Artan dünya nüfusu ve küresel ısınmanın oluşturduğu kuraklığın etkisi ile su gibi ikamesi olmayan bir kaynağın, canlıların ihtiyaçlarını karşılayacak oranda muhafaza edebilmek, dünya kamuoyunun üzerinde durduğu önemli konular arasında yer almaktadır. Akarsulardaki su seviyesinin düşük olduğu dönemlerde gerekli olan su ihtiyaçlarını karşılamak ve akarsuların oluşturmuş olduğu düzensizliği ortadan kaldırılması için biriktirme haznelerinin yapılması gerekmektedir. Bu çalışmamızda Yalova Gökçe Barajının, su seviyesinin oluşmasında etkili olan Sellimandıra Deresinin gelen akım, yağış, buharlaşma, su tahliyesi, ( içme ve kullanma suyu, sulama suyu, dolu savak tahliyesi ) sızıntı suyu ve barajın rezervuar yüzey su seviyesi ölçüm verilerini kullanarak, rezervuardaki mevcut suyun seviyesinin tahmin edilmesi amaçlanmıştır. Son zamanlarda Yapay Sinir Ağlarının birçok uygulama alanında başarı ile kullanılmasından dolayı bu çalışmamızda da yapay sinir ağları modelleri kullanılarak Yalova Gökçe Barajının 2019 yılı su seviyesinin tahmini yapılmıştır. Yapay sinir ağları modellerinin analizinde, çok katmanlı yapay sinir ağı fonksiyonu olan Levenberg-Marquardt eğitim algoritması ile Gradient Descent with Momentum eğitim algoritması kullanılmıştır. Bu yapay sinir ağı modellerinin analizlerinden çıkan sonuçlar karşılaştırılmış olup Levenberg-Marquardt eğitim algoritması ile bulunmuş olan tahmin verilerinin gerçek su seviyesine daha yakın değerlerde olduğu tespit edilmiştir.

Making estimation of river flow with hydrological and methodological data of the past period and using in water resources project studies have been used for a long time in many studies in our country and in the world. In addition to this, along with frequent drought problems in recent years, it is also important to store and use the water in the reservoirs correctly. This study aims to research how two separate Artificial-Neural-Network-Functions, which are generated based on the study of neural networks in the human brain, can be used in areas with a certain reservoir capacity such as dams or ponds and to provide an example of the most appropriate ANN model for predicting the level changes that will occur depending on the next years. In this context, two separate functions have been evaluated. These are the Levenberg-Marquardt-(LM) training function and Gradient-Descent-with-Momentum(GDM) training functions. Here, instead of the best results obtained from the models, the best three model outputs were evaluated and the models were considered from a wide frame. Gökçe dam basin in Marmara region was selected as the Model Study Area. For the period of December 31, 2019 - January 1, 2000, the monthly average data prepared with daily data from DSI and different architectural ANN training models were used to forecast the monthly basin capacities for 2019. It can be stated that LM Training models created in terms of Basin reservoir level estimates give results closer to the measured value.

Incomplete station meteorological records constrain sustainable water-demand planning and climate-adaptation decisions in data-limited regions. Open-data reference evapotranspiration (ETo) products can provide continuous hydroclimatic information for atmospheric-demand screening, but they do not directly represent realized water demand and their use as station substitutes may introduce bias because gridded and station-derived ETo represent related but non-identical calculation and spatial domains. This study develops a station-anchored open-data ETo screening workflow for assessing local product compatibility under incomplete meteorological records. The workflow combines four official Turkish State Meteorological Service (MGM) stations in the eastern Marmara region, station-derived Hargreaves–Samani (HS) ETo for 1990–2020, TerraClimate v1.1 reference ETo, calendar-month and anomaly diagnostics, blocked historical gap-transfer tests in which omitted periods are excluded from correction fitting and tuning, and measured-radiation FAO-56 Penman–Monteith (PM) sensitivity subsets. TerraClimate reproduced the first-order annual ETo cycle strongly (r = 0.972–0.978), but mean TerraClimate-minus-HS bias remained station-dependent (+1.78 to +11.55 mm month⁻¹) and strongly calendar-month-dependent. After removing each series’ calendar-month climatology, Pearson correlation decreased to 0.559–0.849, indicating that the common annual cycle explains a substantial part of the raw agreement. Train-only monthly-bias and nested L2 residual corrections reduced pooled blocked out-of-sample MAE, although no correction method was universally superior across stations. Measured-radiation PM subsets at Kocaeli and Yalova showed PM-minus-HS mean offsets of −4.42% and +3.37%, respectively. The blocked tests evaluate transfer to omitted historical periods rather than prospective forecasting. The results support local, season-aware compatibility screening before gridded ETo is used for climate-resilient water-resource planning; they should not be interpreted as validation of actual ET, crop water use, irrigation withdrawals, or realized water demand.

Reservoirs can alter river responses to climatic water deficits, but distinguishing regulation-associated differences from hydroclimatic variability remains difficult. We use a large-sample matched-basin design to compare full-record meteorological–soil-moisture and meteorological–streamflow coupling between highly regulated basins and hydroclimatically similar near-natural references, with event-based drought propagation examined separately. GRDC-Caravan hydroclimatic records were combined with Global Dam Watch reservoir information, with reservoir exposure represented by a static, basin-level storage-to-discharge proxy. The main analysis retained 394 matched basin pairs for 2002–2018 (17 calendar years) with at least 80% streamflow completeness. The primary metric was zero-lag Pearson coupling computed from pairwise-complete monthly standardized anomaly series across the full analysis period; decoupling was defined as 1 − r₀. Event-based metrics were evaluated separately in eligible pairs. Streamflow showed a modest regulation-associated shift: the median regulated-minus-reference decoupling difference was 0.023 (95% bootstrap CI [0.008, 0.037]), with 57.4% of pairs positive. Soil moisture showed an opposite-sign shift (median ΔDc = −0.017), indicating a compartment contrast rather than a common positive decoupling response. The streamflow signal remained positive under stricter completeness filtering, an extended analysis period, exclusion of the poorest matches, and without-replacement matching, with a complementary sign-flip sensitivity check. Country- and GRDC-subregion cluster bootstraps retained positive intervals, whereas coarse WMO-region clustering and the separate 1991–2020 eligible-subset sensitivity had intervals including zero. Storage-to-discharge intensity was positively but modestly associated with streamflow decoupling (Spearman ρ = 0.241). These results indicate that reservoir regulation is associated with weaker contemporaneous meteorological–streamflow coupling in the sampled basins and support reservoir-aware interpretation of drought signals, while residual geographic and other human water-use confounding, together with unobserved reservoir operation, preclude strict causal attribution.

Monthly precipitation distributions provide screening-level hydroclimatic information relevant to regional water-resource assessment that is not captured by annual or seasonal means alone. This study evaluates observed and projected shifts in monthly precipitation distributions across the nine NOAA Climate Regions of the contiguous United States using PRISM observations and NEX-GDDP-CMIP6 projections. Observed changes were assessed using PRISM monthly precipitation for 1941–2020 by comparing 1941–1980 with 1981–2020, with sensitivity tests using 1981–2014 and an alternative bootstrap block length. Future changes were evaluated using five NEX-GDDP-CMIP6 models under four SSP scenarios, relative to each model’s 1981–2014 historical reference. To strengthen the statistical and water-resource interpretation, the analysis additionally evaluates historical CMIP6 performance against PRISM, tests explicit hypotheses using permutation/bootstrap procedures with Benjamini–Hochberg false-discovery-rate control, and introduces a categorical Hydroclimatic Screening Matrix. The selected models reproduced the regional monthly climatological cycle well, with monthly climatology correlations of 0.943–0.969 and monthly climatology Kling–Gupta Efficiency values of 0.834–0.886, although raw monthly time-series skill was lower. Observed PRISM results show robust mean increases in the Northeast, Upper Midwest, and Ohio Valley, while monthly Q95 increased robustly in the Northeast and South. Future projections indicate broad late-century increases in monthly mean and upper-tail precipitation, especially under SSP5-8.5. Statistical testing showed that late-century SSP5-8.5 projected changes fell outside the empirical PRISM historical-change envelope in 14 of 18 region-metric pairs, whereas upper-tail amplification was not FDR-significant across regions. The Hydroclimatic Screening Matrix identifies aligned planning-relevant signals, emerging upper-tail concerns, seasonal-storage follow-up priorities, and directional-uncertainty cases. These results support precipitation-based screening for identifying where detailed hydrologic impact modeling may be warranted, without treating monthly precipitation metrics as direct runoff, recharge, reservoir-operation, or flood-risk estimates. The observational endpoint of 2020 was selected to form two equal 40-year periods, thereby keeping the sample lengths balanced for distributional comparisons.