The Netherlands has for long been witnessing problems due to subsidence. While urban infrastructure is mostly safeguarded from differential settlement by deep pile foundations, the same cannot be said about greenhouses. The greenhouses of the Netherlands has enabled the country to become the largest exporter of vegetables, second only to the United States of America. It is imperative that these greenhouses which form the backbone of the agriculture infrastructure of the country, are monitored continuously to minimize unprecedented damage.

This thesis aims to study the feasibility of using time series Interferometric Synthetic Aperture Radar (InSAR) as a means to monitor differential settlement in greenhouse structures. The analysis was primarily done using RADARSAT-2 data. In case of translucent surfaces of greenhouses, it was important to firstly identify the physical targets that are associated to scattering centres. This was done by analysing the statistics of the heights of the scatterers which helps in ascertaining where the radar signal is getting back-scattered from. It was inferred that the persistent and distributed scatterers are primarily identified from objects on the roof and outer walls of the greenhouses.

Moreover, the magnitude of deformation estimated from the scatterers have been corroborated with geotechnical data. It was seen that higher magnitudes of deformation was seen in locations with compressible soil types such as clay and peat close to the ground surface. It was also seen that greenhouse structures are prone to differential settlement when the depths of the piles are insufficient in areas with varying soil types. The effect of thermal contributions has also been studied and it was found that the estimation of thermal expansion does not significantly affect the estimated deformations.

From the study, it is evident that time series InSAR offers an effective means to monitor differential settlements in greenhouses. In order to check for differential settlement in individual greenhouses, it is proposed that a persistent scatter interferometric analysis be done initially and if it is seen that the density of these scatterers is insufficient, the analysis can be followed up with time series interferometry of distributed scatterers. Incorporating multiple track directions of radar data increases the number of greenhouses that can be monitored. Moreover, it was also seen that persistent scatterers were identified from additional greenhouses when Sentinel-1 data was used, despite its poorer spatial resolution. For further study, it is recommended that corner reflectors are used to validate the positions of the targets that are identified as persistent and distributed scatterers.

This report represents the baseline study of the Project "Cotton Water", a collaboration between Solidaridad Asia, TU Delft and other participating instututions to improve the livelihood of cotton farmers in the Vidarbha and Marathwada regions of Maharashtra, India. This baseline study was divided in three main steps: a desktop study, where high resolution maps of precipitation, potential evaporation, soil type and landuse were used in conjunction with a smallholder socio-hydrological model to identify 'hotspots' where farmers' capital falls below poverty lines; a field survey, in which farmers were extensively questioned on their financial situation and farming practices as well as their perception of water scarcity and irrigation schemes; and a final synthesis where interventions are analysed with the smallholder socio-hydrological model and a psycho-social analysis of farmer behaviour is delivered alongside a mapping of current water productivity of cotton in the study area. The main results found are that the proposed water harvesting and recharge interventions increase and stabilize yields, and the overall effect on capital are moderate. Other factors that do not impact water availability including fertilizer and labour were found to have notable impacts and should be well understood to accurately improve farmers' situations. Financial aspects including cotton sale prices and loan interest rates had strong impacts on farmers' capital development as well, particularly with high interest rates punishing some farmers. An analysis of good- and poor-performing farmers demonstrated that irrigation in general and micro-irrigation did improve probability of good farmer performance as did increased yields. Older men also showed higher rates of profit, demonstrating that the impact of experience may increase profit margins, even if it necessarily doesn't increase likelihood to adopt interventions. What was found to increase probability of adopting irrigation and irrigation technology was low promotion exposure. It is hypothesized that increased promotion may influence many farmers negatively, fostering an attitude of despair rather than informing them of opportunity. The psycho-social evaluation also found that solutions that are reasonably expensive but not too costly have higher chances of being adopted.
Four main recommendations were made to help improve farmer welfare with respect to the scope specified. It was recommended to: limit promotion and to be more selective and positive with the message; focus on localized water storage interventions to increase farmers' access to water; regulate cotton prices through government intervention or contracts with clothing companies to decrease vulnerability to price changes; and improve access to loans from the government and reduce the role of money lenders who often are the ones charging the greatest interest rates.