A history of the monsoon in southern India between 1730 and 1920 and its impact on society: with a particular focus on Tamil Nadu.

The field of historical climatology has rapidly developed over the past two decades, driven by the understanding that improving the knowledge of the past can help make informed decisions about the future. Most studies currently focus on mid-latitude regions, whereas, this thesis is part of a growing body of work that seeks to expand the methodology into tropical and subtropical regions. Because of India’s colonial past, there is a substantial amount of extant, English language documentation available to researchers and these documents can help to build an understanding of both historical monsoon magnitude and changes in the social-ecological systems of the past. This thesis firstly explores the application of two types of documentary evidence for reconstructing the monsoon of Southern India, with a focus on the northeast monsoon of Tamil Nadu between 1730 and 1920. The first type is terrestrial documentation; this predominantly consists of government records, diaries, correspondence, historical accounts, newspapers and early instrumental records. The content of these documents was calibrated to modern instrumental rainfall, creating a five-point index of northeast monsoon magnitude, the first reconstruction of its length and resolution for the region, which had a strong correlation with modern instrumental data of 0.74, significant at the 0.05 threshold. The reconstruction was extended to present using degraded modern instrumental data: this new dataset showed that there is persistence in the cyclic pattern of increased and decreased monsoon rainfall, with approximately six epochs, each containing a period of above and below normal northeast monsoon rainfall between 1730 and 2017. The quality of this reconstruction was sufficient to investigate teleconnections between the northeast monsoon and the Indian Ocean Dipole and El Niño Southern Oscillation over nearly three centuries. This study demonstrated a non-stationary relationship between northeast monsoon rainfall and each of the two climate drivers, with good agreement between strong El Niño/La Niña and strong positive IOD events and normal-increased or normal-decreased northeast monsoon magnitude. Secondly, this thesis explored the use of data from within ships’ logbooks to reconstruct monsoon magnitude between 1750 and 1920, for both Kerala and Tamil Nadu and for each rainfall season. Two reconstruction methods were explored, composite plus scaling, and principal component regression. Interestingly, due to the unique monsoon climate and the need to adequately resolve it in both time and space, these reconstructions were less successful than their terrestrial counterpart. This thesis goes on to use the information contained within the documentary sources to explore details of historical famine events in the region of modern day Tamil Nadu. These periods of famine were identified and characterised. The management strategies employed by government and their changes over time were identified, revealing a distinct and widespread transition to non-interventionist famine management on the part of the British government at the turn of the nineteenth century. Finally, this thesis uses the new northeast monsoon reconstruction to explore the interwoven relationship between political failure, rainfall and famine