Shallow groundwater tables and associated salinity problems have become dominant features in agricultural areas around the world. This is especially so in arid and semi-arid regions where rising water tables and associated soil salinization are an increasing agricultural and environmental threat. Groundwater has the ability to mobilize and transport soluble salts when it moves through the soil profile. Once the water table reaches a critical depth below the ground surface, evaporation of this water can occur via capillary rise, transporting soluble salts with it to the soil surface. Over time, this leads to an accumulation of salts at soil surface and subsequent land degradation as well as losses in production. Whilst there have been several studies on ground water associated salinity, the movement of groundwater in particular with respect to recharge events and its impact on soil surface salt accumulation phenomenon still remains to be clarified. "How can we explain shallow water table behavior in response to added water events?", "Can we solve the salt accumulation problem by controlling the groundwater level?" are the two main questions to be addressed in this study. Responses to these questions will lead to get more awareness on soil surface salt accumulation phenomenon dominated by shallow water table fluctuations. Ultimately, this will help to overcome the substantial knowledge gap concerning this issue and to suggest appropriate management practices which may save costly reclamation efforts and further land losses.

The general objective of this research is to elucidate soil surface salt accumulation phenomenon induced by saline shallow groundwater fluctuations under arid and semi-arid climate. Specifically, it aims at (i) assessing and analyzing short term groundwater fluctuations in salt affected areas in Tunisia (case study Metouia Oasis), (ii) investigating and modeling shallow groundwater responses to recharge events under controlled conditions at laboratory scale and (iii) elucidating the salt accumulation phenomenon under shallow groundwater conditions. To achieve these objectives a methodology based on field investigations in salt affected areas in Tunisia, laboratory experiments and modeling investigations concerning shallow groundwater fluctuations and salt accumulation, was followed.

Field investigations were conducted in two salt affected areas in Tunisia, namely, Dhraa Tamar (Middle West) and Metouia Oasis (South East). These investigations showed that in such arid and semi-arid environments with saline shallow groundwater tables, the capillary rise of water and salts from the groundwater to the soil surface seems to be a major contributor to salt accumulation within soil profile. Metouia Oasis was subject to a high measurement frequency based assessment (30 minutes interval) of shallow groundwater fluctuation from March to November 2009. The resulted hydrograph showed a dynamic groundwater level fluctuation which was the consequence of climatic conditions as well as several hydrological processes taking place in the observation site and in the upstream areas. Interestingly, water table has been shown to rise during rainless periods owing to lateral subsurface groundwater inflow from irrigated upstream areas in the Oasis. At seasonal basis, groundwater was characterized by a declining phase during spring and summer seasons followed by a rising phase starting from mid-October corresponding to the starting of fall and winter seasons. At daily basis, the hourly changes in water table levels showed an evident diurnal variation throughout the day characterized by a downward trend during daytime induced by the high evaporative demand and an upward trend during night time when less evapotranspiration is occurring. Further, the hydrograph was characterized by several rapid water table rises following recharge events followed by extended periods of water table declines. Such behavior was attributed to the reverse Wieringermeer effect (RWE) which reflects high and rapid water table rise due to small water input.

To get better understanding of the groundwater behavior at the site, we needed to know more about shallow groundwater response to recharge events, specifically in relation to RWE. Hence, experimental and modeling investigations of shallow water table fluctuations inside sandy (Toyoura sand) and clayey (Chiba light clay) soil columns in response to surface and sub-surface recharge events have been carried out. Experimental results showed that small application of water could raise the shallow water table level more than 100 times in depth in the case of Toyoura sand and more than 50 times in the case of Chiba LiC, reflecting a reverse Wieringermeer effect response type of groundwater. This response was dependent on the way of water supply. Hence, in the case of surface recharge the water table response was rapid and large while it was gradual and less pronounced when sub-surface recharge was applied. Nevertheless, the RWE was observed for both surface and sub-surface recharge events. This phenomenon occurs when the water table depth is less or equal to the capillary fringe height. Further, water table rise was associated with a prompt change of pressure head values which exhibited instantaneous fluctuations of centimeters due to the addition of millimeters of water. The recharge volumes leading to such disproportionate water table rise were successfully estimated using a simple analytical model based on the moisture retention curve of the soil and considering the hysteresis effect on soil water dynamics within the capillary fringe zone. Simulations based on this model suggested that the soil water movement inside the soil column was mainly upward from the lower part being wetting up after water supply and that the shallow groundwater had a great effect on the vertical distribution of soil water content and soil water potential. Simulations based on non hysteretic models failed to predict the recharge volumes by either highly overestimating or underestimating the measured data.

To fully understand the effects of saline shallow groundwater on soil salt accumulation as observed at field scale we carried out two laboratory experiments simulating soil salinization under fluctuating and stable saline shallow water tables. Water and salt dynamics as well as evaporation rate were monitored and analyzed for both conditions. The analysis of salt profiles observed in soil columns under controlled conditions showed an enhancement of the build-up of soluble salts in soil columns where groundwater was fluctuating compared to those with constant water table. This suggests that soil salinization can be rapid in areas with fluctuating saline shallow groundwater and can not be accurately predicted if such behavior of groundwater was not considered in the modeling process.

Based on these experimental results, the numerical model HYDRUS-1D was first validated and then used to carry out various numerical scenarios. The results showed that groundwater fluctuation caused more salt to accumulate at soil surface compared to that caused by a stable groundwater. By reducing the evaporation rate through a decrease of the temperature intercepted at soil surface, a significant reduction of salt concentration at soil surface was observed. Moreover, frequent irrigations with small quantities were effective to reduce soil surface salt accumulation induced by saline shallow groundwater.

Furthermore, numerical simulations using the model HYDRUS-1D were carried out in order to investigate the effect of saline shallow groundwater fluctuations on soil surface salt accumulation in Metouia Oasis based on shallow groundwater behavior and quality as well as on climatic conditions prevailing in the Oasis. Three cases simulating field observations were considered: stable water tables at depths 60cm and 115.5cm and fluctuating water table between these two depths. The results showed that the fluctuating groundwater has the ability to mobilize and transport soluble salts rapidly when it moves through the soil profile compared to a stable water table. Moreover, the increase of salt concentration toward soil surface was progressive in the case of scenarios with constant water table and rapid and variable in the case of fluctuating water table. These results can be explained by differences in transport mechanisms between the cases with stable water table and the one with fluctuating water table. In the former the salt is transported toward soil surface mainly by capillary rise of saline groundwater and secondary by molecular diffusion. In the case of fluctuating water table a piston like flow of groundwater is the main process inducing rapid and high salt movement in the soil profile. This suggests that in order to solve the soil salinization problem associated with saline shallow groundwater, the water table must be stabilized at harmless levels and the groundwater fluctuation has to be dampened and retarded by control of groundwater recharge.

塩類集積問題は、世界の100カ国以上で発生している地球規模の環境問題であり、試算によれば毎分3haの土地が塩類集積のために放棄されている。特に、中央アジアやアフリカの乾燥地帯、半乾燥地帯での塩類集積が著しい。地球の温暖化傾向は、これに拍車をかけている。塩類集積は様々な要因から発生するが、本研究では浅層地下水が原因となる場合に着目した。なぜなら、オーストラリアやUSAカリフォルニア州において浅層地下水が塩類集積を加速していると報告されており、チュニジアではかんがい農地の30%が塩分濃度の高い浅層地下水の影響を受けていると報告されているからである。

第1章は序論であり、チュニジアの現地フィールドにおける浅層地下水の水位振動と塩類集積問題の実態、浅層地下水の水位変動に関する従来の土壌物理的知見、浅層地下水と塩類集積問題の関連などを整理し、本研究の目的を、「乾燥地・半乾燥地における浅層地下水振動が塩類集積に及ぼす影響」と設定した。

第2章では、これまでに行われてきたフィールド研究、実験的研究、理論的研究などのレビューを行い、わずかな地下水供給が大きな地下水変動をもたらすことを説明した「ウィーランゲミア効果」説と「逆ウィーランゲミア効果」説の重要性を論証した。

第3章では、チュニジア中央部に位置する最大級のメトゥイアオアシスにおけるフィールド調査の方法を述べた。まず、この地域での地下水位がおよそ1.2 m程度とかなり浅いこと、かんがい水のEC値(電気伝導度)が4~5 mS/cmであるのに対し地下水のEC値が約30 mS/cmであることと、それによって土壌水のEC値が約3.5以上という高い値であること、などを実測に基づいて述べた。そして、30分ごとの地下水位を長期に自動計測する方法について説明した。

第4章では、第3章で述べた方法によりメトゥイアオアシスで取得した地下水位長期変動データ、年間降雨量データ、年間ポテンシャル蒸発量データ、年間気温変化データ、土壌物理性データなどを総合的に解析し、これまでに知られていなかった地下水動態を解明した。すなわち、年間降水量が180mm程度しかないのに、この地区の地下水位は浅いときには60cm、深いときには150cmにまで低下すること、しかも全く降雨の無い4カ月間に8回もの大きな地下水位変動が現れ、その振幅が50cmを超えるほどであることを実証した。これらの結果から、メトゥイアオアシスの浅層地下水位変動は、測定位置における降雨やかんがいによるものと、オアシス全体のどこかで行われるかんがいの地下水補給によるものとの2種類があることを明らかにした。

第5章と第6章では、一般に浅層地下水位の激しい上下振動はなぜ起こるのか、また、その振動は土壌表面における塩類集積にどのような影響を与えるかについて、室内カラム実験を行ったことを述べた。特に、地表面から水を浸透させて地下水を補給した場合、1mmの給水で起こる地下水位上昇が119mm(豊浦砂の場合)及び54mm(千葉ライトクレイの場合)であること、他方、地下から水を補給して地下水を押し上げる場合、1mmの給水で起こる地下水位上昇が182mm(豊浦砂の場合)及び11mm(千葉ライトクレイの場合)であることを述べ、これらの結果は「逆ウィーランゲミア効果」によって良く説明できることを証明した。そして、5日に1回の割合で地下水位を上下に30cm振動させた場合、その15日後には地下水位振動による土壌中の塩分濃度上昇が観測されたことから、浅層地下水位の上下振動は、短期的には土壌中の塩分濃度を上昇させているとの結論に至った。

第7章では、第6章で行ったカラム実験の結果を再現し、長期的な変化を予測するための数値シミュレーションを行った。用いたプログラムはHYDRUS-1Dであり、60日間蒸発の間に深さ60cm地下水から地表面に多量の塩分が上方輸送され、地表の塩類集積が起こることを予測した。

第8章は結論であり、チュニジアのメトゥイアオアシスで従来全く知られていなかった浅層地下水位の振動の実態を測定したこと、降雨量の少ない土地での大きな地下水位変動が「逆ウィーランゲミア効果」によって説明できること、塩分濃度の高い浅層地下水位の振動は、土壌における塩類集積を助長するものであり、振動を小さく抑える技術が必要であることを述べた。

以上要するに、本論文は、乾燥地・半乾燥地における塩類集積問題と浅層地下水位振動との関連性に着目し、チュニジアのフィールドにおける実態調査と実験室内におけるモデル実験、そして数値シミュレーションを用いて、現場で発生する問題の解析を行い、塩類集積を軽減する技術の提言を行ったものであり、学術上、応用上貢献するところが少なくない。よって審査委員一同は本論文が博士(農学)の学位論文として価値あるものと認めた。