Estimation and forecasting Evaporation at Baramati Station, Maharashtra, India

Abstract

[500] Accurate modelling of evaporation is crucial for effective agricultural water management and irrigation scheduling. Evaporation represents a significant component of the hydrological cycle, directly influencing crop water requirements and overall agricultural productivity. The variability of evaporation rates is largely governed by environmental factors such as wind speed, temperature, and humidity. Traditionally evaporation measured using experimental methods, however missing of data due to experimental error or non-recording etc. hampers the traditional measures. Parallelly, as evaporation is an incidental, nonlinear, complex, and unsteady process and the traditional methods involved in estimating the same are penman equation, the Hargreaves equation, and the Blaney-Criddle method. However, these hard data equations stand still when some of the data is missing. Thus, the challenge of missing data and errors in traditional methods often necessitates reliance on univariate modelling approaches, which limit the predictive capability by using only evaporation data for forecasting. The present work focuses on Estimating and Forecasting Evaporation few time steps ahead, within no time. Estimation of Evaporation at time t, t+1(one day) and t+2(two day) in advance at location in Baramati, Maharashtra, India. For these observed values of Maximum and Minimum Temperature (°C), Maximum and Minimum Humidity (%), Average wind Speed(km/hr) and Evaporation(mm)and previous time steps considered to estimate and forecast evaporation few time steps ahead using Multi Gene Genetic programming. The set of models developed can be directly utilised for early alarming system for effective irrigation systems.

Specification

Description:The present study consists of application of Multi Gene genetic programming as an artificial intelligence technique to estimate and forecast the evaporation at real time at Baramati location, Maharashtra, India. The Estimation is done with input parameters: Maximum and Minimum Temperature (°C), Maximum and Minimum Humidity (%), Average wind Speed(km/hr) and Evaporation(mm) at time t. However, the forecasting after 24 hrs and 48 hrs requires input parameters prior data of Maximum and Minimum Temperature (°C), Maximum and Minimum Humidity (%), Average wind Speed(km/hr). Traditionally evaporation is recorded through Pan or lysometer and penman equation, the Hargreaves equation, and the Blaney-Criddle method. However missing data can hinder the process of estimation and forecasting. Using the equation developed by MGGP which consists of prior data as input and forecasting, can take care of missing data. This work aims to estimate and forecast evaporation for short time intervals (t, t+1, and t+2 days) in Baramati, Maharashtra, using observed values of temperature, humidity, and wind speed for estimation of evaporation. One day prior data of temperature, humidity, and wind speed can be used for forecasting of evaporation 24 hrs ahead and 48 hrs ahead. , Claims:1. Application of Multi gene Genetic Programming to real time Evaporation(mm) estimation using Temperature, Humidity, and wind speed as influential parameters.
2. Real time forecast of Evaporation in mm at 24 Hrs and 48 Hrs ahead in time at Baramati station using Temperature, Humidity, and wind speed as influential parameters recorded at time t (one day before).
3. Real time forecast of Evaporation in mm at 24 Hrs and 48 Hrs ahead in time at Baramati station using Temperature, Humidity, and wind speed as influential parameters recorded at time t-1 (one day before).

Documents