time step also leads to a successive increase of the model water discharges, expressed in mm of layer thickness, by a factor of about 2. The regularities in the variations of the parameters in the block describing the formation and transformation of HBV, as well as maximal measured precipitation totals at changes in the time step for rains ≥30 mm/day can be well approximated by power relationships with exponents 0.3–0.6. It was supposed that for high-intensity rains, the dependence of maximal precipitation amounts on the aggregation time interval can have an effect on the values of parameters of HBV-type models.

The open-source low-parametric con cep tual hy dro log i cal model GR4J was used for run off sim u la tions of 17 nested catch ments of the Up per Ussuri River ba sin with ar eas rang ing from 133 to 24400 km2, in clud ing the Kirovskii out let. The data of stan dard hydro meteoro logi cal ob ser va tions of Primorye Administration for Hydrometeorology and Environmental Monitoring were used for the model calibration and verification. Runoff simulations were performed with a daily step. The criteria commonly used in hydrological community, namely, the Nash-Sutcliffe model efficiency coefficient (NSE), coefficient of determination (R2), and bias (BIAS) were used to as sess the mod el ing ef fi ciency. According to the specified criteria, modeling results are mostly in "satisfactory" or better categories. The base of the unit hy dro graph expectedly in creases with the catch ment area, but there were no other relationships found between the values of calibrated model parameters, the annual water content, and the main morphometric pa ram e ters of the catch ments. It was shown for the stud ied catch ments that the sta bi li za tion and the max ima of the mod el ing ef fi ciency scores are reached at the catch ment ar eas of 1200-1700 km2. The in flu ence of rep re sen ta tive ness of the weather sta tion net work on the mod el ing efficiency was demonstrated using complementary meteorological observations from experimental catchments in the Up per Ussuri ba sin for the sim u la tion of the cat a strophic flood in 2016 that was caused by typhoon Lionrock.

This study is focused on the comparison of streamflow composition simulated with three well-known rainfall–runoff (RR) models (ECOMAG, HBV, SWAT) against hydrograph decomposition evaluated with End-Member Mixing Analysis (EMMA). In situ observations at two small mountain testbed catchments located in the south of Pacific Russia are used. All applied RR models and EMMA analysis demonstrate that two neighboring catchments disagree significantly on the mutual dynamics of the runoff sources. The RR models' benchmark test is based on proximity to EMMA hydrograph composition. Different aggregation intervals (season, month, and pentad) were applied to find a reasonable generalization period ensuring the clarity of results. ECOMAG is most conformable to EMMA outcome; HBV reflects flood events well enough; SWAT exhibits distinctive behavior compared to the other models. It is shown that, along with standard efficiency criteria of simulated and observed runoff proximity, EMMA analysis might provide useful auxiliary information for the validation of modelling results.

The results of long-term field studies on small catchments in the upper reaches of the Ussuri River (Primorskii krai, Pacific Russia) are presented. By virtue of modern means of observation, a unique dataset was obtained to record effectively the complicated runoff formation process in small low-mountain river basins. Geochemical and hydrological modeling were used jointly to describe the catchment dynamics, and genetic components of the river flow were assessed to study more thoroughly the runoff processes and to evaluate the runoff modeling accuracy factors.

The efficiency of runoff simulation, the values of parameters, and the dynamics of the estimated runoff components were analyzed for two conceptual hydrological models GR4J and HBV for 17 watersheds in the Upper Ussuri River (Primorsky Krai, Russia) with areas from 138 to 24 400 km 2 . Both models demonstrate an increase in the simulation efficiency (based on NSE criterion) with an increase in the catchment area up to 1–2 thousand km 2 , after which they stabilize within the interval of 0.75–0.85 for the calibration period and 0.70–0.80 for the verification period. The estimates obtained for the HBV model were 5–10% higher than those for the GR4J model. Analysis of the measured and calculated annual runoff maximums over the warm season suggests the conclusion that GR4J model is on the average 5–6% more efficient than the HBV model in calculating the maximal values of rain flood discharges. At the same time, the obtained values of the relative error BIAS demonstrate a more accurate reproduction of the annual average runoff by the HBV model. The main distinctions determining the efficiency of simulation in the study region are as follows: the method of considering the precipitation height increments within altitude belts, the specific features of the calculation of model evapotranspiration, the method for calculating the outflow from conceptual runoff-forming storages in the GR4J and HBV models.

Soil loss rate, suspended load, and suspended sediment discharge were simulated for the drainage basin of the Komarovka R., located at the territory of the former Primorskaya Water Balance, Russian Far East. The study was based on an open-code SWAT model. The model was calibrated and verified using archive measurement data on sediment load and suspended sediment discharge at PWBS by data of Tsentral’nyi hydrological gage (157 km 2 ). The simulation efficiency of the suspended load and suspended sediment discharge corresponds to categories satisfactory and higher. On the average for the drainage basin, the rate of soil washout is 11–12 t/ha per year. The washout rate is maximal in the period of active snow melting. The mean monthly value for April–May is 75 thous. t. The mean annual washout is 171 thous. t, and the maximal is 800 thous. t/year. The mean annual runoff of suspended sediments in the outlet section of the drainage basin is 2.6 thous. t. The highest monthly values of sediment runoff correspond to the period of summer floods in August, when with average of 1 thous. and maximum of 7.7 thous. t. The obtained data on the heterogeneity of the spatial distribution of the simulated values of washout rate are in good agreement with the landscape and climatic features of runoff formation in the drainage river basin. The major errors in the estimation of sediment load are due to the simulation quality of the hydrological regime and dynamics of runoff components that form flood hydrographs.

In this study, the Delft3D-Flow hydrodynamic model was used to determine the characteristics of the main tidal waves and currents in the Penzhina Bay (Sea of Okhotsk, Russia). The initial and boundary conditions are set according to the TPXO9 model data, the salinity and water temperature for each model layer are set based on the reanalysis data of the HYCOM ocean general circulation model, with a time step of 3 h. E-uropean Center for Medium-Range Weather Forecasts ECMWF (European Center for Medium-Range Weather Forecasts) reanalysis ERA-5 was used as meteoforcing. Modeling was performed for the ice-free period from May to September 2005. Maps of cotidal lines, tidal ellipses, and reversibility coefficient for the main tide waves: semidiurnal wave M2 and diurnal wave K1 were constructed. The model was verified by comparing the published and calculated harmonic components for 9 sea level monitoring points located in the Penzhina Bay. Conformance evaluation generally shows a high degree of consistency between model and reference data. The values of the coefficient of determination R2 between the series formed by the model and published harmonic components are in the range of 0.96–0.99. According to the magnitude of relative errors, the simulation results are divided into two consistency categories – high (1.48–2.14%) and satisfactory (2.93–4.27%). Spatial patterns of distribution for the values of relative errors were not found. A certain inconsistency in the results is presumably due to the time discretization of the observation time support and the short sea level monitoring data series used to calculate the published harmonic components at the Penzhina Bay.

Based on data of special observations at the former Primorskaya water balance station, the ratio of surface, subsurface, and ground components of rain floods is considered. The widely known HBV conceptual hydrological model is used as a methodological basis of the study. The runoff simulation quality for all stations is assessed as good and satisfactory based on the common statistical metrics. It is shown that the dynamics of the model runoff components is similar to a priori knowledge about the functioning of small catchments. The range of runoff values at the outlet related to the change in the dominant runoff generation type is estimated. An impact of soil cover and relief on the separation of runoff components is noted. The details of the flood formation in the study region are discussed.

An analysis of the HBV hydrological model parameters values and a comparison of modeling efficiency have been made for small experimental mountain catchments located in the upper reaches of the Ussuri River (Primorsky krai, Russia). For each catchment, the model is calibrated for individual years and, after that, for the entire observing period. The sets of parameters are used in calculations based on independent data for periods not included in the interval of parameters optimization. A calibration of the model shows that an optimal set of the parameters values for one year can differ greatly from the parameters for another year within a single catchment. Appropriate values of the modeling efficiency for the entire observations period are provided with using sets of parameters calibrated for seasons with high floods caused by heavy rainfall. The periods of low and medium water flow are characterized by relatively high errors in calculating the water runoff, which decreases the overall efficiency and reliability of results in modeling the entire period of data availability. It is shown that, for the simulation of a high runoff by the HBV model, it is sufficient to have 2 or 3 years of instrumental observations, including measurements of one or two significant flood events. A formal transfer of model parameters from one small catchment to nearby similar objects is only possible for assessments of rainfall floods characteristics. In other cases, such an approach leads to an inappropriate structure of the model water balance components and causes errors in runoff calculations.

Based on the results of modeling, the probability of productivity exceedance of aquafarm cultivation of the giant oyster Crassostrea gigas and the Pacific mussel Mytilus trossulus were calculated for Voevoda Bay (Russkii Island, Peter the Great Bay, Sea of Japan). The fields of hydrodynamics and salinity were calculated using the Delft3D Flow model, taking into account the inflow of freshwater from the catchment area of the bay. The parameters of the production of organic matter and its redistribution over the water area were calculated using the DELWAQ model. The results of hydrodynamic and biogeochemical modeling were used as input data in the FARM management model to estimate the ensured values of the productivity of aquatic organisms. The maps of the spatial distribution of the ensured values of the model productivity of bottom and cage plantations were constructed. The possibilities of their use in planning mariculture activities are discussed.

The article analyzes the integral atmospheric water content by data of continuous GPS/GLONASS observations over 2017–2019 at 13 observation points of the GNSS network in the territory of the Primorski Krai. The procedure of IWV evaluation is based on the decomposition of the total zenith tropospheric delay of satellite signal into the hydrostatic and humidity components. The GNSS-estimates of IWV were verified using radiosonde data from two weather stations of the Primorski Department of Hydrometeorological Service. The correlation between the results of GNSS and radiosonde observations was 0.93–0.99. The obtained IWV estimates were compared with Global Forecast System data. The comparison involved 16 nodes of the model grid nearest to each sensor. It was shown that the correlation of GNSS estimates of IWV with GFS data at the moment of forecast is, on the average, >0.90; starting from the forecast advance time of 48 h, the coefficient of correlation decreases to 0.60. The coefficient of correlation with GFS is 0.85–0.97 in the warm season and ≤0.60 in the cold season. The analysis of the spatial distribution of the correlation coefficient showed that the measured IWV values are linearly related with the PWEA model values, which refer to the grid nodes that show lesser elevation difference with the GNSS point. The results of the study suggest the conclusion that the expenses for the acquisition and processing of GNSS data are minimal and the operation of grid points does not depend on weather conditions. This makes the results of GNSS sounding promising for use in regional atmospheric models.