RECOMMENDATIONS 



Some understanding of the water balance could help explain variations in 

 vegetation parameters. The most important measurements related to water balance 

 are: available soil moisture, precipitation, average temperature and day 

 lengths, and a measure or calculation of evaporation or evapotranspiration. 



The amount of soil moisture available to plants is the moisture content of 

 the soil minus the permanent wilting percentage. Soil moisture is determined by 

 weighing fresh soil samples (by depth intervals), drying the samples at 105 

 degrees C, and weighing the dry samples. Soil moisture should be measured 

 monthly from April to September. The permanent wilting percentage can be 

 approximated (for non-saline soils) by measuring the soil moisture remaining 

 after samples have been subjected to 15 atmospheres of pressure and allowed to 

 drain. 



Good candidates for available moisture determination include the following 

 communities: Stco-Agsm/Bogr (which has a good mix of dominant species), 

 Bogr/Agsm (representative of the flatter lowlands), Agsp/Bogr-Caf i 

 (representative of uplands with a cool-season, mid-grass dominant, and warm- 

 land, cool-season, short-grass subdominants) , Ansc (warm-season mid-grass), and 

 Arca/Agsm/Bogr (sagebrush and water run-in areas). After initial preparation, 

 field work will take about one day per month and can be performed by a local 

 resident after a training session. Samples can be processed at the University 

 of Montana. 



Precipitation should be measured at the Dreyer Ranch headquarters year- 

 round, and near the exclosures on Nelson Creek from April to September. 

 (Approximate cost for equipment, less shipping: standard rain and snow gauge, 

 $160; stand, $40; wind screen, $240.) Temperature should be recorded on a 

 thermograph at Dreyer Ranch headquarters from April to September. (Approximate 

 cost for equipment, less shipping: Dickson Minicorder (4-inch, 7-day, key- 

 wound), $140 plus charts, or Weathertronics circular thermograph (as above), 

 $150 plus charts; small louvered shelter (without support), $80. 



More precise productivity measurements are recommended to improve the power 

 of conclusions and to allow quantification of the relationship between climatic 

 conditions and vegetation. In 1977, about twenty-four samples were taken in each 

 exclosure. The number of plots sampled in 1978 was determined from 1977 data by 

 calculating the number of plots necessary to obtain a sample mean within 20% of 

 the population mean at the 80% probability level. Through similar calculations, 

 the number of plots clipped in 1979 was determined from 1977 and 1978 data. 



To improve sampling precision, more stringent criteria were imposed In 1980, by 

 determining the number of samples necessary to obtain a sample mean within 20% 

 of the population mean at the 90% probability level, using 1979 data with 

 reference to the number of samples taken in 1977 and 1978. 



In the future, even more stringent criteria should be employed in 

 determining sample size. Communities sampled for available soil moisture should 

 be sampled for productivity to attain a sample mean within 10% of the population 

 mean at the 90% probability level. Other exclosures should be sampled to attain 

 an 80% probability of being within 10% of the population mean. Productivity 

 data from 1977 through 1980, with the possible omission of data severely 

 affected by grasshoppers, should be used to calculate the number of samples to 



-32- 



