Compendium

Hydrology and the Global Water Cycle

23 min read

Hydrology and the Global Water Cycle

Hydrology is the geoscience of water on, above, and below Earth’s surface — the spatial distribution, temporal variability, flux pathways, and chemical-biological interactions of liquid, solid, and vapor H2O. The discipline straddles atmospheric science (precipitation generation), surface-water engineering (streamflow, reservoirs), and subsurface science (vadose-zone flow, aquifer dynamics). The water cycle is the most energetic mass-flux loop on the planet: roughly 1.4e9 km3 of total H2O is stored across reservoirs that exchange ~577 Tm3 yr-1 (teracubic metres per year, equal to 5.77e14 m3 yr-1) through evaporation, condensation, precipitation, and runoff. Because energy and water cycles are coupled (~50 % of insolation absorbed at the surface drives latent-heat flux), perturbations to climate forcings (greenhouse gas concentrations, aerosol burdens, land-cover change) propagate directly into hydrologic response — heavier rain, longer droughts, declining mountain snowpack, altered groundwater recharge. This note compiles the quantitative architecture of the global water budget, the principal precipitation-generating systems, the measurement infrastructure, the canonical rainfall-runoff and flood-frequency methods, the institutional frameworks that govern water resources, and the climate-change signal in hydrologic statistics.

1. Global water budget — stocks and fluxes

1.1 Stocks (volumetric inventory)

Total Earth water mass is approximately 1.386e9 km3 (Shiklomanov 1993; Trenberth-Smith-Qian-Dai-Fasullo 2007). Distribution by reservoir:

  • Oceans and seas: 1.338e9 km3 = 96.5 % of total water.
  • Ice and snow (glaciers, ice caps, permafrost ground ice): 2.41e7 km3 = 1.74 %. Antarctic ice sheet alone holds ~2.65e7 km3 ice volume (~58 m sea-level equivalent if melted, Fretwell BEDMAP2 2013); Greenland ice sheet 2.85e6 km3 (~7.4 m SLE); mountain glaciers 0.158e6 km3 (~0.32 m SLE, Farinotti 2019).
  • Groundwater: 2.34e7 km3 = 1.69 %. Of this, ~50 % is saline; freshwater groundwater is ~1.05e7 km3, of which ~30 % is “modern” groundwater (recharged within last 50 yr per Gleeson-Befus-Jasechko-Luijendijk-Cardenas 2016).
  • Surface freshwater (lakes, rivers, wetlands): 1.785e5 km3 = 0.013 %. Lakes hold 176 400 km3 (Lake Baikal alone ~23 600 km3, 22 % of unfrozen surface freshwater); rivers ~2 120 km3 at any instant.
  • Soil moisture: 16 500 km3.
  • Atmospheric water vapour: 12 900 km3 (mean ~25 mm column-integrated precipitable water; Trenberth-Fasullo-Mackaro 2011).
  • Biological water (plants, animals): 1 120 km3.
  • Permafrost ground ice (separate from glacial ice): ~3e5 km3.

1.2 Fluxes (annual exchange)

Trenberth-Fasullo-Mackaro 2011 budget (revised from Trenberth 1998 with reanalysis + satellite era data):

  • Precipitation over oceans: 373 Tm3 yr-1 (= 1027 mm yr-1 over 361.9e6 km2 ocean area).
  • Precipitation over land: 113 Tm3 yr-1 (758 mm yr-1 over 148.9e6 km2). Total global precipitation ~486 Tm3 yr-1 (505 Tm3 yr-1 in older Shiklomanov budgets used in textbooks).
  • Evapotranspiration from oceans: 413 Tm3 yr-1 (1141 mm yr-1).
  • Evapotranspiration from land: 73 Tm3 yr-1 (490 mm yr-1) — partitioned roughly 39 % transpiration, 32 % canopy interception evaporation, 20 % soil evaporation, 6 % open-water evaporation, 3 % snow sublimation (Schlesinger-Jasechko 2014; Good-Noone-Bowen 2015).
  • Net atmospheric moisture transport ocean-to-land: 40 Tm3 yr-1 (closure: 113 − 73 = 40).
  • River runoff land-to-ocean: 38 Tm3 yr-1 (~1.2e6 m3 s-1 mean global discharge; Dai-Trenberth 2002). Largest single river: Amazon ~6.6 Tm3 yr-1 (~210 000 m3 s-1), ~17 % of global runoff.
  • Groundwater discharge to ocean (submarine + coastal): ~2 Tm3 yr-1 (Moore 2010; uncertain).
  • Net glacial mass loss (currently): ~0.4 Tm3 yr-1 (Greenland + Antarctica + mountain glaciers, IMBIE 2020 + Hugonnet 2021).

The budget closes within ~5 % once endorheic basin closure and ice-sheet contributions are included.

1.3 Residence times

Residence time tau = stock / flux:

  • Atmosphere: tau ≈ 12 900 / 486 000 = 0.027 yr ≈ 9.7 d (rapid).
  • Soil moisture: ~1–2 mo.
  • Rivers: 2 120 / 38 000 ≈ 0.056 yr ≈ 20 d.
  • Lakes: ~17–100 yr depending on basin.
  • Shallow groundwater: ~100–10 000 yr.
  • Deep “fossil” groundwater (Nubian Sandstone, Ogallala recharge fraction): up to 10 000–1 000 000 yr (Sturchio 2004 36Cl dating).
  • Oceans: 1.338e9 / 486 000 ≈ 2 750 yr for full cycling.
  • Antarctic ice sheet: ~17 000 yr full turnover.

2. Atmospheric water — precipitation systems

2.1 Convective precipitation

Convective storms form where conditional instability (CAPE > 0) plus a trigger (orographic lift, mesoscale convergence, frontal forcing, diurnal heating) drives parcel ascent. Cumulonimbus updrafts reach 100 hPa (~16 km in tropics) with vertical velocities 10–50 m s-1; latent heat release ~334 kJ kg-1 for freezing plus 2 500 kJ kg-1 for condensation drives buoyancy.

  • Single-cell convection: ~30 min lifetime, ~10 km diameter, rainfall rates 10–100 mm h-1.
  • Multicell clusters: 1–4 h lifetimes, regenerating via gust-front lifting.
  • Supercells: rotating mesocyclones, 1–6 h, severe hail + tornadoes; Browning 1964 classic identification.
  • Mesoscale convective systems (MCS) and mesoscale convective complexes (MCC): Maddox 1980 defined MCC as cloud shield >50 000 km2 with -32 °C IR top for >6 h; account for ~50 % of warm-season precipitation in US Great Plains and Sahel.
  • Orographic convection: forced lifting over terrain; Smith 1979 mountain-wave theory; Roe 2005 orographic precipitation review.

2.2 Stratiform precipitation

Wide, gently rising air masses (5–50 cm s-1) along extratropical fronts or in tropical anvil outflow produce stratiform rain (1–10 mm h-1) with bright-band radar signature at the 0 °C isotherm (melting layer; Heymsfield-Bansemer-Schmitt 2015). Stratiform fraction of total rainfall: ~40 % global, ~70 % in midlatitudes, ~25 % tropics (Schumacher-Houze 2003 TRMM analysis).

2.3 Tropical circulation

  • Intertropical Convergence Zone (ITCZ): low-level trade-wind convergence zone, mean position ~5° N over annual cycle; migrates seasonally over land (5°–25° N in NH summer over Africa/India). Provides ~30 % of global precipitation. Schneider-Bischoff-Haug 2014 review on ITCZ controls (cross-equatorial energy transport).
  • Monsoons: thermally driven seasonal wind reversals. South Asian summer monsoon delivers ~80 % of India’s annual rainfall in JJAS (June–September); Indian Ocean Dipole and ENSO modulate strength. West African Monsoon, East Asian Monsoon, North American Monsoon (Arizona/Mexico), South American Monsoon. Webster-Magana-Palmer-Shukla-Tomas-Yanai-Yasunari 1998 classic review.
  • Walker circulation: zonal east-west tropical Pacific overturning, weakens with El Niño. ENSO (Wyrtki 1975) — warm phase shifts convection eastward, reducing W Pacific/Maritime Continent rainfall, enhancing E Pacific.

2.4 Extratropical cyclones

Baroclinic instability (Charney 1947; Eady 1949) along midlatitude jet drives cyclogenesis. Norwegian school (Bjerknes-Solberg 1922) cold-warm-occluded front model. Comma-cloud structure, stratiform rain ahead of warm front, narrow convective lines along cold front. Bomb cyclones: explosive deepening >24 hPa/24 h normalised to 60° latitude (Sanders-Gyakum 1980). North Atlantic and North Pacific storm tracks deliver winter precipitation to W Europe and W North America.

2.5 Atmospheric rivers (ARs)

Narrow corridors of intense horizontal water-vapour transport in lower troposphere, 400–500 km wide, >2 000 km long, integrated vapour transport (IVT) >250 kg m-1 s-1. Discovered by Newell-Newell-Zhu-Scott 1992 (“tropospheric rivers”); coined “atmospheric rivers” by Zhu-Newell 1998. Ralph et al. (CW3E Scripps) leading research group. AR Scale 1–5 by Ralph-Rutz-Cordeira-Dettinger-Anderson-Reynolds 2019 (BAMS):

  • AR 1: weak/primarily beneficial (IVT 250–500, <24 h).
  • AR 3: strong, mostly beneficial mostly hazardous.
  • AR 5: exceptional, primarily hazardous (IVT >1 250 kg m-1 s-1, >48 h).

Deliver 30–50 % of West Coast US annual precipitation in 5–10 events per winter; 80 % of major flood events (Ralph-Dettinger 2012). The “Pineapple Express” subset originates near Hawaii. California megafloods: 1861–62 ARkStorm scenario produced ~10 ft of standing water in Central Valley (Engstrom 1996; Porter 2011 USGS ARkStorm modelling). 2023 California AR sequence (9 ARs Dec 2022 – Jan 2023) caused $4.6 B damages and 22 fatalities (NOAA NCEI). Climate-change response: 25 % increase in AR frequency by end-century RCP 8.5 (Espinoza-Waliser 2018), with intensified extreme tails.

2.6 Tropical cyclones

Saffir-Simpson Hurricane Wind Scale (Simpson-Saffir 1971):

Category1-min sustained wind (m s-1)(mph)Storm-surge typical (m)
TS17–3239–73<1
133–4274–951.2–1.5
243–4996–1101.8–2.4
3 (major)50–58111–1292.7–3.7
4 (major)58–70130–1564.0–5.5
5 (major)>70>157>5.5

WMO basin classifications: hurricane (Atlantic, NE Pacific), typhoon (NW Pacific), severe tropical cyclone (SW Pacific, S Indian). Rapid intensification (RI): wind-speed increase >15.4 m s-1 (30 kt) in 24 h (Kaplan-DeMaria 2003); now more common (Bhatia-Vecchi-Knutson-Murakami-Kossin 2019).

3. Measurement infrastructure

3.1 Rain gauges

Tipping-bucket (0.2 mm or 0.01” resolution), weighing gauges (Geonor T-200B), optical disdrometers (Parsivel, OTT). Cooperative Observer Program (COOP, NWS, 1890+) ~8 700 US stations. WMO Global Telecommunication System (GTS) distributes synoptic precipitation. Gauge undercatch from wind 10–50 % for snow (Goodison-Louie-Yang 1998 WMO intercomparison).

3.2 Weather radar

  • NEXRAD WSR-88D (Weather Surveillance Radar 1988 Doppler): 159 S-band (2.7–3.0 GHz, ~10 cm wavelength) radars across US + territories, deployed 1988–1997. Dual-polarisation upgrade 2011–2013 added differential reflectivity (Zdr), specific differential phase (Kdp), and correlation coefficient for hydrometeor classification (Ryzhkov-Zrnic 2019 monograph).
  • Z–R relationship: Marshall-Palmer 1948 Z = 200 R^1.6 default; convective Z = 300 R^1.4 (NEXRAD WSR-88D Hydrologic Algorithm).
  • European OPERA radar mosaic; Japanese AMeDAS + JMA radar; Australia BoM network.

3.3 Satellite precipitation

  • TRMM (Tropical Rainfall Measuring Mission, NASA-JAXA 1997–2015): first spaceborne precipitation radar at Ku band (13.8 GHz), 35° N-S coverage, 4 km horizontal resolution.
  • GPM (Global Precipitation Measurement, NASA-JAXA 2014+): Core Observatory carries Dual-frequency Precipitation Radar (DPR — Ku 13.6 GHz + Ka 35.5 GHz) and GPM Microwave Imager (GMI, 13 channels 10–183 GHz). Global coverage 65° N-S. IMERG (Integrated Multi-satellitE Retrievals for GPM) merges constellation passive microwave + IR geostationary at 0.1° 30-min resolution (Huffman 2019).
  • Other satellite products: CMORPH (Joyce-Janowiak-Arkin-Xie 2004), PERSIANN (Sorooshian-Hsu 2000), MSWEP (Beck 2017 multi-source merge).
  • GRACE/GRACE-FO (Gravity Recovery and Climate Experiment): twin satellites measure terrestrial water storage anomalies via gravity changes, 300 km resolution monthly. Famiglietti 2014 + Rodell-Famiglietti-Wiese-Reager-Beaudoing-Landerer-Lo 2018 (“Emerging trends in global freshwater availability”) found 21 of 37 major aquifers in net depletion. Tapley 2019 GRACE 15-year retrospective.

4. Rainfall-runoff and surface hydrology

4.1 Infiltration

  • Horton 1933 infiltration capacity decay.
  • Green-Ampt 1911: piston-flow analytical solution with wetting-front capillary suction psi_f, hydraulic conductivity K_s, moisture deficit dtheta — f(t) = K_s [1 + (psi_f · dtheta)/F(t)].
  • Philip 1957 two-term series: f(t) = (1/2) S t^(-1/2) + K_s.
  • Richards 1931 equation: dtheta/dt = del · [K(theta) grad H], governing equation for unsaturated flow.

4.2 SCS Curve Number method

USDA SCS (now NRCS) National Engineering Handbook Section 4 (1956+, updated 1986). Direct runoff Q = (P − Ia)² / (P − Ia + S) when P > Ia, where S = (25 400 / CN) − 254 mm (CN dimensionless 30–100; Ia = 0.2 S initial abstraction). Widely used in design despite theoretical limitations.

4.3 Flow routing

  • Kinematic wave (Lighthill-Whitham 1955; for steep slopes, friction-dominated): dQ/dt + c dQ/dx = q_lateral, where c is celerity.
  • Diffusion wave: includes pressure term, captures backwater.
  • Dynamic wave (full Saint-Venant 1871 equations): continuity + momentum. HEC-RAS (US Army Corps Hydraulic Engineering Center River Analysis System) is industry-standard 1D + 2D unsteady solver.

4.4 Distributed hydrologic models

  • VIC (Variable Infiltration Capacity, Liang-Lettenmaier-Wood-Burges 1994).
  • WRF-Hydro / NOAA National Water Model (NWM, operational 2016+, 2.7 M reach NHDPlus stream network, hourly forecasts).
  • MIKE-SHE (DHI).
  • ParFlow (Maxwell-Kollet 2008 coupled subsurface-overland).
  • CLM-PRMS coupling for land-surface model integration.
  • JULES (Joint UK Land Environment Simulator, Best 2011).
  • WaterGAP (Döll-Kaspar-Lehner 2003) global hydrological model.

5. Flood frequency analysis

5.1 Series construction

  • Annual maximum series (AMS): largest peak each water year.
  • Partial duration series (PDS) / peaks-over-threshold (POT): all events above threshold; correlated POT events screened (Lang-Ouarda-Bobée 1999).

5.2 Distributions

  • Log-Pearson Type III (LPIII): US federal standard per Bulletin 17B (1982) and updated Bulletin 17C (England-Cohn-Faber-Stedinger-Thomas-Veilleux 2018). Three parameters fit to log-transformed flows.
  • Generalised Extreme Value (GEV, Jenkinson 1955): unifies Gumbel (xi=0), Fréchet (xi>0), Weibull (xi<0).
  • Generalised Pareto for POT (Pickands 1975).
  • L-moments estimation (Hosking 1990; Hosking-Wallis 1997 Regional Frequency Analysis monograph). L-moments more robust than conventional moments under heavy tails.

5.3 Regional frequency analysis

Pool data across hydrologically homogeneous regions (Dalrymple 1960 index-flood; Hosking-Wallis 1997). Reduces sampling variance for long return periods (100-yr, 500-yr). Climate-change nonstationarity invalidates classical assumption: Milly-Betancourt-Falkenmark-Hirsch-Kundzewicz-Lettenmaier-Stouffer 2008 (“Stationarity is dead”). Nonstationary models (Cheng-AghaKouchak 2014 GEV with time-varying location).

6. Drought metrics and droughts

  • PDSI (Palmer Drought Severity Index, Palmer 1965): soil-moisture water-balance with two-layer bucket, calibrated to climatology. PDSI < -3 severe; < -4 extreme. Self-calibrating PDSI (Wells-Goddard-Hayes 2004).
  • SPI (Standardised Precipitation Index, McKee-Doesken-Kleist 1993): fit precipitation aggregates (1-mo, 3-mo, … 48-mo) to gamma, transform to standard normal. SPI < -1 moderate drought; < -2 extreme. WMO recommended (Hayes 2011).
  • SPEI (Standardised Precipitation Evapotranspiration Index, Vicente-Serrano-Beguería-López-Moreno 2010): like SPI but uses P − PET (potential evapotranspiration), capturing warming-driven aridification.
  • Crop Moisture Index, Surface Water Supply Index, US Drought Monitor (Svoboda 2002 weekly synthesis).
  • Flash drought: rapid intensification, often <30 d (Otkin-Svoboda-Hunt-Ford-Anderson-Hain-Basara 2018 BAMS). 2012 US Central Plains and 2019 SE US examples.

Major recent droughts:

  • US Western megadrought 2000–2024 — Williams-Cook-Smerdon-Cook-Anchukaitis-Bolles-Cook-Coats-Marvel-Smith 2022 found driest 22-yr period in 1 200 yr; ~42 % attributable to anthropogenic warming.
  • Cape Town “Day Zero” 2017–18: 5–7× more likely with climate change (Otto-Wolski-Lehner-Tebaldi-vanOldenborgh-Hogesteeger-Singh-Holden-Fucik-vanGarderen-NewmanThakur 2018).
  • Chennai 2019: city reservoirs near-empty by June.
  • Horn of Africa 2020–23: five failed rainy seasons.
  • Western Europe 2022: 5–20× more likely under CC (Vicente-Serrano + WWA 2022).
  • Bangalore 2024: water crisis affecting 14 M residents.

7. Groundwater

7.1 Hydrogeology basics

Darcy’s law (Darcy 1856): q = − K grad h, where K is hydraulic conductivity (m s-1, ~1e-12 for clays to ~1e-2 for gravels). Transmissivity T = K · b for aquifer thickness b. Storativity S (confined): elastic + water expansion, ~1e-5 to 1e-3. Specific yield S_y (unconfined): drainable porosity, 0.01–0.3.

Theis 1935 transient well solution: drawdown s(r,t) = (Q / 4 pi T) W(u), where u = r² S / (4 T t) and W is the well function. Cooper-Jacob 1946 simplification for late-time. Hantush-Jacob 1955 leaky aquifer.

7.2 Major aquifers and depletion

GRACE-derived trends (Famiglietti 2014; Rodell 2018; Richey-Thomas-Lo-Famiglietti 2015):

  • Ogallala (High Plains Aquifer): 174 000 km3 storage, ~3 km3 yr-1 net depletion (mean), localised depletion >50 % since 1950 in TX panhandle, Kansas. Pre-development storage 4 000 km3; ~330 km3 removed by 2017 (McGuire 2017 USGS).
  • California Central Valley: 130 km3 lost 1962–2014 (Famiglietti-Lo-Ho-Bethune-Anderson-Syed-Swenson-deLinage-Rodell 2011); cumulative subsidence >8 m near Mendota (Galloway-Riley 1999, Faunt-Sneed-Traum-Brandt 2016).
  • North China Plain: 8 cm yr-1 water-level decline.
  • Northern India/Pakistan: Rodell-Velicogna-Famiglietti 2009 17 ± 4 km3 yr-1 loss.
  • Arabian Aquifer System (largest withdrawal stress globally per Richey 2015), Nubian Sandstone, Sahara.

7.3 Subsidence

Aquifer-system compaction from effective-stress increase. Mexico City sinks 35 cm yr-1 in worst zones; Jakarta 25 cm yr-1; San Joaquin Valley up to 30 cm yr-1 during droughts (InSAR Sentinel-1 monitoring; Smith-Bawden-Sneed 2017).

7.4 Regulation — SGMA

California Sustainable Groundwater Management Act 2014 (SB 1168, AB 1739, SB 1319) requires “high” and “medium” priority basins to form Groundwater Sustainability Agencies (GSAs), adopt Groundwater Sustainability Plans (GSPs), and achieve sustainability by 2040 (or 2042 for critically over-drafted basins). First major US groundwater regulation. Implementation contested; DWR rejected six SJV basin GSPs 2023.

US federal: no comprehensive groundwater statute; SDWA (Safe Drinking Water Act 1974) regulates wellhead protection; Class V UIC for injection.

8. Water resources management

8.1 Major reservoirs

  • Three Gorges (Yangtze, China, 2003): 39.3 km3 storage, 22.5 GW installed (largest hydropower).
  • Lake Mead (Colorado, Hoover Dam 1936): 35.2 km3 at 1 229 ft full pool; 2024 level ~1 060 ft, ~33 % full.
  • Lake Powell (Colorado, Glen Canyon Dam 1966): 33.3 km3 at 3 700 ft; 2024 ~33 %.
  • Bonneville Dam (Columbia 1938), Grand Coulee (Columbia 1942) — Columbia Basin Project.
  • Aswan High Dam (Nile, Egypt, 1970): 132 km3 Lake Nasser. Sediment trapping reduced Nile delta sediment supply 98 %, contributing to delta erosion and Mediterranean fisheries collapse (Stanley-Warne 1993).
  • Itaipu (Paraná, Brazil-Paraguay 1984): 14 GW.
  • Mosul (Tigris, Iraq), Atatürk (Euphrates, Türkiye) — transboundary tensions.

8.2 Withdrawals by sector

Global water withdrawals ~4 000 km3 yr-1 (FAO AQUASTAT 2020):

  • Irrigation/agriculture: 70 %.
  • Industrial: 19 %.
  • Municipal: 11 %.

Withdrawal vs consumption: irrigation consumption rate ~50 %, thermal-electric ~3 %, municipal ~15 %.

8.3 Irrigation

Surface: flood/furrow ~40 % efficient. Sprinkler 70–85 %. Drip/micro 85–95 % (Howell 2003). India + China + US + Pakistan top irrigators; ~330 Mha global irrigated (FAO 2020).

8.4 Water rights

Western US prior appropriation (“first in time, first in right”) vs riparian (eastern US, English common law). California has hybrid system. Colorado River Compact 1922 allocated 7.5 MAF (million acre-feet) upper basin + 7.5 MAF lower basin + Mexico Treaty 1944 1.5 MAF — total ~16.5 MAF planned vs ~12–13 MAF mean natural flow 2000+ (Schmidt-Yackulic-Kuhn 2023). Seven basin states + 30 federally recognised tribes + Mexico negotiating post-2026 interim guidelines (2007 Interim Guidelines expire). Drought Contingency Plans 2019; 2026 framework still under negotiation as of 2024 IBCA effort.

Sacramento-San Joaquin Delta: California Bay Delta Conservation Plan (BDCP) evolved into California WaterFix (twin-tunnels Brown 2018), shelved by Newsom to single Delta Conveyance Project. Voluntary Agreements 2022+ for instream flow contributions.

8.5 Conveyance

  • California State Water Project (Oroville 1968, 3 380 km3, California Aqueduct 1 129 km to S California).
  • Central Arizona Project (CAP, 1985, 540 km).
  • Central Valley Project (USBR, Shasta 1945 + Friant 1942 + delta-mendota canal).
  • Mojave Aqueduct supplies Antelope/Victor valleys.
  • Roman aqueducts (Aqua Appia 312 BCE +); Han Dynasty Grand Canal.
  • UAE/Saudi desalination pipelines (Taweelah, Jebel Ali).

8.6 Desalination

Global capacity ~100 km3 yr-1 (2024). Reverse osmosis (RO) dominates ~70 % capacity. Energy 3–4 kWh m-3 for RO (theoretical thermodynamic minimum ~1 kWh m-3 for seawater); cost $0.5–1.0 m-3 commodity 2024 in low-cost megaprojects.

  • Israel: Sorek (624 ML d-1 = 228 Mm3 yr-1, 2013) + Sorek B (548 ML d-1, 2023) + Ashkelon + Hadera + Palmachim — provides ~80 % of municipal water.
  • Saudi Arabia: Ras Al-Khair (1 036 ML d-1 hybrid MSF + RO, 2014).
  • UAE: Taweelah (909 ML d-1).
  • Carlsbad CA: 190 ML d-1 (50 MGD), operational 2015, 2 200 acre-ft water cost.
  • Solar PV + RO emerging: Australian remote, Cape Verde (Brilliant Planet), Middle East (NEOM).
  • Brine disposal: outfall mixing, brine pond evaporation, salt recovery; environmental concerns at high-salinity outfalls.

8.7 Wastewater treatment

  • Primary (sedimentation), secondary (activated sludge, trickling filter), tertiary (filtration, disinfection).
  • Activated sludge (Ardern-Lockett 1914) with aeration basins + secondary clarifier + return-activated-sludge (RAS).
  • Membrane bioreactor (MBR): submerged or external membranes replace clarifier; higher MLSS (8–12 g L-1), smaller footprint.
  • Advanced treatment: microfiltration + reverse osmosis + UV-AOP (advanced oxidation, peroxide + UV) for indirect/direct potable reuse.

Notable reuse:

  • Singapore NEWater (2003+): 5 plants, ~40 % of water demand met by reclaimed water; injected into reservoirs or supplied direct non-potable.
  • Orange County Water District GWRS (Groundwater Replenishment System): 379 ML d-1 (100 MGD) expanded 2023, world’s largest indirect potable reuse, recharges Talbert injection barrier + Anaheim spreading basins.

8.8 Stormwater and green infrastructure

Philadelphia Green City Clean Waters (2011, 1.5 B). DC RiverSmart. Portland OR Tabor to River.

8.9 US federal water policy (recent)

  • Bipartisan Infrastructure Law 2021 (PL 117-58) 15 B lead-pipe replacement, 11.7 B Clean Water SRF, $5 B Western drought (Aging Infrastructure + WaterSMART).
  • Inflation Reduction Act 2022 $4 B for Western drought + Lake Mead/Powell stabilisation.
  • Safe Drinking Water Act (SDWA 1974, amended 1986, 1996): EPA + state primacy. National Primary Drinking Water Regulations cover ~90 contaminants. PFAS rules 2024 (4 ng L-1 PFOA/PFOS MCL).
  • Clean Water Act (CWA 1972): NPDES discharge permits + Section 404 dredge-and-fill (USACE). Sackett v EPA 2023 (598 US 651) narrowed “waters of the United States” to require continuous-surface-water connection to traditional navigable waters, removing ~half of wetlands from federal jurisdiction (see employment-and-environmental-law for legal analysis).

9. Climate change in hydrology

9.1 Thermodynamic scaling

Clausius-Clapeyron: saturation vapour pressure de_s/dT ≈ 6.5 % K-1 at typical surface temperatures. Extreme precipitation intensification scales with CC (Trenberth 1999; Allen-Ingram 2002). Pall-Allen-Stone-Aina-Stott-Nozawa-Hilberts-Lohmann-Allen 2011 attributed UK Autumn 2000 floods to anthropogenic warming. Sub-daily extremes scale 6–7 % K-1; some studies find super-CC scaling 14 % K-1 for hourly extremes due to convective dynamics (Lenderink-vanMeijgaard 2008).

9.2 Pattern changes

  • “Wet wet wetter, dry dry drier” (Held-Soden 2006 paradigm): expansion of P − E spatial pattern. Subtropical drying, tropical/midlatitude wetting. Recent revisions (Greve 2014, Byrne-O’Gorman 2015) noted land-only signal weaker than ocean.
  • Hadley cell expansion ~0.5° lat decade-1 (Lu-Vecchi-Reichler 2007; Grise-Davis 2020), pushing subtropical dry zones poleward.
  • Storm-track shifts: jet poleward shift in SH (Polvani-Previdi-Deser 2011); NH less robust.

9.3 Snow and snowpack

  • Mountain snowpack decline: Western US -23 % on average since 1950s (Mote-Hamlet-Clark-Lettenmaier 2005, updated Mote-Li-Lettenmaier-Xiao-Engel 2018). SWE/precipitation ratio declining.
  • Earlier snowmelt timing (-1 to -4 weeks since 1950, Stewart-Cayan-Dettinger 2005).
  • Rain-on-snow flood risk shifting (McCabe-Hay-Clark 2007).
  • Snow drought (Harpold-Dettinger-Rajagopal 2017): warm snow drought (precipitation falls as rain) vs dry snow drought.

9.4 Compound flooding

Coincident pluvial + fluvial + coastal/storm-surge flooding. Wahl-Jain-Bender-Meyers-Luther 2015 documented increasing compound flood frequency at US coastal cities. Bevacqua-Vousdoukas-Zappa-Hodges-Shepherd-Maraun-Mentaschi-Feyen 2020 European compound flood analysis. Hurricane Harvey 2017 (Houston, $125 B damages, NOAA NCEI) and Florence 2018 (Carolinas) prototypical events.

9.5 IPCC AR6 confidence (WG1 Chapter 11, Ch 8 Water Cycle Changes)

  • Heavy precipitation intensification — high confidence at global scale.
  • Drought (agricultural-ecological) — medium-to-high confidence in Mediterranean, W North America, S Africa, Australia.
  • Streamflow flood changes — low-to-medium confidence regional signal.
  • Snowpack decline — high confidence NH mountains.

10. Modelling tools

  • Earth system models: CESM (NCAR), GFDL, HadGEM, IPSL, MPI, NorESM — CMIP6 archive at ESGF.
  • Regional: WRF (Weather Research and Forecasting), CCAM, RegCM, COSMO-CLM.
  • Land-surface models: CLM (Community Land Model, integrated with CESM), JULES, Noah-MP, ORCHIDEE, ISBA.
  • Global hydrology: WaterGAP (Frankfurt), PCR-GLOBWB (Utrecht), H08 (NIES Japan), CWatM (IIASA).
  • Distributed catchment: ParFlow-CLM coupled subsurface-overland-land surface; CLM-PRMS coupling.
  • Cloud-resolving: LES (Large Eddy Simulation), SAM (System for Atmospheric Modeling Khairoutdinov-Randall 2003), DYAMOND project (Stevens-Satoh-Auger 2019).
  • Cloud microphysics schemes: Lin-Farley-Orville 1983 single-moment, Morrison-Gettelman 2008 two-moment, Thompson-Field-Rasmussen-Hall 2008 aerosol-aware, P3 (Predicted Particle Properties, Morrison-Milbrandt 2015).

11. Surface energy balance and evapotranspiration

11.1 Penman-Monteith

Penman 1948 + Monteith 1965 derived the canonical ET equation combining energy balance with aerodynamic + surface-resistance terms:

lambda_v · ET = (Delta · (R_n − G) + rho_a · c_p · (e_s − e_a) / r_a) / (Delta + gamma · (1 + r_s / r_a))

where lambda_v is latent heat of vaporisation (~2.45 MJ kg-1 at 20 °C), Delta is slope of saturation vapour pressure curve, R_n net radiation, G ground heat flux, rho_a air density, c_p specific heat, e_s − e_a vapour pressure deficit, r_a aerodynamic resistance, r_s surface (stomatal) resistance, gamma psychrometric constant. The FAO-56 Penman-Monteith (Allen-Pereira-Raes-Smith 1998) standardises reference ET (ET_0) computation for irrigation scheduling.

11.2 Priestley-Taylor

Priestley-Taylor 1972 simplification for saturated surfaces:

ET = alpha · (Delta / (Delta + gamma)) · (R_n − G) / lambda_v

with alpha ≈ 1.26 (Priestley-Taylor coefficient). Widely used in land-surface modelling.

11.3 Eddy covariance

Direct turbulent-flux measurement via fast-response sonic anemometer + open- or closed-path H2O analyser at 10–20 Hz. AmeriFlux + FLUXNET (1996+, Baldocchi-Falge-Gu-Olson-Hollinger-Running-Anthoni-Bernhofer-Davis-Evans-Fuentes-Goldstein-Katul-Law-Lee-Malhi-Meyers-Munger-Oechel-Paw-Pilegaard-Schmid-Valentini-Verma-Vesala-Wilson-Wofsy 2001 BAMS) network of 700+ towers globally provides ground-truth ET + CO2 + sensible-heat flux.

11.4 Remote sensing ET

  • MODIS MOD16 (Mu-Heinsch-Zhao-Running 2007 + 2011): global ET at 1 km, Penman-Monteith driven by MODIS LAI + reanalysis meteorology.
  • ECOSTRESS (ISS-mounted, 2018+): land surface temperature + ET at 70 m resolution.
  • SEBAL / METRIC (Surface Energy Balance Algorithm for Land / Mapping Evapotranspiration at high Resolution with Internalized Calibration, Bastiaanssen 1998 + Allen 2007).
  • DisALEXI / GOES (Anderson 2007).

12. Snow and cryosphere hydrology

12.1 SWE measurement

  • Snow pillow (NRCS SNOTEL network, ~900 western US stations, ETI Instruments 3 m fluid-filled bags).
  • Snow course (manual sampling, monthly).
  • Cosmic-ray neutron sensing (CRNS, ~30 m footprint, Schattan 2017).
  • Airborne lidar (Painter et al. ASO Airborne Snow Observatory 2013+; now NASA + USDA).
  • Passive microwave (SSM/I, AMSR-E, AMSR2): SWE retrieval at 25 km; saturates at >150 mm SWE.
  • SAR + dry-snow penetration; Sentinel-1 InSAR + Tan Lab.
  • Future: NASA SnowEx (2017+) + SWESARR + ESA CIMR + ROSE-L upcoming.

12.2 Snowmelt modelling

  • Temperature-index (degree-day): M = a · (T_air − T_base) where a ≈ 3–5 mm K-1 d-1.
  • Energy balance (PrISM, SnowModel, CROCUS, SNOWPACK Lehning 2002).
  • SnowPILS at Mammoth (multiple sensors).

12.3 Glacier mass balance

  • Geodetic via DEM differencing (Hugonnet-McNabb-Berthier-Menounos-Nuth-Girod-Farinotti-Huss-Dussaillant-Brun-Kääb 2021 Nature global glacier mass loss 2000–19 of 267 ± 16 Gt yr-1).
  • Surface mass balance via stakes + pits.
  • GRACE for ice-sheet integrated change.
  • Total contribution to sea-level rise from glaciers + ice sheets ~1.8 mm yr-1 of 3.4 mm yr-1 total (IPCC AR6).

13. Water quality

13.1 Major contaminants

  • Pathogens (E. coli, Cryptosporidium, Giardia, norovirus): SDWA Surface Water Treatment Rule 1989 + Long Term 2 ESWTR 2006. Cryptosporidium Milwaukee 1993 outbreak (~400 000 ill, 50+ dead).
  • Disinfection byproducts (DBPs): trihalomethanes (THMs), haloacetic acids (HAAs), bromate. SDWA MCL TTHM 80 µg L-1, HAA5 60 µg L-1.
  • Lead + Copper Rule (LCR 1991, revised LCRR 2021, LCRI proposed 2024): action level Pb 15 µg L-1 → proposed reduction to 10 µg L-1; Flint MI 2014-15 crisis (Hanna-Attisha 2016 Am J Public Health).
  • Nitrate (NO3): MCL 10 mg N L-1; methaemoglobinemia (blue-baby syndrome).
  • Arsenic: MCL revised 50 → 10 µg L-1 (2001); Bangladesh + India crisis (Smith 2000 Bull WHO).
  • PFAS (per- and polyfluoroalkyl substances): “forever chemicals”, >12 000 compounds. EPA Apr 2024 final National Primary Drinking Water Regulation: MCL 4 ng L-1 PFOA + PFOS individually; HI for mixtures; $1 B+ utility costs.
  • Microplastics: California SB 1422 + SB 1263 monitoring requirements 2022+; no federal MCL.

13.2 TMDLs and nutrient pollution

Clean Water Act §303(d) impaired waters; Total Maximum Daily Loads (TMDLs). Chesapeake Bay TMDL 2010 largest (multi-state Bay TMDL for N + P + sediment). Gulf of Mexico hypoxic zone (~15 000 km2 2024, NOAA + LUMCON 2024 cruise): N from Mississippi-Atchafalaya Basin agriculture; Hypoxia Task Force 2008+.

13.3 Harmful algal blooms

  • Lake Erie 2014 Toledo OH water-supply shutdown (microcystin from cyanobacteria HAB).
  • Florida red tide (Karenia brevis annual events).
  • HAB climate-driven intensification (Wells 2020).

14. Paleo-hydrology and climate variability

14.1 Tree rings

Dendrochronology: ring widths reflect growing-season moisture. Cook-Meko-Stahle-Cleaveland 1999 NA drought atlas; Cook-Anchukaitis-Wright-Stahle-Buckley-Wang-Cole-Krusic-Esper 2010 + 2020 NADA + LDA. Williams 2022 megadrought reconstruction.

14.2 Speleothems

Stalagmite delta-18O records (Wang-Cheng-Edwards 2001 Hulu Cave China; monsoon proxy).

14.3 Lake sediments

Diatom-inferred salinity (Fritz 1991), pollen-inferred precipitation, varve thickness.

14.4 ENSO + PDO + AMO

  • ENSO indices: NINO3.4, SOI; reconstructed via tree rings (Cook 2008) + corals back ~700 yr (Cobb 2003).
  • Pacific Decadal Oscillation (Mantua 1997).
  • Atlantic Multidecadal Oscillation (Kerr 2000); recently reframed as forced response (Mann 2021).
  • Influence on US drought + flood frequency (Cole-Cook 1998).

15. Operational hydrology and forecasting

15.1 US National Water Model

NOAA NWM v3.0 (operational 2023): 2.7 M reach NHDPlus stream-network forecasts, ~1 km grid land surface (Noah-MP), 1-h hourly + 18-h short + 10-day medium + 30-day long-range ensembles. Distribution via OWP + Geo-EDF data portal.

15.2 European Flood Awareness System (EFAS)

Copernicus EMS service; coordinates flood warnings across European basins.

15.3 GloFAS

Global Flood Awareness System; 10-day + seasonal river-flow forecasts.

15.4 Reservoir operations

  • US Army Corps of Engineers (USACE) ResSim + HEC-ResPRM.
  • Optimisation under uncertainty: stochastic dynamic programming (SDP), reinforcement learning emerging.
  • Forecast-Informed Reservoir Operations (FIRO): Lake Mendocino FIRO pilot 2014–24 increased storage 28 % using AR forecasts (Jasperse 2017 + Talbot 2023).

16. Hydroclimate ML

  • Sun-Lawrence-Kollet-Maxwell-Condon-Tarboton-Tijerina-Salazar 2021 (Frontiers Big Data) machine-learning hydrology review.
  • ML for streamflow: Kratzert-Klotz-Brandstetter-Gauch-Hochreiter-Nearing 2019 LSTM streamflow prediction outperformed conceptual models.
  • ML for precipitation downscaling: cross-ref ai-and-machine-learning-for-climate.
  • Google Flood Hub (FloodHub 2018+ originally India + Bangladesh, global expansion 2023): operational AI-based flood forecasting in 80+ countries; Nature 2024 paper (Nearing-Cohen-Dube-Gauch-Gilon-Harrigan-Hassidim-Klotz-Kratzert-Metzger-Nevo-Pappenberger-Prudhomme-Shalev-Shenzis-Tekalign-Weitzner-Matias).

Adjacent

Graph View