Inflow Forecasting for Hydro Catchments Ross Woods and Alistair McKerchar NIWA Christchurch
Inflows Water flowing into hydro storages Usually measured by monitoring the levels and outflows from hydro storages Inflow = Outflow + (Rate of increase of storage) Variable at many timescales, different from place to place Why make forecasts? How?
Nearly 80 years of levels for Lake Wakatipu Data courtesy Contact Energy Ltd
Main hydropower storages Lake Storage Capacities 2000 Capacity (GWh) 1600 1200 800 400 0 Taupo Rotoaira Moawhango Waikaremoana Cobb Coleridge Tekapo Pukaki Ohau Benmore Hawea Wanaka Wakatipu Te Anau Manapouri Mahinerangi Lake
Forecasting for Three Timescales Weather systems: 0-3 days from now Seasonal forecasting: 1 week to several months Interannual variability: Decadal-scale
Timescale: 0-3 days Why? How much electricity can we make? Flushing flows; High flow management What we need to know What water is already in the catchment (river flows, soil water, groundwater)? How will weather system affect the rainfall, snowfall and evaporative demand? With this information, we can use a model to calculate snow melt, runoff from land into rivers, flow along rivers to lakes This can be done in more or less detail
TOPNET - 1 Snowpack, canopy and root-zone submodels over a shallow water table Topo. controls Snowpack Melt Throughfall Recharge Rain, Evap Canopy Root zone Sat. zone Surface flow Subsurface flow Sub-basin outflow Other sub-basins: each one is unique River Network
TOPNET - 2 Sub-basin outflows are connected to river network routing Model gives results for every sub-basin and every river reach, every hour
The Clutha River Catchment L. Wanaka L. Hawea Shotover R. L. Wakatipu Alexandra L. Roxburgh Pomahaka R. Balclutha (20,500 km 2 )
The Catchment Model
RAMS Rainfall Forecasts over Clutha Catchment Alexandra Balclutha
Real Time Forecasts for Alexandra - Nov. 1999 using RAMS forecast on 20 km grid 1200 1000 Alexandra Inflow (um/h) 800 600 400 200 Successful flood forecast on a large event in a slowlyresponding catchment Diamonds mark forecast time at 0800 each day 0 10-Nov-99 12-Nov-99 14-Nov-99 16-Nov-99 18-Nov-99 20-Nov-99
Next Generation of Forecasts Operational Forecasting for catchments, with linked weather and hydrology models Operational Forecasting for Regions, with linked weather and hydrology models
Timescale: 1 week 6 months Why? Is there a hydro drought coming? When do we need pre-empt with thermal? What we need to know for inflow forecasts: How much rainfall/snowfall? What temperatures? No deterministic answers: uncertainty is central. Aim to reduce the uncertainty This can be done in more or less detail
Hermitage monthly rainfall (1931-1996) Seasonal 1500 Patterns 1000 Monthly rainfall (mm) 500 0 JAN MAR MAY JUL SEP NOV FEB APR JUN AUG OCT DEC Lake Wanaka monthly inflows 1931-1995 Max Min 75% 25% Median Hermitage monthly rainfall 800 600 Monthly inflow (m3 400 200 0 JAN FEB MAR APR MAY JUNE JULY AUG SEPT OCT NOV DEC Max Min 75% 25% Median Lake Wanaka monthly mean inflows
The Past as a guide to the Future 1. It is common to forecast the performance of energy systems over the next season by starting from the current lake storage, and then using weekly inflow sequences from each of the last 70 years of inflows. This lets us sample the variability of inflows at this time of year. BUT, the unspoken assumption is that each of the last 70 years is equally likely in the next 3 months. Once we are in a strong El Nino or La Nina, some years are much more likely than others. 2. Some energy system models use auto-regressive models to forecast streamflows: on average these might be useful, but extreme seasons have a different persistence structure 3. Use ENSO -streamflow or rainfall-streamflow forecasting
250 Lake Taupo Natural Spring Inflows 1935-1992 200 150 100 Spring inflo 50-30 -20-10 0 10 20 30 Spring SOI
Lake Taupo natural spring inflows 1935-1992 250 200 150 100 Non-Outlier Max Non-Outlier Min Spring inflo 50 < -5 (-5,5] > 5 Spring SOI 75% 25% Median
1200 Clutha lakes summer inflow vs spring SOI 1000 800 600 400 200 Summer i 0-30 -20-10 0 10 20 30 Spring SOI
Clutha lake summer inflows vs spring SOI 1200 1000 800 600 400 Inflow (m3/s) 200 0 <= -5 (-5,5] > 5 Spring SOI Max Min 75% 25% Median
NIWA National Climate Centre and National Centre for Water Resources Forecasts each month of the streamflows for the coming 3 months www.niwa.co.nz/ncc Based on: climate forecasts, current status of river flow and soil moisture, expert knowledge and statistical analysis of past data. Forecasts developed as the probability of Above normal, Normal and Below normal flows. In the long run each of these occurs 33% of the time. So random guesses have 33% hit rate. We have had 40-45% hit rate over the last 3 years www.niwa.co.nz/ncwr - water quality, quantity, lakes, groundwater,
Next Steps Develop new methods for forecasting climate 1. Statistical weather generators 2. Regional climate models Both of these are ways to generate many sequences of daily rain and temperature for the coming season (conditioned on the current oceanatmosphere status ENSO etc) Either way, we need to convert these climate forecasts to inflows: can use the same model as for the 0-3 day forecasting.
Topnet Example of Regional Modelling in Northland Topo. controls Snowpack Melt Throughfall Recharge Rain, Evap Canopy Root zone Sat. zone Surface flow Subsurface flow Sub-basin outflow Other sub-basins: each one is unique River Network
Timescale: Interannual Why? Investment in new generation; Assessing the expected performance of existing system Fact: Inflows can be very different from one decade to another Up to 15% changes in South Island Inflow differences caused mainly by rainfall variability. This is climate variability (as opposed to climate change: e.g. rise in temperature & rain in west and south of South Island by 2050) Rainfall differences are caused by slow changes in ocean-atmosphere circulation, called Interdecadal Pacific Oscillation (IPO, PDO) 1945-1977, 1978-1999?, 1999?-
South Island West Coast Rainfall
Clutha River at Balclutha 1947/48 to 1998/99 (data courtesy Contact Energy Ltd) Annual mean flow (m 3 /s) 900 700 500 Mean = 536 m 3 /s Mean = 612 m 3 /s 300 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
Lake Te Anau inflow, 1931/32 to 2001/02 (data courtesy FRST, Meridian Energy Ltd & The Market Place) Annual mean inflow (m 3 /s) 500 1977/78-1998/99 Mean =305 m 3 /s 400 1931/32-1945/46 1946/47-1976/77 Mean = 277 m 3 /s Mean = 269 m 3 /s 300 200 100 Jan-30 Jan-40 Jan-50 Jan-60 Jan-70 Jan-80 Jan-90 Jan-00
Summary 0-3 days: weather system forecasting and a model of catchment processes 1 week 6 months: Effects such as ENSO can shift the distribution of inflows far from the typical distribution Interdecadal: Differences between decades of order 15% - be careful which years of data you use! For these inflow forecasts to be useful, we need better links with energy system modellers, esp for advice on how best to communicate inflow forecasts