This can be partially explained by the following. At night CAM plants convert starch to PEP and in doing so generate NADH. This NADH is used to reduce The OAA (that is formed when PEP is carboxylated by HCO3) to malate which is stored in the vacuole. In the light, the malate is de carboxylated to release carbon dioxide, pyruvate and, importantly in relation to your question, NADPH. This NADPH can be used in the Calvin-Benson cycle to reduce 1,3-BPGA to 3-PGaldehyde. Since the NADPH has been supplied without the action of PSII, and ATP required in the Calvin-Benson cycle by photophosphorylation less water is split and less oxygen released.

Here you can find a quite good explanation: http://5e.plantphys.net/article.php?id=400

The mechanism of photo respiration particularly for CAM plants is central to avoid oxygen toxicity.

This is a very good question.  I appreciate the answer projected by Surendra Chandrasekhar Sabat.  Photosynthesis in CAM is not spatially separated.   The RuBP Case is endowed with both carboxylation and oxygenation sites.  The CO2 conc is very high  in CAM plants due to maleate decarboxylation and O2 from exterior can not enter the plant, as stomata are closed during day time.  Any  small amount of Photosynthetically generated O2 can be taken care of by RuBP carboxylate/oxygenase   through photo respiration 

I would try an actual CAM leaf disk in an oxygen electrode, but the interplay with vacuoles and their large buffer volume of malate is a possibility.

I think the sum of above answers is providing good explanation for this very interesting answer. As a rule CAM plants show high rates of cyclic electron transport (PSI activity is significantly higher as compared to PSII activity). - Therefore chlorophyll fluorescence measurements are not easy to interprete in CAM plants. - Moreover, CAM plants depend on photorespiration capacity.

As we know, water splitting is the source for H that is transferred NADPH to Calvin cycle to make carbohydrate. So one water molecules should be split for production of each NADPH+H+ that is used to produce one PGA. Consequently, one molecule of O2 would evolve per each fixed CO2. If this O2 is destined to be used in photorespiration, then RUBISCO should carry out one oxidative reaction per each carboxylation reaction that brings the net photosynthesis rate to zero. Therefore, the role of photorespiration in photosynthetic O2 detoxification seems hard to imagine. Thanks Gustavo for the useful link but still it appears that the provided scenarios there, are not enough. The evolved O2 is destined to be released to air in some way. It could not concentrate inside the plant, and I believe that this is overlooked in that link. This should happen during the night, while the stomata are open. We can imagine some kind of reversible antioxidation reactions or something like that. Even so, this is only in theory, and we need some evidence. I wonder if this question is not covered by any research, or it might be that this part of photosynthesis of CAM plants has been remained under the shadow of their odd CO2 fixation habit discovery?

Here is a very simple answer: Oxygen can diffuse through biological membranes. Thus the produced oxygen can diffuse into the environment even if the stomata are closed. (Just consider the respiration of not CAM plants: in the night most of them have closed stomata but they can take up oxygen from the environment for their respiration processes.)

Dear Bela,

It is correct that O2 like CO2 can diffuse through biological membranes, but we had better considering that with closed stomata, they should diffuse through the cuticle as well. If  this amount of Oxygene could diffuse out then we could imagine that the same molar amount of CO2 to diffuse inside via the same path? In this case, there would remain no reason for stomata to open at night. However, I believe that as CAM plants are native to harsh environments, they are well sealed to prevent water loss and gas exchange in the same time.  The fact that non CAM plants could absorb oxygen during the nights with closed stomata could be due to that there remains microscopic pores open where slight diffusion might be possible. However, they are not sealed in the same degree as CAM plants are.  Another point to consider is that the O2 needed for respiration at night is far less than what is produced during photosynthesis.

In fact if plants could evolve some other material instead of cuticle that let gas exchange while blocking water loss,  we could reduce the water problem in a large degree. Maybe some time this could happen by advances in science of materials and genetic engineering.

Ulrich Lüttge directly discusses the question of O2 accumulation during phase III of CAM behind closed stomata in the light period - Lüttge U (2002) CO2-concentrating: consequences in crassulacean acid metabolism. J Exp Bot vol. 53, pp. 2131 - 2142. He quotes a range of measured O2:CO2 ratios for CAM species during in phase III in the light and concludes that the ratio is 0.1 to 0.5 of the ratio in ambient air and thus that the CO2 concentration relative to the leaf internal O2 concentration would be sufficient to favour the carboxylase activity of Rubisco over the oxygenase activity. Thus, photorespiration would be reduced to negligible levels during phase III in the most efficient CAM species that have the lowest O2:CO2 ratios. For full details see page 2137 and onwards of the article by Lüttge (2002).

In general, not specifically for CAM plants,   i  would also add that O2 generated by the light reaction, have very  small contribution to the overall amount of O2 (21% of the athmosphere). this is in  sharp  contrast to CO2  which is found in small amount in the athmosphere (0.04%), and therefore CO2  generated, accumulated  or consumed by the plants have much stronger effect on CO2  concentration.

Thank you James for sharing the interesting article by Lüttge. As I understand the presence of a oxidative stress due to high O2 concentration is confirmed in CAM plants and a role for anti-oxidative response system is proposed. However it seems that  the fate of produced O2  remains unclear. The following text is from that article:

"Phase III oxidative stress

The O2‐concentrating consequence of CO2‐concentrating in planta has been demonstrated above. Studies of O2‐exchange by mass spectroscopy of stable oxygen isotopes (16O, 18O) and the main electron flow of chloroplasts have revealed the strong O2 production in CAM plants in the light (Thomas et al., 1987). Experiments measuring O2 evolution in the leaf disc O2‐electrode, where leaf discs of CAM plants have been artificially exposed to 5% ambient CO2, were performed (Adams et al., 1987; Adams and Osmond, 1988; Borland and Griffiths, 1989; Maxwell et al., 1998), and Osmond et al. (1996) note the action of the ‘O2‐pump’ leading ‘to ever‐increasing O2‐concentration in the experimental chamber as malate decarboxylation proceeds’. Thus Phase III CO2‐concentrating not only stays short of preventing photorespiration and photoinhibition (see above), but the corresponding O2‐concentrating even may be one of the major adverse consequences.

High piO2 supports formation of aggressive reactive oxygen species (ROS) such as hydrogen peroxide (H2O2), superoxide (O2•–) and the hydroxyl radical (OH•). CAM plants are especially equipped to deal with this oxidative stress by the increased expression of an antioxidative response system (ARS). In the C3/CAM‐intermediate Sedum album this appeared to be especially important during the C3 to CAM transition when the activities of ascorbate peroxidase, superoxide dismutase (SOD), gluthatione reductase, and monodehydroascorbate reductase as ROS scavenging enzymes were increased (Castillo, 1996). In the C3/CAM‐intermediate M. crystallinum, where CAM‐induction is elicited by salt stress, it is often difficult to distinguish ARS expression in response to salinity from requirements of CAM, for example, the up‐regulation of cytosolic CuZn‐dependent SOD (Hurst and Ratajczak, 2002). It appears, however, that up‐regulation of mitochrondrial Mn‐dependent SOD is a typical reaction to the induction of CAM (Miszalski et al., 1998; Broetto et al., 2002

Very interesting, important indeed, discussions. I want to answer the question straightly.

CAM plants handle the O2 evolved in photosynthesis during daytime by the following ways-

1) Mitochondrial respiration

2) Photorespiration, and

3) Formation of some reactive oxygen species (ROS) as adverse consequence

You can go through the book- Plant Physiology by Lincon Taiz and Eduardo Zeiger (Chapter 8, Topic 8.11: Photorespiration in CAM plants).

Several contributions about toxic oxygen and photorespiration but the main question is that leaf disk O2 electrode chembers can show avariety of responses, and every CAM plant needs experiments more than theory. Might I add that the very small stomatal conduction for H2O allows some O2 to excape from leves during a CAM day. Eventually shall one take i minf that  near all O2 is from photosynthesis but CO2 is low for the large fossil sinks produced by carbonates and fossil lignin of land trees. We thrive for an high O2..

CAM ("crassulacean acid metabolism") plants also initially attach CO 2 to PEP and form OAA. However, instead of fixing carbon during the day and pumping the OAA to other cells, CAM plants fix carbon at night and store the OAA in large vacuoles within the cell. This allows them to have their stomates open in the cool of the evening, avoiding water loss, and to use the CO 2 for the Calvin-Benson cycle during the day, when it can be driven by the sun's energy. CAM plants are more common than C4 plants and include cacti and a wide variety of other succulent plants.

• Cactus, pineapples have different adaptations to Hot, Dry Climates.

• Plants that use the CAM Pathway Open their Stomata at NIGHT and Close during the DAY, the opposite of what other plants do.

• At NIGHT, CAM Plants take in CO2 and fix into Organic Compounds.

• During the DAY, CO2 is released from these Compounds and enters the Calvin Cycle.

• TEMPORAL SEPARATION

• Because CAM Plants have their Stomata open at night, they grow very Slowly, But they lose LESS Water than C3 or C4 Plants.

The stomatal closure of CAM plants allows some residual conduction for small molecules like O2 (stomata are not valves), and intercellular spaces with spongy structrure are flexible. I would point out thet Hadavi question is about oxygen and even if oxydative damage requires O2 the levels insude a CAM leaf are not toxic. To generate radicals is requirred a complex action of cascades linked to cell integrity and senescence. CO2 cycling for wasteful and requires reductants that are diverted from growth. The chlorenchima proportion to whole tissue is reduced, but again Agave and pineapple are not slow growing. . 

Sorry about spelling, my phone is not so clever and I'm growing short of sight: The stomatal closure of CAM plants allows some residual conduction for small molecules like O2 (stomata are not valves), and intercellular spaces with spongy structrure are flexible. I would point out thet Hadavi question is about oxygen and even if oxydative damage requires O2 the levels inside a CAM leaf are not toxic. To generate radicals is required a complex action of cascades linked to cell integrity and senescence. CO2 cycling is wasteful and requires reductants that are removed from growth. The chlorenchima proportion to whole tissue is reduced, but again Agave and pineapple are not slow growing. .Of course several CAM are very resistant to photoxidation and phoinhibition much more than C3 or C4.