Development of a Browse Database

Principle Investigators

Merle M. Moore, Associate Director of Grounds Maintenance and Horticulture Denver Zoological Gardens

Nancy A. Irlbeck, PhD., Department of Animal Sciences Colorado State University Alison Kirschner, Masters of Science Student Colorado State University


The use of browse (leaves and twigs from shrubs, trees and herbaceous plants) is becoming widely practiced as a form of captive animal enrichment and, in some cases, an important nutritional supplement to an animal’s regular diet. To date little research has critically evaluated the nutritional value or health implications of feeding browse. The Denver Zoological Gardens, in conjunction with the Department of Animal Sciences at Colorado State University (CSU), is currently engaged in a study of landscape plants commonly fed as browse to animals in zoos. Also involved, as cooperating researchers in this initial study, are the North Carolina Zoo, the Phoenix Zoo, the Columbus (Ohio) Zoo, and the Palm Beach Zoo. We are also exchanging data with Dr. Ellen Dierenfeld who is conducting a similar study at the Wildlife Conservation Center of the Bronx Zoo. Our ultimate objective is to produce a database of information on animal browse supplied from landscape plants. By doing so we hope to provide a greater understanding of the utilization of browse and enable zoo staff to make more efficient and informed feeding decisions. We have developed a standardized protocol to evaluate variability in nutrient content and chemical composition of browse by season and location. We are also investigating species preferences and potential health implications of various browse plants.

Study Technique

Each of the participating zoos is currently collecting browse samples two to three times during the growing season (Spring, Summer and Fall). For each plant in the study twenty (20) branches, 10 inches (25 cm) in length from current season’s growth, are collected and labeled “Sample A”. In addition, two (2) branches, 10 inches (25 cm) in length, also of current season’s growth, are collected at the same time, from the same plant, or group of plants, and labeled “Sample B”. Sample “B” was placed in a Zip Lock bag to prevent moisture loss.

For each “Sample A” plant, the leaves were separated from the stems. Each fraction (leaves and stems) are then dried in a laboratory oven between 50-60 degrees Centigrade for 48 hours. Upon removal from the drying oven, the samples were weighed to account for the leaf:stem ratio (dry matter basis). Each fraction is then ground using a 1 mm screen in a Wiley laboratory mill.

For each “Sample B” plant, leaves and stems are separated, and then dried at 100 degrees Centigrade for 48 hours to determine the harvest dry matter concentration for each plant. Samples of each plant fraction are analyzed for dry matter; ash content; crude protein using the combustion method; detergent fiber analyses including neutral detergent fibers (NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL); and mineral content using Inductively Coupled Plasma Emissions Spectroscopy using National Institute of Standards and Technology as standard reference material (Ca, K, Mg, Na, P, Cu, Fe, Mn, Zn and Mo). From these tests we are able to measure the amount (percentage) of each mineral in the sample.

Wet Chemistry Analyses

All plants are composed of cells having fibrous cell walls for support and protection. Within the plant cells are several soluble compounds, most of which are highly digestible. Fiber digestibility can be a limitation to plant utilization. Forage quality is thus determined by knowing as much as possible about the cell wall material and its fiber content. The detergent fiber analyses (NDF, ADF, ADL) was sequentially determined using the ANKOM method, utilizing a heat resistant enzyme for the NDF. To do this a given sample of ground and dried plant material is subjected to a sequential treatment of three different chemical reagents. Using a base reagent (NDF) the amount of hemicellulose, cellulose and lignin are determined. Then an acid reagent (ADF) is used to remove the hemicellulose, leaving the cellulose and lignin. This is followed with a 72% sulfuric acid treatment (ADL) which eliminates the cellulose, leaving only the lignin.

The analyses of NDF, ADF and ADL to determine forage quality is critical to an understanding of the research data we are compiling in regard to landscape plants used for browse. Acid Detergent Fiber (ADF) is the percentage of highly indigestible plant material in a feed or forage. It is a plant constituent that is insoluble in acid detergent. ADF differs from crude fiber in that ADF contains silica. Silica and lignin in plants are associated with low digestibility. The lower the ADF, the more feed an animal can digest. For instance, in Colorado, we can observe the wild elk herds in Rocky Mountain National Park moving to alpine meadows to graze after the calves are born. They do this because native grasses occurring at the lower montane life zone have begun to accumulate large amounts of silica and lignin by late June. Moving to the alpine meadows insures the elk cows and calves a two-month long supply of nutritious, highly digestible grass to graze on. Another outcome of this phase of chemical testing is a measurement of the total nitrogen concentration in the plant, usually expressed as crude protein. This is a useful measure to determine that an adequate intake of nitrogen is occurring in the animals’ diet.

Neutral Detergent Fiber (NDF) is the percentage of cell wall material or plant structure in a feed. It is insoluble in neutral detergent and is only partially available to animals. The lower the NDF percentage, the more an animal will eat, so a low percentage of NDF is desirable. In other words, low NDF usually indicates high intake by the animal. The final analyses, ADL, indicates the amount/percentage of lignin in a given plant sample. Being non-digestible, lignin cannot be used by ruminants, or any animal, for energy.

At this stage of our research project, completing these time and material consuming series of wet chemistries is necessary in order to provide the data needed to calibrate the Near Infrared Reflectance Spectroscope (NIRS) for future analysis of landscape plants. Many years of grain and forage research has resulted in the development of this analysis technique, utilizing Near Infrared Reflectance Spectroscopy (NIRS) to very quickly analyze a sample of forage plant material. However, this equipment has never been calibrated for testing landscape plant material. It takes, at a very minimum, doing the wet chemistry for 500 or more plant samples to compile sufficient data to do this calibration of the NIRS. The data from several thousand samples would be far better, thereby greatly increasing the accuracy of future NIRS readings.

Database Development

Over the past three years of our study, a graduate student at Colorado State University has been preparing and inputting browse data into a computerized database, created using Microsoft Access-Version 8 software. Information compiled from the nutrient analysis of browse, and from the AZH Animal Browse Survey published in 1998, will be incorporated into the database. A literature search is also on going, and pertinent data, when found, will be incorporated into the database with relevant citations being indicated.

Research Objective

When completed, the browse database will be made available to zoos, and other interested parties, managing exotic animal species in captivity. It will contain a readily accessible compilation of plant common and scientific names, nutrient content based on harvest time and region, possible concerns (for example, potentially harmful secondary plant compounds), animal species that can and cannot be fed a specific browse and the source of literature as it applies. We expect a person looking up browse information to be able to access it, through cross-indexing, by plant name, geographic region, secondary compounds found in browse species, and animal species that browse can be fed or not fed to.

Current Challenges

One of the major challenges we face at this time is processing a sufficient number of samples (wet chemistries) to be able to accurately calibrate the NIRS and establishing the base-line equations. There are significant cost implications in both chemical reagents and personnel to run these tests and compile the data. For instance, each crude protein analyses costs $5.00/sample with the minimum 500 projected samples costing $2,500. Each Detergent Fiber Analyses costs $9.20/sample or $4,600 for 500 samples. These figures are exclusive of the cost for personnel to process each sample which is rapidly approaching $24,000 contributed by CSU. Funding to support this research project has also come from a recent grant of $3,000 from AZH and an initial investment of over $15,000 from the Denver Zoological Gardens Conservation Biology Department. Additional funding is currently being sought by the Development Department staff at the Denver Zoological Gardens for continuing support of project expenses. Another challenge that we will be facing a little further down the road is how best to distribute the information generated by this research. Do we charge a fee to access the information, using the monies raised to support additional research, or do we distribute the information free, as a service? We will certainly be investigating the many options available to do this as our research project progress.

I believe the ultimate outcome of this project will greatly benefit zoos in general and zoo horticulturists in particular. As we strive to utilize horticulture science in support of more humane and enriching animal exhibits in zoos, projects like this browse data base project will make available to us the technologically advanced methods and equipment to accomplish this important objective.

Literature Cited

Kirschner, A.C., Irlbeck, N.A., Moore, M. M. 1998. Implications of Zoological Database. Comparative Nutrition Society Proceedings, Banff, Canada, August 1998

Lehr, T.L., Moore, M. M., Williams, S., Irlbeck, N.A. 1997, The establishment of a zoological browse database. Nutrition Advisory Group Proceedings, Fort Worth, Texas, October 1997