The harvest of Macrocystis off the Baja California Peninsula shows a seasonal pattern with minimum values in winter, and the maximum during spring and summer, then decreasing in autumn (Fig. 10). The spring and summer harvests were greater (P < 0.05) than winter and autumn, and the harvest of winter was the lowest (P < 0.05). In general the harvest of all beds had the same pattern. In the beds at Punta Mezquite (03), Salsipuedes (04), and Bahia de la Soledad (08), which were more frequently exploited, this pattern is evident, and less so the beds less harvested, such as Playas de Tijuana (02), Isla Todos Santos (05), and San Isidro (12). A similar behavior was found when the harvest obtained per hour of ship harvest (CPUE) for 1993 — 1999 was analyzed. The highest harvest/hour was during May to August (75 t/hour). These values were significantly different (P < 0.05) to both periods: February to April (62 t/hour) and September to December (54 t/hour), which were lower (Fig. 11).

3.2 Relation harvest-effort

During 1956 to 1999, the harvest of Macrocystis increased as a function of the level of effort (number of trips) (r = 0.98, Fig. 12) and similarly when the effort was measured as number of hours of ship harvest (r = 0.85) for 1993 — to 1999 (Fig. 13).

4. Discussion

From 1958 to 2004, the average harvest of Macrocystis was 26,000 t, which was about 50% of the standing crop estimated by Casas-Valdez et al. (1985) and Hernandez et al. (1989a, 1989b, 1991), who evaluated the biomass and standing crop of Macrocystis using aerial photography and field work along the area of the distribution of this kelp. From Islas Coronado to Bahia del Rosario they estimated a standing crop of 40,000 t in summer 1985 and 63,000 t in summer 1986. This species of seaweed has a high growth rate (13 — 21 cm/day) (Hernandez, 1996) and its regeneration rate is high.

The lowest harvest and effort recorded in category I can be related to: a) the harvest being suspended in beds 11 (1978), 06 (1985), 07 (1984), 02 (1991), and 01 (1993), b) the long distance from the beds to the base port, bed 12 (12 h 20 min), 13 (13 h), 14 (16,5 h), and 15 (20 h). The highest harvest and effort recorded in category III can be related to a) a high productivity of the bed and, b) the short distance from the bed to the base port (5 h). In relation to the previous information, Roberto Marcos (com. pers.) noted that the quantity of effort used at each bed depended on the productivity of the bed and its cost of operation, which are related principally to the distance that the ship most run from the base port to the bed. Guzman et al. (1971) and Corona (1985) mention that the more productive beds for 1956 — 1968 and 1974 — 1985 were the beds 03, 04, 08, 09, and 10 that are in categories II and III of this study. The largest harvest of Macrocystis was in spring and summer and the lowest in winter.

Along the northwest coast of the Baja California Peninsula the greatest upwellings are during spring and summer (Casas-Valdez, 2001) and have high nutrient concentrations and lower temperatures (Lynn & Sympson, 1987; Pares & O’Brien, 1989) that favor the development of Macrocystis fronds (Tegner & Dayton, 1987; Tegner et al., 1996; Lada et al., 1999). Growth studies in situ showed that the lower temperatures of spring enhance the growth rate of Macrocystis (Gonzalez et al., 1991) and also the increase of nutrients (Zimmerman & Kremer,

1986) . Casas-Valdez et al. (1985) and Hernandez-Carmona et al. (1989a, 1989b, 1991) evaluated the biomass and standing crop of Macrocystis along their natural distribution and found the largest surface and biomass of the beds in spring (45,000 t) and summer (63,000 t). They noted that these values were three times greater than those in winter (14,000).

Playa Tijuana 1958 1964 1970 1976 1982 1988 1994 2000

1958 1964 1970 1976 1982 1988 1994 2000

Fig. 3. Data series of harvest and effort of the Macrocystis pyrifera beds: Islas Coronados, Playas de Tijuana, Punta Mezquite, Salsipuedes, Isla Todos Santos and San Miguel y El Sauzal. Harvest, effort.

1958 1964 1970 1976 1982 1988 1994 2000 Year

Fig. 4. Data series of harvest and effort of the Macrocystis pyrifera beds: Punta Banda, Bahia de la Soledad, Santo Tomas, Punta China, Punta San Jose and Punta San Isidro.

Harvest, effort.

Isla San Martin

Punta San Telmo

H 1 1 1 1 1 1 1 1 1—I—г 1958 1964 1970 1976 1982 1988

1994 2000

Isla Coronado 1958 1964 1970 1976 1982 19! Year 1994 2000
Isla Todos Santos Year	San Miguel Sauzal Year

Fig. 7. Data series of harvest per unit effort (CPUE) of the Macrocystis pyrifera beds: Islas Coronados, Playas de Tijuana, Punta Mezquite, Salsipuedes, Isla Todos Santos and San Miguel y El Sauzal.

Fig. 8. Data series of harvest per unit effort (CPUE) of the Macrocystis pyrifera beds: Punta Banda, Bahia de la Soledad, Santo Tomas, Punta China, Punta San Jose and Punta San Isidro.

Isla San Martin	Punta San Telmo
Fig. 9. Data series of harvest and effort of the Macrocystis pyrifera beds: Punta San Telmo, Isla San Martin and Bahia del Rosario.

Fig. 10. Seasonal variation of the harvest of Macrocystis pyrifera in Baja California Peninsula. ± 2 SD.

The CPUE was used as indicator of abundance for Gelidium robustum a red seaweed that is harvested along in the west coast of the Baja California Peninsula from 1956 to the present. The unit of effort selected for this fishery was the fishing equipment (a boat with three fishermen) and the CPUE was expressed as harvest/boat (Casas-Valdez et al., 2001). They used the CPUE to determine the relationship of the abundance of Gelidium with both temperature and upwelling. As an indicator of the abundance of Macrocystis, Tegner et al.

(1996) compared data on the maximum canopy of the kelp forest and size of the annual harvest of Macrocystis for California, and they chose harvest size as the most useful data to relate to environmental variables. They pointed out that harvest size was a reflection of
changes in consumer demand, harvest productivity, and natural disturbances. They also noted that this variable has the advantage of integrating growth over a long period and has less subjectivity in its measurement.

In our study, we considered that the CPUE shows the changes in the abundance of Macrocystis better than only the harvest, because the size of the harvest varies according to the amount of effort used and not only as a function of the abundance. Furthermore, the use of the CPUE is cheaper than the use of aerial photography and field work to determine the variations in the abundance of this resource. Casas-Valdez et al. (2003) mentioned that the harvest/trip is a reasonable indicator of the Macrocystis abundance, because about 60% of the alga biomass is present in the surface canopy (North, 1968), and almost 95% of its production takes place in the first meter of the top of the water column, and the kelp is harvested at a maximum depth of 1.2 m. Furthermore the ship operations were the same at all beds and did not change over the study period. We considered that the harvest/hour is a better indicator.

The surplus production models of Schaefer and Fox were used to assess the fishery condition of Gelidium off the Baja California Peninsula from 1985 to 1997. The results have shown that the resource is not overexploited (Casas-Valdez et al., 2005). In this study we tried to use these surplus models for the data of Macrocystis, but the fit was not satisfactory. This occurred because an increased effort produced increased harvest. To fit these models, it is necessary to count, along with the catch, effort, and CPUE data, an ample range of fishing effort levels, preferably including those that correspond to the level of overexplotation in the curve (IATTC, 1999). The linear relation (correlation) found between the harvest and the effort used for the Macrocystis fishery means that the fishery was in the eumetric growth segment of the curve of the Schaefer model and therefore it is possible to conclude that there have not been negative effects of the harvest on the resource. It is considered that the effort has not been increased, due to the fact that the demand for Macrocystis has not been increased either. In fact, the harvest drastically decreased in 2005, when the principal company that was buying this kelp as raw material for the alginate production ceased buying it (Roberto Marcos com. pers.).

5. Conclusions

The Macrocystis fishery along the Mexican Pacific coast did not show signals of over exploitation due to increases in the effort corresponding to increases in the harvest, and the CPUE has been maintained almost constant since the begging of the harvesting of this resource until now (2004), with the exception of the years when "El Nino" event was present.

Along the northwest coast of the Baja California Peninsula, the highest harvest of Macrocystis was found in spring and summer, when the greatest upwellings ocurre in agreement with high nutrient concentrations and lower temperatures.

The harvest per unit of effort (CPUE) was more stable in the beds where more effort was used, as in the beds at Punta Mezquite, Salsipuedes, Bahia de La Soledad, Santo Tomas and Punta China, whereas in the beds where less effort was used the CPUE was more variable.

6. Acknowledgment

Thanks to Productos del Pacifico, S. A. de C. V. for providing the data of harvest of Macrocystis. We really appreciate the adviser of Roberto Marcos Ramirez. Thanks to Dr. Ellis Glazier for editing this English-language text. Margarita Casas Valdez and Daniel Lluch Belda are fellows of COFAA-IPN and EDI-IPN.