Drought is one of the most challenging disasters that impact the natural and cultural ecosystems across the world, especially in the climate dependent sectors of arid and semi-arid areas. The aim of this article is to share the experiences gained and enhance the readers’ awareness on the status of drought and process of the early warning systems (EWS) in south India. Drought status of three agroecologically different states is included in this article, such as Kerala, Tamil Nadu and Telangana. As far as Tamil Nadu is concerned, Karur, Thuthukudi, Krishnagiri, Namakkal, Trichy and Thirunelveli districts are water scarce compared to other districts in the state. The districts such as Wayanad, Thiruvananthapuram, Idukki and Palakkad in Kerala have received lesser rainfall compared to the other parts of the state during the period 1981 to 2019. In Telangana, the mandals such as Nagarkurnool, Jogulamba-Gadwal, Wanaparthy, Mahabubnagar Nalgonda and Yedadri are frequently hit by dry spells and droughts. As a case study, weather early warning dissemination, carried out at Parambikulam Aliyar basin, Coimbatore, Tamil Nadu, during Khariff and Rabi seasons, using IMDs medium and extended range forecast is also elaborated in particular in the article. As far as the accuracy of forecast is concerned, probability of false detection (false alarm rate) was found to be 0.81 for Khariff and 0.30 for Rabi season, indicating the need for better performance in the accuracy of dry spell early warning, disaster preparedness and response. In-spite of this, access to early warning has supported the farmers during harvest and land preparation with a utility score of 72% and 59%, respectively. In Parambikulam Aliyar basin, remote sensing products such as MODIS-NDVI, NDWI and TWI was also used to identify the real-time progression of monthly vegetative condition for Kharif and Rabi seasons. NDVI values were used to monitor the district level vegetation condition and compared it with the drought year 2016, the difference in area under barren land was 76% less during Khariff, 2021 and 44% during Rabi, 2021.This study is a compilation of lessons learned from different states and the existing knowledge and practice in early warnings, and recommends the need for a holistic approach in drought and dry spell monitoring along with better accuracy and dissemination to minimize climate-related shocks in agriculture.

Climate change is often linked with record-breaking heavy or poor rainfall events, unprecedented storms, extreme day and night time temperatures, etc. It may have a marked impact on climate-sensitive sectors and associated livelihoods. Block-level weather forecasting is a new-fangled dimension of agrometeorological services (AAS) in the country and is getting popularized as a climate-smart farming strategy. Studies on the economic impact of these microlevel advisories are uncommon. Agromet advisory services (AAS) play a critical role as an early warning service and preparedness among the maize farmers in the Parambikulam–Aliyar Basin, as this area still needs to widen and deepen its AWS network to reach the village level. In this article, the responses of the maize farmers of Parambikulam–Aliyar Basin on AAS were analyzed. AAS were provided to early and late Rabi farmers during the year 2020–2022. An automatic weather station was installed at the farmers’ field to understand the real-time weather. Forecast data from the India Meteorological Department (IMD) were used to provide agromet advisory services. Therefore, the present study deserves special focus. Social media and other ICT tools were used for AAS dissemination purposes. A crop simulation model (CSM), DSSAT4.7cereal maize, was used for assessing maize yield in the present scenario and under the elevated GHGs scenario under climate change. Our findings suggest that the AAS significantly supported the farmers in sustaining production. The AAS were helpful for the farmers during the dry spells in the late samba (2021–2022) to provide critical irrigation and during heavy rainfall events at the events of harvest during early and late Rabi (2021–22). Published research articles on the verification of weather forecasts from South India are scanty. This article also tries to understand the reliability of forecasts. Findings from the verification suggest that rainfall represented a fairly good forecast for the season, though erratic, with an accuracy score or HI score of 0.77 and an HK score of 0.60, and the probability of detection (PoD) of hits was found to be 0.91. Verification shows that the forecasted relative humidity observed showed a fairly good correlation, with an R² value of 0.52. These findings suggest that enhancing model forecast accuracy can enhance the reliability and utility of AAS as a climate-smart adaptation option. This study recommends that AAS can act as a valuable input to alleviate the impacts of hydrometeorological disasters on maize crop production in the basin. There is a huge demand for quality weather forecasts with respect to accuracy, resolution, and lead time, which is increasing across the country. Externally funded research studies such as ours are an added advantage to bridge the gap in AAS dissemination to a great extent.

Evaluation of climate-change impacts on crop yield for the north-east agroclimatic region of Tamil Nadu, was assessed during 2014, using DSSAT4.5 model under IPCC, RCP4.5 emission trajectory. Results showed that, the yields of the major crops, viz. rice (Oryza sativa L.), groundnut (Arachis hypogaea L.) and sugarcane (Saccharum sp.), tend to decrease towards the end of 21st century. In this purview, this study was an endeavour to investigate the easiest agronomic adaptation options, i.e. sowing-date adjustment to offset the projected climate-change impacts on yield. Advancing or delaying sowing dates were found to be giving better yield under climate change scenario. Results for rice showed that, adjusting the sowing dates can have better yield rates in future. Experimental outcomes showed that rice crop sown on 1st November led to a yield increase of 2.18%, and this would be the most appropriate response to offset the impacts. The groundnut sown on 10th December and 1 January were performing well, with a yield increment of 7.51 and 7.50%, respectively, by the end of century. For sugarcane, planting season extends to 6 months in the study area. The results revealed that, planting on 15th March can increase the yield by 3.5% in the near century period under RCP 4.5. The overall agreement between simulated and observed rice yields was 74% with an R2 and (RMSE) values of 0.92 and 807 kg/ha respectively. The overall agreement between simulated and observed groundnut yield was 82% with R2 , root mean square error (RMSE) values of 0.71 and 408.98 kg/ha respectively. The agreement between simulated and observed sugarcane yields was 76% with 0.76 and 25.82 tonnes/ha as R2 and RMSE values respectively. The challenge is to disseminate this information for the benefit of the farmers as part of awareness creation, preparedness and capacity building. Therefore, researchers and agriculturists have a major role to play in taking forward, scientific findings for the benefit of the stakeholders.