The intensification of agricultural production in the Mekong Delta has faced serious challenges with respect to increased use of agrochemicals and especially pesticides. The indiscriminate use of pesticide could potentially impact on the long-term food production, environmental and human health in the delta. The aim of this thesis was to investigate the negative side effects of the current use of pesticides on climbing perch (Anabas testudineus) in rice fields using brain acetylcholinesterase (hereafter referred to as AChE) activity as a biomarker. The empirical work, on which this thesis is based, includes structured questionnaires, laboratory and field experiments. First, a field survey using questionnaires was carried out to gain a better understanding of the current state of rice farming systems, the use of pesticides and attitude to pest management strategies among rice and rice-fish farmers, as well as to provide basic information for the set-up of the laboratory and field experiments. Secondly, laboratory studies were conducted to clarify if the selected insecticides applied alone and in mixtures caused negative side effects on climbing perch fingerlings. Thirdly, further toxicity studies were carried out, under rice field conditions, to further investigate the toxicity effects of the insecticides, applied alone, in mixtures and under sequential applications, on climbing perch fingerlings. The results showed that although there were a more selective use of pesticides and an increased awareness among farmers of the negative side effects of pesticides in 2007 as compared to 1999, the current use of pesticide in the Mekong Delta still cause many problems to the environment and human health. Chlorpyrifos ethyl (hereafter referred to as CPF) was found to cause a significant and more prolonged inhibition on the brain AChE activity in climbing perch than fenobucarb (hereafter referred to as F). The inhibition by the mixture of CPF and F were significantly higher than the inhibition by only F, but less prolonged and significant lower than the inhibition by only CPF. The results suggest that the combined effect from a mixture of F and CPF can create both additive effects initially and later antagonistic effects. CPF and F applied at concentrations used by farmers, either as separate doses, in a mixture or in sequential doses, decreased the brain AChE activity, growth and survival rates in climbing perch. The results demonstrate that brain AChE activity in climbing perch is a relevant biomarker for monitoring of exposure to, and sub-lethal impacts from organophosphates and carbamates under tropical conditions. The result also shows that 2-PAM re-activate the brain AChE activity, and can be used as an alternative method to assess the AChE inhibition level in organisms recently exposed to OP’s, in situation where it may be difficult to find unexposed individuals as controls. In conclusion, this thesis shows that the current use of pesticides in the Mekong Delta has a negative effect on climbing perch living in rice fields. It indicates that a sustained long-term food production in the Mekong Delta must be based on ecological principles, taking advantages of ecosystem biodiversity and productivity, and not through intensified use of pesticides. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Submitted. Paper 5: Submitted.</p>

Field and greenhouse experiments were conducted to determine barnyardgrass control as affected by glufosinate and graminicide tank-mixtures, application timing of tank-mixtures of graminicides plus glufosinate and application time of day of tank-mixtures of glufosinate and clethodim. When increased rates of graminicide were tank-mixed with glufosinate, barnyardgrass control was unaffected by quizalofop-P plus glufosinate; however, clethodim plus glufosinate control in the field indicated the potential for reduced barnyardgrass control. When evaluating increasing glufosinate rates tank-mixed with graminicides, barnyardgrass control was not negatively affected by the combination of glufosinate and graminicides. The difference in soybean yield among the graminicides may indicate that the cyclohexanedione herbicides had a slight yield advantage over the aryloxyphenoxypropionate herbicides due to potential increased levels of barnyardgrass control. Applications of glufosinate alone provide variable control throughout the growing season in both field and greenhouse experiments. Although barnyardgrass control in the field was not affected by glufosinate application timing, data from the greenhouse indicates potential exists for reduced control if glufosinate is applied 1 or 3 d before graminicides. Clethodim was unaffected by application time of day; however, glufosinate applications at midnight reduced barnyardgrass control compared to applications made at noon and 6 P.M. Applications at 6 A.M. also reduced barnyardgrass efficacy compared to applications at 6 P.M. Environmental factors such as temperature and light at the time of application are likely responsible for the time of day effects observed in these studies. For maximum benefit from incorporating graminicides into a glufosinate weed control system, fluazifop-P, quizalofop-P, clethodim and sethoxydim should be applied with glufosinate at 594 or 890 g ai ha-1. Sequential treatments of glufosinate should be applied 7 d prior to a graminicide application or 1, 3 or 7 d after a graminicide application. To optimize barnyardgrass efficacy with tank mixtures of glufosinate and clethodim, applications should be made at noon or early evening to avoid potential time of day effects.

Field studies were conducted to test three application methods for applying antagonistic herbicide combinations: 1) tank mix (TMX), 2) mix-in-line (MIL), and 3) separate boom (SPB). Sethoxydim applied with bentazon, glyphosate applied with dicamba or 2,4-D, and clethodim applied with dicamba or 2,4-D had higher efficacy when applied using the SPB method. Antagonism of all the herbicide combinations above was observed when applied using the TMX and MIL methods. In some cases, antagonism was avoided when using the SPB method. Three application methods tested in greenhouse studies were 1) TMX, 2) synthetic auxin applied first (AAF), and 3) synthetic auxin applied second (AAS). The AAS application method resulted in higher weed control than the TMX and AAF methods. Analysis done through liquid chromatography mass spectrometry supported the greenhouse results with higher rates of glyphosate detected with the AAS method.

Quizalofop and florpyrauxifen-benzyl are both new herbicides for rice in the midsouthern U.S. Quizalofop is only effective for control of monocot weed species; therefore, mixtures of florpyrauxifen-benzyl with quizalofop could be beneficial in acetyl CoA carboxylase (ACCase)-resistant rice. Field experiments were conducted at the Delta Research and Extension Center in Stoneville, MS, in 2017 and 2018 to evaluate control of monocot and dicot weed species with sequential applications of quizalofop including auxinic herbicides in the first or second treatment. Other field experiments in 2017 and 2018 evaluated sequential applications of different rates of quizalofop with florpyrauxifen-benzyl included in treatments immediately prior to flooding. A final field experiment in 2017 and 2018 evaluated growth and yield of six ACCase-resistant rice cultivars and advanced lines following POST applications of florpyrauxifen-benzyl.