Research status of medium and long term hydrological forecasting

Short-term flood forecasting is very important for reservoir flood control and river flood control and rescue, but its prediction period is too short, and it is often difficult to meet the requirements of flood control and drought control, reservoir scheduling, hydropower station operation and shipping management, and the growth forecast period is the national economy.

Departments are becoming more and more urgent. However, the prediction period of hydrological forecasting is limited by the hydrological law itself, and it cannot be very long, for example, the prediction period of flood forecasting in river sections cannot exceed the propagation time of flood waves in river sections, and the prediction period of rainfall runoff forecasts cannot exceed the confluence time of river basins. Therefore, to carry out hydrological forecasting in the medium and long period, it is necessary to introduce meteorological factors and closely integrate hydrological forecasting with meteorological forecasting.

Hydrological forecasting is divided into flood forecasts for floods in flood seasons of major rivers, lakes and reservoirs according to the characteristics of water conditions and forecast contents: flood levels, flood flows and flood processes; Dry water forecasts for water levels, flows and river network storage in the main dry season. Ice forecasts for the freezing and ablation process of water bodies, including freezing date, ice thickness, thawing date and drift ice.

Typhoon surge forecasts mainly forecast the sharp increase in water volume and water level rise in the estuary section of the sea; **Sediment forecasts for river sediment content and reservoir sediment scouring. According to the prediction period: short-term forecasts with a prediction period of a few hours to a few days; Medium-term forecasts with a foresight period of more than 2 5 days and less than 10 days (or 15 days); It is generally considered that the prediction period is more than 15 days, and the long-term hydrological forecast is within 1 year; Ultra-long-term hydrological forecasts with a foresight period of more than one year are called.

Empirical and semi-empirical methods, hydrological modelling methods and statistical forecasting methods are generally used in hydrological forecasting.

1. The change of all hydrological elements has its specific physical mechanism, and it is the basic principle to explain the rationality of the prediction factors from the physical causes, and to start from the analysis of the physical mechanisms of the formation of hydrological phenomena, so that the forecasting model is based on strict physical causes, which should be followed in the future medium and long-term hydrological forecasting and other hydrological forecasts.

2. The hydrological system is a complex giant system, and the temporal and spatial variation of hydrological elements has a high degree of nonlinearity. It is difficult to make significant progress in medium- and long-term hydrological forecasting methods, which are mainly based on linear methods. According to the nonlinear characteristics of hydrological changes, it is necessary to break through the shackles of traditional concepts, introduce new analytical approaches and promote multidisciplinary collaboration.

Fuzzy analysis, artificial neural networks, and other nonlinear scientific methods may be effective new ways of analysis.

3. The complexity of the hydrological system and the uncertainty of the change of hydrological elements determine that it is impossible to comprehensively improve the reliability of medium and long-term hydrological forecasting calculations, decision-making and decision-making by using a single method or pinning hopes on the improvement of mathematical statistical methods. The combination of various methods, i.e., the use of the so-called coupling approach, especially the deterministic artificial neural network method combined with the uncertain random, fuzzy, gray, chaotic and other methods and their various couplings, will play an increasingly important role in medium- and long-term hydrological forecasting.

4. Insufficient or poor data representativeness is one of the important reasons restricting the development of medium and long-term hydrological forecasting. It is necessary to actively carry out the observation and accumulation of data required for medium- and long-term hydrological forecasting, especially the changes in forecasting factors and influencing factors caused by human activities and global warming, which have a great impact on medium- and long-term hydrological forecasting, and their monitoring is undoubtedly an effective way to improve the forecasting effect.

5. The trend of medium- and long-term hydrological changes is mainly affected by the changes of large-scale hydrometeorological elements, and the changes of marine factors, especially the El Niño and La Niña phenomena, have a certain control effect on continental hydrometeorology, and the grasp and interpretation of their changes will play an important role in improving the accuracy of medium- and long-term hydrological forecasts.

According to the hydrometeorological elements in the early stage, the method of genetic analysis and mathematical statistics is used to scientifically carry out the hydrological elements in the future for a long period of time (ten to one year), which is called medium and long-term hydrological forecasting. 1. Medium- and long-term hydrological forecasting usually refers to the hydrological forecast that the prediction period exceeds the maximum confluence time of the basin and is more than 3 days and less than 1 year. The main contents of the forecast are:

Runoff, estuarine water level, ice clearing, river channel and erosion changes, drought and flood trends, etc. The main influencing factors for the long-term change of hydrological situation include atmospheric circulation, solar activity, underlying surface conditions, other astronomical and geophysical factors (such as changes in the earth's rotation speed, volcanic eruptions, relative positions of planets, etc.), and human activities.

Medium- and long-term hydrological forecasting models, there is currently no one model that works for all hydrological sequences. The applicability of forecasting models is an issue that still needs to be studied in depth. Based on the principle of specific analysis of specific problems, the medium- and long-term forecasting of a specific hydrological series often requires analysis, experimentation, testing and other steps to finally find a suitable forecasting model.

The research on medium- and long-term hydrological forecasting is still in the development stage, and compared with short-term hydrological forecasting, it lags behind the actual requirements of production. In the research of forecast theory, more attention is paid to the statistical correlation characteristics of hydrological series, and relatively little attention is paid to the physical genesis. In terms of forecasting methods, there is not enough research on the effectiveness of various methods, which makes it difficult for existing methods to be popularized and applied in practice. In terms of the practical application of the forecast results, the medium- and long-term hydrological forecasts mainly play a reference role in the macro management of water resources. The Hydrological Information Code states:

In order to meet the needs of production development, qualified water units can issue medium and long-term water outlook, only provide leadership to grasp the reference, not as the basis for taking specific measures, and pay attention to the actual changes, at any time to supplement the outlook. "The root cause of this situation is the lack of understanding of medium- and long-term hydrological forecasts, and the lack of attention to slippage. With the rapid development of the national economy and the enhancement of the awareness of natural resource development and environmental protection, the departments in the basin have higher and higher requirements for flood control and water supply prediction.

The demand for production will inevitably lead to the development of Shouxiaodong technology, so the research on medium and long-term hydrological forecasting will receive more and more attention from scientific researchers.

Hydrological forecasting refers to the qualitative or quantitative hydrological situation of a certain water body, a certain area or a hydrological station in a certain period of time in the future based on the previous or current hydrometeorological data. It is of great significance to flood control, drought control, rational use of water resources, and national defense. Techniques and operations for scientific** and forecasting future hydrological conditions.

Hydrological forecasting is an important aspect of hydrology for economic and social services, especially for the forecasting of catastrophic hydrological phenomena and the short-term, medium-term and long-term forecasting of the comprehensive utilization of large-scale water conservancy hubs. China has carried out forecasting services for flood water level and flow, dry water level and flow, sediment content, various ice conditions, water quality, etc.