Many people now agree that one of the most serious – if not the most dangerous – global issues facing humanity is climate change. Global temperatures continue to rise, extreme weather events are becoming frequent, and countless species face the threat of extinction. Meanwhile, emerging technologies such as artificial intelligence (AI) are being increasingly used to address environmental challenges and slow the pace of climate change.
Extreme weather phenomena such as heatwaves, cold spells, heavy rains, droughts, and storms can cause severe damage to infrastructure, especially in urban areas, and endanger public health. AI has improved forecasting accuracy, helping reduce the risks of injury and death. It is now widely used to create more precise climate models and reduce prediction bias.
By integrating AI with numerical climate simulations, scientists can fill data gaps and decrease uncertainty in climate projections. AI has successfully anticipated El Niño, tropical instability waves, and shifts in global temperatures, and is increasingly used to predict extreme weather events like heatwaves, droughts, and intense rainfalls, which are becoming more common due to global climate change.
Local impacts in Kurdistan
The Kurdistan Region is already experiencing the effects of climate change, and the situation continues to deteriorate. According to a study on the severity of drought from 1998 to 2017, vegetation made up 52.4% of the total area in 1998. However, as a result of recurrent droughts, this percentage dramatically declined after 2000. As of now, vegetation only covers 12% of the land. This significant reduction in vegetation is compounded by rising greenhouse gas emissions, particularly carbon dioxide, which, according to the World Meteorological Organization, have reached alarming levels. These environmental shifts have led to more unpredictable and severe weather patterns, making accurate forecasting even more critical for Kurdistan.
As recently as March 2025, much of Kurdistan experienced unusually dry and cold weather, with temperatures so low. At the same time, temperatures are rising rapidly, resulting in large and sudden fluctuations. In April, there were also several days of dust storms across the region, making it difficult for people to go outside. These extreme weather swings, combined with erratic precipitation and frequent dust storms, highlight the rapid pace of change in regional weather patterns and underscore the importance of precise forecasting. Just like last year, the region endured devastating torrents in the city of Zakho and Duhok Governorate, along with severe droughts and flash floods in Erbil and Sulaymaniyah.
These events inflicted significant damage to infrastructure and public health systems. One of the major contributing factors was the lack of early warning systems. Unfortunately, many residents still distrust weather forecasts due to previous inaccuracies.
AI in global forecasting
AI-based forecasting may help bridge this gap. Tools such as NASA’s Earth Science Satellites, NOAA’s GOES-R Series, and ESA’s Sentinel Satellites – on which the Kurdistan Region should rely – can enhance prediction accuracy by processing enormous volumes of data from satellites, weather stations, and ground sensors.
For instance, a 2019 study conducted by a team led by Melinda Pullman used deep learning algorithms to identify hailstorms by analyzing infrared brightness temperatures and other hail-related parameters. Meanwhile, a 2021 study conducted by a team led by Kasuni Adikari evaluated the predictive capabilities of multiple AI models – including the adaptive neuro-fuzzy inference system, wavelet decomposition function, convolutional neural networks, and long short-term memory networks – for forecasting floods and droughts. These technologies have significantly improved modelling and forecasting, helping to reduce the uncertainty that typically surrounds sudden shifts in weather patterns.
The application of AI in forecasting meteorological parameters like temperature, precipitation, and extreme weather events has proven effective. This is especially vital for Kurdistan, where catastrophic weather events are becoming more frequent. AI could predict future droughts, temperature increases, and regional rainfall patterns. The region has already experienced a 2°C increase in average temperature over recent decades, with the U.S. Agency for International Development noting in 2017 that Iraq’s mean annual temperature had been rising since the 1950s. Projections suggest that days with temperatures exceeding 35°C will increase significantly between 2020 and 2060, further emphasizing the need for AI-driven models to detect and respond to these changes.
Meteorological models improve when machine learning algorithms analyze large datasets of historical and recent weather conditions like temperature, precipitation, and wind speed. A 2020 study by Zhijun Chen compared three machine learning models—deep neural networks, time convolution neural networks, and short-term memory neural networks—to estimate daily evapotranspiration in Northeast China using techniques like support vector machines, random forest models, and empirical formulas. Researchers found that distributed lagged nonlinear models outperformed traditional cross-correlation functions for detecting time lags and selecting predictive variables. Standardized precipitation evapotranspiration indices were also more accurately predicted by machine learning methods than older nonlinear models using artificial neural networks.
Agriculture and land management
In addition to predicting the weather, AI-powered satellites collect vast volumes of data related to land resources, improving the efficiency of spatial land planning and sustainable development. AI can optimize agricultural land use, predict soil respiration, and track land-use changes. In 2019, Lopez Santos explored artificial neural networks in sustainable development, finding that orchard productivity was influenced by planting conditions, environmental compatibility, and growers’ knowledge.
Another 2019 study by Al-Dousari et al. applied support vector machines and neural networks to assess and predict land cover and soil use in Kuwait, using neural networks and linear regression to evaluate subsurface soil characteristics and estimate soil respiration more accurately.
These AI models are especially helpful in Kurdistan, where farming has become harder due to climate change. Droughts have destroyed nearly half of the region’s agricultural fields, resulting in significant crop losses, particularly affecting key crops such as wheat and barley.
Government and institutional responses
It is noteworthy that the Kurdistan Regional Government has taken action to fight climate change, showing considerable interest in environmental and technological issues, particularly through the Ninth Cabinet, which supports the deployment of cutting-edge technology and climate-related projects.
Universities such as Salahaddin University, the University of Kurdistan Hewler, and the University of Zakho have established dedicated centers for AI research, and several commercial and governmental institutions are also pursuing climate-tech solutions. While the region has many skilled professionals, greater public awareness and real-world application of these technologies are still needed to fully address the growing climate crisis.
Integrating AI into weather forecasting is essential to tackling immediate threats and broader climate challenges in the region. AI can clearly enhance the accuracy of predictions regarding temperature changes, precipitation, and extreme weather events and thus provides valuable tools for improving disaster preparedness, public health responses, and urban development plans.
Diman Zuhair Jacksi is a lecturer and AI Coordinator at the College of Engineering, University of Zakho.
